Pruner: parallel pruning

refs #62

.circleci/config.yml

@@ -1,386 +0,0 @@
-version: 2.1
-orbs:
-  # NB: this is not the Go version, but the Orb version
-  # https://circleci.com/developer/orbs/orb/circleci/go#usage-go-modules-cache
-  go: circleci/go@1.11.0
-
-commands:
-  setup-home-local-bin:
-    steps:
-      - run:
-          shell: /bin/bash -euo pipefail
-          command: |
-            mkdir -p "$HOME/.local/bin"
-            line='export PATH="$HOME/.local/bin:$PATH"'
-            if grep "$line" $BASH_ENV >/dev/null; then
-              echo "$line" >> $BASH_ENV
-            fi
-
-  # NOTE: UV version is defined in .uv-version file at repository root
-  install-docdep:
-    steps:
-      - run:
-          name: Read UV version from .uv-version
-          command: |
-            UV_VERSION=$(cat .uv-version)
-            echo "export UV_VERSION=$UV_VERSION" >> $BASH_ENV
-      # Python is managed by uv - it will automatically download the version
-      # specified in docs/.python-version when needed
-      - run:
-          name: Install uv
-          command: curl -LsSf https://astral.sh/uv/${UV_VERSION}/install.sh | sh
-      - run:
-          name: Add uv to PATH and set cache dir
-          command: |
-            echo 'export PATH="$HOME/.local/bin:$PATH"' >> $BASH_ENV
-            echo 'export UV_CACHE_DIR="$HOME/.cache/uv"' >> $BASH_ENV
-      - restore_cache:
-          name: Restore uv cache
-          keys:
-            - uv-cache-v1-${UV_VERSION}-{{ checksum "docs/uv.lock" }}
-            - uv-cache-v1-${UV_VERSION}-
-
-  save-uv-cache:
-    steps:
-      - run:
-          name: Prune uv cache for CI
-          command: uv cache prune --ci
-      - save_cache:
-          name: Save uv cache
-          key: uv-cache-v1-${UV_VERSION}-{{ checksum "docs/uv.lock" }}
-          paths:
-            - ~/.cache/uv
-
-  install-zfs-from-source:
-    parameters:
-      zfs_release:
-        type: string
-    steps:
-      - run:
-          name: Record kernel version for cache key
-          command: uname -r > /tmp/kernel-version
-      - restore_cache:
-          name: Restore ZFS native debs cache
-          keys:
-            - zfs-debs-v2-<<parameters.zfs_release>>-{{ checksum "/tmp/kernel-version" }}
-      - run:
-          # https://openzfs.github.io/openzfs-docs/Developer%20Resources/Building%20ZFS.html
-          name: Build ZFS <<parameters.zfs_release>> native debs (if not cached)
-          no_output_timeout: 20m
-          command: |
-            # CircleCI machine images have pyenv Python 3.13 shadowing the system
-            # Python 3.12. The apt python3-* packages (setuptools, cffi, etc.) only
-            # install for the system Python, and ZFS's dpkg-buildpackage needs them
-            # for --enable-pyzfs. Use the system Python so apt packages are visible.
-            export PYENV_VERSION=system
-            if [ -d /tmp/zfs-debs ]; then
-              echo "ZFS debs cache hit, skipping build"
-              exit 0
-            fi
-            sudo apt-get update
-            sudo apt-get install -y \
-              alien autoconf automake build-essential debhelper-compat dh-autoreconf \
-              dh-dkms dh-python dkms fakeroot gawk git libaio-dev libattr1-dev \
-              libblkid-dev libcurl4-openssl-dev libelf-dev libffi-dev libpam0g-dev \
-              libssl-dev libtirpc-dev libtool libudev-dev linux-headers-$(uname -r) lsb-release \
-              parallel po-debconf python3 python3-all-dev python3-cffi python3-dev \
-              python3-packaging python3-setuptools python3-sphinx uuid-dev zlib1g-dev
-            mkdir -p /tmp/zfs-build
-            cd /tmp/zfs-build
-            git clone --depth 1 --branch <<parameters.zfs_release>> https://github.com/openzfs/zfs.git zfs-src
-            cd zfs-src
-            sh autogen.sh
-            ./configure
-            make native-deb
-            mkdir -p /tmp/zfs-debs
-            find /tmp/zfs-build -name '*.deb' -exec mv -t /tmp/zfs-debs/ {} +
-      - save_cache:
-          name: Save ZFS native debs cache
-          key: zfs-debs-v2-<<parameters.zfs_release>>-{{ checksum "/tmp/kernel-version" }}
-          paths:
-            - /tmp/zfs-debs
-      - run:
-          name: Install ZFS <<parameters.zfs_release>>
-          command: |
-            ls /tmp/zfs-debs/
-            # Only install the debs we need. Exclude dracut (conflicts with
-            # initramfs-tools), dkms (we have prebuilt modules), and packages
-            # we don't need (pyzfs, test, doc).
-            sudo apt-get install -y $(find /tmp/zfs-debs -name '*.deb' \
-              ! -name '*dracut*' \
-              ! -name '*dkms*' \
-              ! -name '*initramfs*' \
-              ! -name '*pyzfs*' \
-              ! -name '*doc*' \
-              ! -name '*test*' \
-              -print)
-            sudo modprobe zfs
-      - run:
-          name: Verify ZFS version
-          command: |
-            sudo zfs version
-            sudo zpool version
-
-  docs-publish-sh:
-    parameters:
-      push:
-        type: boolean
-    steps:
-      - checkout
-      - run:
-          command: |
-            git config --global user.email "zreplbot@cschwarz.com"
-            git config --global user.name "zrepl-github-io-ci"
-
-      # Configure git to use the GitHub token for HTTPS authentication.
-      # The token is stored in the 'zrepl-github-io-deploy' context.
-      - when:
-          condition: << parameters.push >>
-          steps:
-            - run:
-                name: Configure git to use GitHub token for push
-                # GITHUB_PAGES_TOKEN is from the 'zrepl-github-io-deploy' context.
-                # CircleCI's secret masking automatically redacts context variables in logs.
-                command: |
-                  # Unset CircleCI's SSH URL rewriting that checkout step configured
-                  git config --global --unset-all url."ssh://git@github.com".insteadOf || true
-                  # Set up credential helper with GitHub token
-                  git config --global credential.helper store
-                  echo "https://x-access-token:${GITHUB_PAGES_TOKEN}@github.com" > ~/.git-credentials
-                  chmod 600 ~/.git-credentials
-
-      # caller must install-docdep
-      - when:
-          condition: << parameters.push >>
-          steps:
-            - run: bash -x docs/publish.sh -c -a -P
-      - when:
-          condition:
-            not: << parameters.push >>
-          steps:
-            - run: bash -x docs/publish.sh -c -a
-
-parameters:
-  do_ci:
-    type: boolean
-    default: true
-
-  do_release:
-    type: boolean
-    default: false
-
-workflows:
-  version: 2
-
-  ci:
-    when: << pipeline.parameters.do_ci >>
-    jobs:
-      - run-docs-publish-sh:
-          name: quickcheck-docs
-          push: false
-      - quickcheck-go:
-          name: quickcheck-go-amd64-linux-1.25.7
-          goversion: &latest-go-release "1.25.7"
-          goos: linux
-          goarch: amd64
-      - test-go:
-          goversion: *latest-go-release
-      - quickcheck-go:
-          requires:
-            - quickcheck-go-amd64-linux-1.25.7 #quickcheck-go-smoketest.name
-          matrix:
-            alias: quickcheck-go-matrix
-            parameters:
-              goversion: [*latest-go-release, "1.24.13"]
-              goos: ["linux", "freebsd"]
-              goarch: ["amd64", "arm64"]
-            exclude:
-              # don't re-do quickcheck-go-smoketest
-              - goversion: *latest-go-release
-                goos: linux
-                goarch: amd64
-      - platformtest:
-          matrix:
-            parameters:
-              goversion: [*latest-go-release]
-              goos: ["linux"]
-              goarch: ["amd64"]
-              zfs_release: ["zfs-2.2.9", "zfs-2.3.5", "zfs-2.4.0"]
-          requires:
-            - test-go
-            - quickcheck-go-<< matrix.goarch >>-<< matrix.goos >>-<< matrix.goversion >>
-
-  release:
-    when: << pipeline.parameters.do_release >>
-    jobs:
-      - release-build
-      - release-deb:
-          requires:
-            - release-build
-      - release-rpm:
-          requires:
-            - release-build
-      - release-upload:
-          requires:
-            - release-build
-            - release-deb
-            - release-rpm
-
-  publish-zrepl.github.io:
-    jobs:
-      - run-docs-publish-sh:
-          name: publish-zrepl.github.io
-          push: true
-          context:
-            - zrepl-github-io-deploy
-          filters:
-            branches:
-              only:
-                - master
-
-jobs:
-  run-docs-publish-sh:
-    parameters:
-      push:
-        type: boolean
-    docker:
-      - image: cimg/base:current
-    steps:
-      - checkout
-      - install-docdep
-      - docs-publish-sh:
-          push: << parameters.push >>
-      - save-uv-cache
-
-  quickcheck-go:
-    parameters:
-      goversion:
-        type: string
-      goos:
-        type: string
-      goarch:
-        type: string
-    docker:
-      - image: &cimg_with_modern_go cimg/go:1.25
-    environment:
-      GOOS: <<parameters.goos>>
-      GOARCH: <<parameters.goarch>>
-      GOTOOLCHAIN: "go<<parameters.goversion>>"
-
-    steps:
-      - checkout
-
-      - go/load-cache:
-          key: quickcheck-<<parameters.goversion>>
-      - run: make build/install
-      - run: go mod download
-      - run: cd build && go mod download
-      - go/save-cache:
-          key: quickcheck-<<parameters.goversion>>
-
-      # ensure all code has been generated
-      - run: make generate
-      - run: |
-          if output=$(git status --porcelain) && [ -z "$output" ]; then
-            echo "Working directory clean"
-          else
-            echo "Uncommitted changes"
-            echo ""
-            echo "$output"
-            exit 1
-          fi
-
-      # other checks
-      - run: make zrepl-bin test-platform-bin
-      - run: make vet
-      - run: make lint
-
-      - store_artifacts:
-          path: artifacts
-      - persist_to_workspace:
-          root: .
-          paths: [.]
-
-  platformtest:
-    parameters:
-      goversion:
-        type: string
-      goos:
-        type: string
-      goarch:
-        type: string
-      zfs_release:
-        type: string
-    machine:
-      # pinned (not :current) to keep ZFS build cache valid across runs
-      image: ubuntu-2404:2025.09.1
-    resource_class: medium
-    environment:
-      GOOS: <<parameters.goos>>
-      GOARCH: <<parameters.goarch>>
-    steps:
-      - attach_workspace:
-          at: .
-      - install-zfs-from-source:
-          zfs_release: <<parameters.zfs_release>>
-      - run: sudo make test-platform GOOS="$GOOS" GOARCH="$GOARCH"
-
-  test-go:
-    parameters:
-      goversion:
-        type: string
-    docker:
-      - image: *cimg_with_modern_go
-    environment:
-      GOTOOLCHAIN: "go<<parameters.goversion>>"
-    steps:
-      - checkout
-      - go/load-cache:
-          key: make-test-go
-      - run: make test-go
-      - go/save-cache:
-          key: make-test-go
-
-  release-build:
-    machine:
-      image: &release-vm-image "ubuntu-2404:current"
-    resource_class: large
-    steps:
-      - checkout
-      - run: make release-docker
-      - persist_to_workspace:
-          root: .
-          paths: [.]
-  release-deb:
-    machine:
-      image: *release-vm-image
-    steps:
-      - attach_workspace:
-          at: .
-      - run: make debs-docker
-      - persist_to_workspace:
-          root: .
-          paths:
-            - "artifacts/*.deb"
-
-  release-rpm:
-    machine:
-      image: *release-vm-image
-    steps:
-      - attach_workspace:
-          at: .
-      - run: make rpms-docker
-      - persist_to_workspace:
-          root: .
-          paths:
-            - "artifacts/*.rpm"
-
-  release-upload:
-    docker:
-      - image: cimg/base:2024.09
-    steps:
-      - attach_workspace:
-          at: .
-      - run: make wrapup-and-checksum
-      - store_artifacts:
-          path: artifacts/release

.circleci/download_artifacts.py

@@ -1,84 +0,0 @@
-import argparse
-from pathlib import Path
-import re
-
-import requests
-
-import time
-
-import os
-import argparse
-from pathlib import Path
-
-
-circle_token = os.environ.get('CIRCLE_TOKEN')
-if not circle_token:
-    raise ValueError('CIRCLE_TOKEN environment variable not set')
-
-parser = argparse.ArgumentParser(description='Download artifacts from CircleCI')
-parser.add_argument('build_num', type=str, help='Build number')
-parser.add_argument('dst', type=Path, help='Destination directory')
-parser.add_argument('--prefix', type=str, default='', help='Filter for prefix')
-parser.add_argument('--match', type=str, default='.*', help='Only include paths matching the given regex')
-
-args = parser.parse_args()
-
-res = requests.get(
-    f"https://circleci.com/api/v1.1/project/github/zrepl/zrepl/{args.build_num}/artifacts",
-    headers={
-        "Circle-Token": circle_token,
-    },
-)
-res.raise_for_status()
-
-# https://circleci.com/docs/api/v1/index.html#artifacts-of-a-job
-# [ {
-#   "path" : "raw-test-output/go-test-report.xml",
-#   "pretty_path" : "raw-test-output/go-test-report.xml",
-#   "node_index" : 0,
-#   "url" : "https://24-88881093-gh.circle-artifacts.com/0/raw-test-output/go-test-report.xml"
-# }, {
-#   "path" : "raw-test-output/go-test.out",
-#   "pretty_path" : "raw-test-output/go-test.out",
-#   "node_index" : 0,
-#   "url" : "https://24-88881093-gh.circle-artifacts.com/0/raw-test-output/go-test.out"
-# } ]
-res = res.json()
-
-for artifact in res:
-    if not artifact["pretty_path"].startswith(args.prefix):
-        continue
-    if not re.match(args.match, artifact["pretty_path"]):
-        continue
-    stripped = artifact["pretty_path"][len(args.prefix):]
-    print(f"Downloading {artifact['pretty_path']} to {args.dst / stripped}")
-    artifact_rel = Path(stripped)
-    artifact_dst = args.dst / artifact_rel
-    artifact_dst.parent.mkdir(parents=True, exist_ok=True)
-
-    res = requests.get(
-        artifact["url"],
-        headers={
-            "Circle-Token": circle_token,
-        },
-        stream=True,
-    )
-    res.raise_for_status()
-
-    total_size = int(res.headers.get("Content-Length", 0))
-    block_size = 128 * 1024
-    with open(artifact_dst, "wb") as f:
-        progress = 0
-        start_time = time.time()
-        for chunk in res.iter_content(chunk_size=block_size):
-            f.write(chunk)
-            progress += len(chunk)
-            percent = progress / total_size * 100
-            elapsed_time = time.time() - start_time
-            if elapsed_time >= 5:
-                print(f"Downloaded {progress}/{total_size} bytes ({percent:.2f}%)", end="\r")
-                start_time = time.time()
-        print(f"Downloaded {progress}/{total_size} bytes ({percent:.2f}%)")
-    print("Download complete!")
-
-print("All files downloaded")

.github/FUNDING.yml

@@ -1,5 +0,0 @@
-github: problame
-patreon: zrepl
-liberapay: zrepl
-custom: https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=R5QSXJVYHGX96
-ko_fi: problame

.github/dependabot.yml

@@ -1,9 +0,0 @@
-version: 2
-updates:
-  # Docs use Python only for static site generation.
-  - package-ecosystem: "pip"
-    directory: /docs
-    schedule:
-      interval: "weekly"
-    ignore:
-      - dependency-name: "*"

.gitignore

@@ -1,6 +1,5 @@
 # Build
 artifacts/
-build/install.tmp
 
 # Golang
 vendor/

.golangci.yml

@@ -1,42 +0,0 @@
-version: "2"
-linters:
-  enable:
-    - revive
-  settings:
-    revive:
-      rules:
-        - name: time-equal
-  exclusions:
-    generated: lax
-    presets:
-      - comments
-      - common-false-positives
-      - legacy
-      - std-error-handling
-    rules:
-      - linters:
-          - errcheck
-        path: _test\.go
-      - linters:
-          - staticcheck
-        text: 'SA9003:'
-      - linters:
-          - staticcheck
-        text: '(QF1001|QF1011|ST1012|QF1008|ST1005|ST1023|QF1003|ST1006|ST1001|QF1004):'
-    paths:
-      - third_party$
-      - builtin$
-      - examples$
-formatters:
-  enable:
-    - goimports
-  settings:
-    goimports:
-      local-prefixes:
-        - github.com/zrepl/zrepl
-  exclusions:
-    generated: lax
-    paths:
-      - third_party$
-      - builtin$
-      - examples$

.uv-version

@@ -1 +0,0 @@
-0.9.30

Gopkg.lock

@@ -0,0 +1,123 @@
+# This file is autogenerated, do not edit; changes may be undone by the next 'dep ensure'.
+
+
+[[projects]]
+  name = "github.com/davecgh/go-spew"
+  packages = ["spew"]
+  revision = "346938d642f2ec3594ed81d874461961cd0faa76"
+  version = "v1.1.0"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/dustin/go-humanize"
+  packages = ["."]
+  revision = "bb3d318650d48840a39aa21a027c6630e198e626"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/ftrvxmtrx/fd"
+  packages = ["."]
+  revision = "c6d800382fff6dc1412f34269f71b7f83bd059ad"
+
+[[projects]]
+  name = "github.com/go-logfmt/logfmt"
+  packages = ["."]
+  revision = "390ab7935ee28ec6b286364bba9b4dd6410cb3d5"
+  version = "v0.3.0"
+
+[[projects]]
+  branch = "v2"
+  name = "github.com/go-yaml/yaml"
+  packages = ["."]
+  revision = "eb3733d160e74a9c7e442f435eb3bea458e1d19f"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/jinzhu/copier"
+  packages = ["."]
+  revision = "db4671f3a9b8df855e993f7c94ec5ef1ffb0a23b"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/kr/logfmt"
+  packages = ["."]
+  revision = "b84e30acd515aadc4b783ad4ff83aff3299bdfe0"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/kr/pretty"
+  packages = ["."]
+  revision = "cfb55aafdaf3ec08f0db22699ab822c50091b1c4"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/kr/text"
+  packages = ["."]
+  revision = "7cafcd837844e784b526369c9bce262804aebc60"
+
+[[projects]]
+  name = "github.com/mattn/go-isatty"
+  packages = ["."]
+  revision = "0360b2af4f38e8d38c7fce2a9f4e702702d73a39"
+  version = "v0.0.3"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/mitchellh/mapstructure"
+  packages = ["."]
+  revision = "d0303fe809921458f417bcf828397a65db30a7e4"
+
+[[projects]]
+  name = "github.com/pkg/errors"
+  packages = ["."]
+  revision = "645ef00459ed84a119197bfb8d8205042c6df63d"
+  version = "v0.8.0"
+
+[[projects]]
+  name = "github.com/pmezard/go-difflib"
+  packages = ["difflib"]
+  revision = "792786c7400a136282c1664665ae0a8db921c6c2"
+  version = "v1.0.0"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/problame/go-netssh"
+  packages = ["."]
+  revision = "ffa145d2506e222977205e7666a9722d6b9959ac"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/problame/go-rwccmd"
+  packages = ["."]
+  revision = "391d2c78c8404a9683d79f75dd24ab53040f89f7"
+
+[[projects]]
+  branch = "master"
+  name = "github.com/spf13/cobra"
+  packages = ["."]
+  revision = "b78744579491c1ceeaaa3b40205e56b0591b93a3"
+
+[[projects]]
+  name = "github.com/spf13/pflag"
+  packages = ["."]
+  revision = "e57e3eeb33f795204c1ca35f56c44f83227c6e66"
+  version = "v1.0.0"
+
+[[projects]]
+  name = "github.com/stretchr/testify"
+  packages = ["assert"]
+  revision = "69483b4bd14f5845b5a1e55bca19e954e827f1d0"
+  version = "v1.1.4"
+
+[[projects]]
+  branch = "master"
+  name = "golang.org/x/sys"
+  packages = ["unix"]
+  revision = "bf42f188b9bc6f2cf5b8ee5a912ef1aedd0eba4c"
+
+[solve-meta]
+  analyzer-name = "dep"
+  analyzer-version = 1
+  inputs-digest = "fcfc66322b5f9dbe329411b7688a4e9dcebe3c529c1dadf4f78bcdb2bcfb8fbf"
+  solver-name = "gps-cdcl"
+  solver-version = 1

Gopkg.toml

@@ -0,0 +1,53 @@
+ignored = [ "github.com/inconshreveable/mousetrap" ]
+
+[[constraint]]
+  branch = "master"
+  name = "github.com/ftrvxmtrx/fd"
+
+[[constraint]]
+  branch = "master"
+  name = "github.com/jinzhu/copier"
+
+[[constraint]]
+  branch = "master"
+  name = "github.com/kr/pretty"
+
+[[constraint]]
+  branch = "master"
+  name = "github.com/mitchellh/go-homedir"
+
+[[constraint]]
+  branch = "master"
+  name = "github.com/mitchellh/mapstructure"
+
+[[constraint]]
+  name = "github.com/pkg/errors"
+  version = "0.8.0"
+
+[[constraint]]
+  branch = "master"
+  name = "github.com/spf13/cobra"
+
+[[constraint]]
+  name = "github.com/spf13/viper"
+  version = "1.0.0"
+
+[[constraint]]
+  name = "github.com/stretchr/testify"
+  version = "1.1.4"
+
+[[constraint]]
+  branch = "v2"
+  name = "github.com/go-yaml/yaml"
+
+[[constraint]]
+  name = "github.com/go-logfmt/logfmt"
+  version = "*"
+
+[[constraint]]
+  name = "github.com/problame/go-rwccmd"
+  branch = "master"
+
+[[constraint]]
+  name = "github.com/problame/go-netssh"
+  branch = "master"

LICENSE

@@ -1,6 +1,6 @@
 MIT License
 
-Copyright (c) 2017-2018 Christian Schwarz
+Copyright (c) 2017 Christian Schwarz
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

Makefile

@@ -1,391 +1,106 @@
-.PHONY: generate build test vet cover release docs docs-clean clean format lint platformtest
+.PHONY: generate build test vet cover release docs docs-clean clean vendordeps
-.PHONY: release release-noarch
+.DEFAULT_GOAL := build
-.DEFAULT_GOAL := zrepl-bin
+
+ROOT := github.com/zrepl/zrepl
+SUBPKGS := cmd logger rpc util zfs
+
+_TESTPKGS := $(ROOT) $(foreach p,$(SUBPKGS),$(ROOT)/$(p))
 
 ARTIFACTDIR := artifacts
 
-ifdef ZREPL_VERSION
+ifndef ZREPL_VERSION
-    _ZREPL_VERSION := $(ZREPL_VERSION)
+    ZREPL_VERSION := $(shell git describe --dirty 2>/dev/null || echo "ZREPL_BUILD_INVALID_VERSION" )
-endif
+    ifeq ($(ZREPL_VERSION),ZREPL_BUILD_INVALID_VERSION) # can't use .SHELLSTATUS because Debian Stretch is still on gmake 4.1
-ifndef _ZREPL_VERSION
-    _ZREPL_VERSION := $(shell git describe --always --dirty 2>/dev/null || echo "ZREPL_BUILD_INVALID_VERSION" )
-    ifeq ($(_ZREPL_VERSION),ZREPL_BUILD_INVALID_VERSION) # can't use .SHELLSTATUS because Debian Stretch is still on gmake 4.1
         $(error cannot infer variable ZREPL_VERSION using git and variable is not overriden by make invocation)
     endif
 endif
+GO_LDFLAGS := "-X github.com/zrepl/zrepl/cmd.zreplVersion=$(ZREPL_VERSION)"
 
-ZREPL_PACKAGE_RELEASE := 1
+GO_BUILD := go build -ldflags $(GO_LDFLAGS)
 
-GO := go
+RELEASE_BINS := $(ARTIFACTDIR)/zrepl-freebsd-amd64 $(ARTIFACTDIR)/zrepl-linux-amd64
-GOOS ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOOS"')
+RELEASE_NOARCH := $(ARTIFACTDIR)/zrepl-noarch.tar
-GOARCH ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOARCH"')
+THIS_PLATFORM_RELEASE_BIN := $(shell bash -c 'source <(go env) && echo "zrepl-$${GOOS}-$${GOARCH}"' )
-GOARM ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOARM"')
-GOHOSTOS ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOHOSTOS"')
-GOHOSTARCH ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOHOSTARCH"')
-GO_ENV_VARS := CGO_ENABLED=0
-GO_LDFLAGS := "-X github.com/zrepl/zrepl/internal/version.zreplVersion=$(_ZREPL_VERSION)"
-GO_MOD_READONLY := -mod=readonly
-GO_EXTRA_BUILDFLAGS :=
-GO_BUILDFLAGS := $(GO_MOD_READONLY) $(GO_EXTRA_BUILDFLAGS)
-GO_BUILD := $(GO_ENV_VARS) $(GO) build $(GO_BUILDFLAGS) -ldflags $(GO_LDFLAGS)
-GOCOVMERGE := gocovmerge
-RELEASE_GOVERSION ?= go1.25.7
-STRIPPED_GOVERSION := $(subst go,,$(RELEASE_GOVERSION))
-RELEASE_DOCKER_BASEIMAGE ?= golang:$(STRIPPED_GOVERSION)
-RELEASE_DOCKER_CACHEMOUNT :=
 
-ifneq ($(GOARM),)
+vendordeps:
-	ZREPL_TARGET_TUPLE := $(GOOS)-$(GOARCH)v$(GOARM)
+	dep ensure -v -vendor-only
-else
-	ZREPL_TARGET_TUPLE := $(GOOS)-$(GOARCH)
-endif
 
+generate: #not part of the build, must do that manually
+	@for pkg in $(_TESTPKGS); do\
+		go generate "$$pkg" || exit 1; \
+	done;
 
-ifneq ($(RELEASE_DOCKER_CACHEMOUNT),)
+build:
-	_RELEASE_DOCKER_CACHEMOUNT := -v $(RELEASE_DOCKER_CACHEMOUNT)/mod:/go/pkg/mod -v $(RELEASE_DOCKER_CACHEMOUNT)/xdg-cache:/.cache/go-build
+	@echo "INFO: In case of missing dependencies, run 'make vendordeps'"
-.PHONY: release-docker-mkcachemount
+	$(GO_BUILD) -o "$(ARTIFACTDIR)/zrepl"
-release-docker-mkcachemount:
-	mkdir -p $(RELEASE_DOCKER_CACHEMOUNT)
-	mkdir -p $(RELEASE_DOCKER_CACHEMOUNT)/mod
-	mkdir -p $(RELEASE_DOCKER_CACHEMOUNT)/xdg-cache
-else
-	_RELEASE_DOCKER_CACHEMOUNT :=
-.PHONY: release-docker-mkcachemount
-release-docker-mkcachemount:
-	# nothing to do
-endif
 
-##################### PRODUCING A RELEASE #############
+test:
-.PHONY: release wrapup-and-checksum check-git-clean sign clean ensure-release-toolchain
+	@for pkg in $(_TESTPKGS); do \
-
+		echo "Testing $$pkg"; \
-ensure-release-toolchain:
+		go test "$$pkg" || exit 1; \
-	# ensure the toolchain is actually the one we expect
+	done;
-	test $(RELEASE_GOVERSION) = "$$($(GO_ENV_VARS) $(GO) env GOVERSION)"
-
-release: ensure-release-toolchain
-	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="vet"
-	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="lint"
-	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="zrepl-bin"
-	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="test-platform-bin"
-	$(MAKE) noarch
-
-release-docker: $(ARTIFACTDIR) release-docker-mkcachemount
-	sed 's/FROM.*!SUBSTITUTED_BY_MAKEFILE/FROM $(RELEASE_DOCKER_BASEIMAGE)/' build/build.Dockerfile > $(ARTIFACTDIR)/build.Dockerfile
-	docker build -t zrepl_release --pull \
-		--build-arg BUILD_UID=$$(id -u) \
-		--build-arg BUILD_GID=$$(id -g) \
-		-f $(ARTIFACTDIR)/build.Dockerfile .
-	docker run --rm -i$$(test -t 0 && echo t) \
-		$(_RELEASE_DOCKER_CACHEMOUNT) \
-		-v $(CURDIR):/src \
-		zrepl_release \
-		make release \
-			GOOS=$(GOOS) GOARCH=$(GOARCH) GOARM=$(GOARM) \
-			ZREPL_VERSION=$(ZREPL_VERSION) ZREPL_PACKAGE_RELEASE=$(ZREPL_PACKAGE_RELEASE) \
-			RELEASE_GOVERSION=$(RELEASE_GOVERSION)
-
-debs-docker:
-	$(MAKE) _debs_or_rpms_docker _DEB_OR_RPM=deb
-rpms-docker:
-	$(MAKE) _debs_or_rpms_docker _DEB_OR_RPM=rpm
-_debs_or_rpms_docker: # artifacts/_zrepl.zsh_completion artifacts/bash_completion docs zrepl-bin
-	$(MAKE) $(_DEB_OR_RPM)-docker GOOS=linux GOARCH=amd64
-	$(MAKE) $(_DEB_OR_RPM)-docker GOOS=linux GOARCH=arm64
-	$(MAKE) $(_DEB_OR_RPM)-docker GOOS=linux GOARCH=arm GOARM=7
-	$(MAKE) $(_DEB_OR_RPM)-docker GOOS=linux GOARCH=386
-
-rpm: $(ARTIFACTDIR) # artifacts/_zrepl.zsh_completion artifacts/bash_completion docs zrepl-bin
-	$(eval _ZREPL_RPM_VERSION := $(subst -,.,$(_ZREPL_VERSION)))
-	$(eval _ZREPL_RPM_TOPDIR_ABS := $(CURDIR)/$(ARTIFACTDIR)/rpmbuild)
-	rm -rf "$(_ZREPL_RPM_TOPDIR_ABS)"
-	mkdir "$(_ZREPL_RPM_TOPDIR_ABS)"
-	for d in BUILD BUILDROOT RPMS SOURCES SPECS SRPMS; do \
-		mkdir -p "$(_ZREPL_RPM_TOPDIR_ABS)/$$d"; \
-	done
-	sed \
-		-e "s/^Version:.*/Version:          $(_ZREPL_RPM_VERSION)/g" \
-		-e "s/^Release:.*/Release:          $(ZREPL_PACKAGE_RELEASE)/g" \
-	    packaging/rpm/zrepl.spec \
-		>  $(_ZREPL_RPM_TOPDIR_ABS)/SPECS/zrepl.spec
-
-	# see /usr/lib/rpm/platform
-ifeq ($(GOARCH),amd64)
-	$(eval _ZREPL_RPMBUILD_TARGET := x86_64)
-else ifeq ($(GOARCH), 386)
-	$(eval _ZREPL_RPMBUILD_TARGET := i386)
-else ifeq ($(GOARCH), arm64)
-	$(eval _ZREPL_RPMBUILD_TARGET := aarch64)
-else ifeq ($(GOARCH), arm)
-	$(eval _ZREPL_RPMBUILD_TARGET := armv7hl)
-else
-	$(eval _ZREPL_RPMBUILD_TARGET := $(GOARCH))
-endif
-	rpmbuild \
-		--build-in-place \
-		--define "_sourcedir $(CURDIR)" \
-		--define "_topdir $(_ZREPL_RPM_TOPDIR_ABS)" \
-		--define "_zrepl_binary_filename zrepl-$(ZREPL_TARGET_TUPLE)" \
-		--target $(_ZREPL_RPMBUILD_TARGET) \
-		-bb "$(_ZREPL_RPM_TOPDIR_ABS)"/SPECS/zrepl.spec
-	cp "$(_ZREPL_RPM_TOPDIR_ABS)"/RPMS/$(_ZREPL_RPMBUILD_TARGET)/zrepl-$(_ZREPL_RPM_VERSION)*.rpm $(ARTIFACTDIR)/
-
-rpm-docker:
-	docker build -t zrepl_rpm_pkg --pull -f packaging/rpm/Dockerfile .
-	docker run --rm -i -v $(CURDIR):/build/src -u $$(id -u):$$(id -g) \
-		zrepl_rpm_pkg \
-		make rpm \
-			GOOS=$(GOOS) GOARCH=$(GOARCH) GOARM=$(GOARM) \
-			ZREPL_VERSION=$(ZREPL_VERSION) ZREPL_PACKAGE_RELEASE=$(ZREPL_PACKAGE_RELEASE)
-
-
-deb: $(ARTIFACTDIR) # artifacts/_zrepl.zsh_completion artifacts/bash_completion docs zrepl-bin
-
-	cp packaging/deb/debian/changelog.template packaging/deb/debian/changelog
-	sed -i 's/DATE_DASH_R_OUTPUT/$(shell date -R)/' packaging/deb/debian/changelog
-	VERSION="$(subst -,.,$(_ZREPL_VERSION))-$(ZREPL_PACKAGE_RELEASE)"; \
-		export VERSION="$${VERSION#v}"; \
-		sed -i 's/VERSION/'"$$VERSION"'/' packaging/deb/debian/changelog
-
-ifeq ($(GOARCH), arm)
-	$(eval  DEB_HOST_ARCH := armhf)
-else ifeq ($(GOARCH), 386)
-	$(eval DEB_HOST_ARCH := i386)
-else
-	$(eval DEB_HOST_ARCH := $(GOARCH))
-endif
-
-	export ZREPL_DPKG_ZREPL_BINARY_FILENAME=zrepl-$(ZREPL_TARGET_TUPLE); \
-		dpkg-buildpackage -b --no-sign --host-arch $(DEB_HOST_ARCH)
-	cp ../*.deb artifacts/
-
-deb-docker:
-	docker build -t zrepl_debian_pkg --pull -f packaging/deb/Dockerfile .
-	# Use a small open file limit to make fakeroot work. If we don't
-	# specify it, docker daemon will use its file limit. I don't know
-	# what changed (Docker, its systemd service, its Go version). But I
-	# observed fakeroot iterating close(i) up to i > 1000000, which costs
-	# a good amount of CPU time and makes the build slow.
-	docker run --rm -i -v $(CURDIR):/build/src -u $$(id -u):$$(id -g) \
-		--ulimit nofile=1024:1024 \
-		zrepl_debian_pkg \
-		make deb \
-			GOOS=$(GOOS) GOARCH=$(GOARCH) GOARM=$(GOARM) \
-			ZREPL_VERSION=$(ZREPL_VERSION) ZREPL_PACKAGE_RELEASE=$(ZREPL_PACKAGE_RELEASE)
-
-# expects `release`, `deb` & `rpm` targets to have run before
-NOARCH_TARBALL := $(ARTIFACTDIR)/zrepl-noarch.tar
-wrapup-and-checksum:
-	rm -f $(NOARCH_TARBALL)
-	tar --mtime='1970-01-01' --sort=name \
-		--transform 's/$(ARTIFACTDIR)/zrepl-$(_ZREPL_VERSION)-noarch/' \
-		--transform 's#dist#zrepl-$(_ZREPL_VERSION)-noarch/dist#' \
-		--transform 's#internal/config/samples#zrepl-$(_ZREPL_VERSION)-noarch/config#' \
-		-acf $(NOARCH_TARBALL) \
-		$(ARTIFACTDIR)/docs/html \
-		$(ARTIFACTDIR)/bash_completion \
-		$(ARTIFACTDIR)/_zrepl.zsh_completion \
-		$(ARTIFACTDIR)/go_env.txt \
-		dist \
-		internal/config/samples
-	rm -rf "$(ARTIFACTDIR)/release"
-	mkdir -p "$(ARTIFACTDIR)/release"
-	cp -l $(ARTIFACTDIR)/zrepl* \
-		$(ARTIFACTDIR)/platformtest-* \
-		"$(ARTIFACTDIR)/release"
-	cd "$(ARTIFACTDIR)/release" && sha512sum $$(ls | sort) > sha512sum.txt
-
-check-git-clean:
-	@# note that we use ZREPL_VERSION and not _ZREPL_VERSION because we want to detect the override
-	@if git describe --always --dirty 2>/dev/null | grep dirty >/dev/null; then \
-        echo '[INFO] either git reports checkout is dirty or git is not installed or this is not a git checkout'; \
-		if [ "$(ZREPL_VERSION)" = "" ]; then \
-			echo '[WARN] git checkout is dirty and make variable ZREPL_VERSION was not used to override'; \
-			git status; \
-			echo "git diff:";  \
-			git diff | cat; \
-			exit 1; \
-		fi; \
-	fi;
-
-tag-release:
-	test -n "$(ZREPL_TAG_VERSION)" || exit 1
-	git tag -u '328A6627FA98061D!' -m "$(ZREPL_TAG_VERSION)" "$(ZREPL_TAG_VERSION)"
-
-sign:
-	gpg -u '328A6627FA98061D!' \
-		--armor \
-		--detach-sign $(ARTIFACTDIR)/release/sha512sum.txt
-
-clean: docs-clean
-	rm -rf "$(ARTIFACTDIR)"
-
-download-circleci-release:
-	rm -rf "$(ARTIFACTDIR)"
-	mkdir -p "$(ARTIFACTDIR)/release"
-	python3 .circleci/download_artifacts.py --prefix 'artifacts/release/' "$(BUILD_NUM)" "$(ARTIFACTDIR)/release"
-
-##################### MULTI-ARCH HELPERS #####################
-
-_run_make_foreach_target_tuple:
-	if [ "$(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG)" = "" ]; then \
-		echo "RUN_MAKE_FOREACH_TARGET_TUPLE_ARG must be set"; \
-		exit 1; \
-	fi
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=amd64
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=386
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=arm	  GOARM=7
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=arm64
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=amd64
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=arm64
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=arm     GOARM=7
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=386
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=darwin    GOARCH=amd64
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=solaris   GOARCH=amd64
-	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=illumos   GOARCH=amd64
-
-##################### REGULAR TARGETS #####################
-.PHONY: lint test-go test-platform cover-merge cover-html vet zrepl-bin test-platform-bin
-
-lint: build/install
-	$(GO_ENV_VARS) build/install/golangci_lint/golangci-lint run ./...
 
 vet:
-	$(GO_ENV_VARS) $(GO) vet $(GO_BUILDFLAGS) ./...
+	@for pkg in $(_TESTPKGS); do \
-
+		echo "Vetting $$pkg"; \
-test-go: $(ARTIFACTDIR)
+		go vet "$$pkg" || exit 1; \
-	rm -f "$(ARTIFACTDIR)/gotest.cover"
+	done;
-ifeq ($(COVER),1)
-	$(GO_ENV_VARS) $(GO) test $(GO_BUILDFLAGS) \
-		-coverpkg github.com/zrepl/zrepl/... \
-		-covermode atomic \
-		-coverprofile "$(ARTIFACTDIR)/gotest.cover" \
-		./...
-else
-	$(GO_ENV_VARS) $(GO) test $(GO_BUILDFLAGS) \
-		./...
-endif
-
-zrepl-bin:
-	$(GO_BUILD) -o "$(ARTIFACTDIR)/zrepl-$(ZREPL_TARGET_TUPLE)"
-
-COVER_PLATFORM_BIN_PATH := $(ARTIFACTDIR)/platformtest-cover-$(ZREPL_TARGET_TUPLE)
-cover-platform-bin:
-	$(GO_ENV_VARS) $(GO) test $(GO_BUILDFLAGS) \
-		-c -o "$(COVER_PLATFORM_BIN_PATH)" \
-		-covermode=atomic -cover -coverpkg github.com/zrepl/zrepl/... \
-		./platformtest/harness
-cover-platform:
-	# do not track dependency on cover-platform-bin to allow build of binary outside of test VM
-	export _TEST_PLATFORM_CMD="$(COVER_PLATFORM_BIN_PATH) \
-		-test.coverprofile \"$(ARTIFACTDIR)/platformtest.cover\" \
-		-test.v \
-		__DEVEL--i-heard-you-like-tests"; \
-		$(MAKE) _test-or-cover-platform-impl
-
-TEST_PLATFORM_BIN_PATH := $(ARTIFACTDIR)/platformtest-$(ZREPL_TARGET_TUPLE)
-test-platform-bin:
-	$(GO_BUILD) -o "$(TEST_PLATFORM_BIN_PATH)" ./internal/platformtest/harness
-test-platform:
-	export _TEST_PLATFORM_CMD="\"$(TEST_PLATFORM_BIN_PATH)\""; \
-		$(MAKE) _test-or-cover-platform-impl
-
-ZREPL_PLATFORMTEST_POOLNAME := zreplplatformtest
-ZREPL_PLATFORMTEST_IMAGEPATH := /tmp/zreplplatformtest.pool.img
-ZREPL_PLATFORMTEST_MOUNTPOINT := /tmp/zreplplatformtest.pool
-ZREPL_PLATFORMTEST_ZFS_LOG := /tmp/zreplplatformtest.zfs.log
-# ZREPL_PLATFORMTEST_STOP_AND_KEEP := -failure.stop-and-keep-pool
-ZREPL_PLATFORMTEST_ARGS :=
-_test-or-cover-platform-impl: $(ARTIFACTDIR)
-ifndef _TEST_PLATFORM_CMD
-	$(error _TEST_PLATFORM_CMD is undefined, caller 'cover-platform' or 'test-platform' should have defined it)
-endif
-	rm -f "$(ZREPL_PLATFORMTEST_ZFS_LOG)"
-	rm -f "$(ARTIFACTDIR)/platformtest.cover"
-	internal/platformtest/logmockzfs/logzfsenv "$(ZREPL_PLATFORMTEST_ZFS_LOG)" `which zfs` \
-		$(_TEST_PLATFORM_CMD) \
-		-poolname "$(ZREPL_PLATFORMTEST_POOLNAME)" \
-		-imagepath "$(ZREPL_PLATFORMTEST_IMAGEPATH)" \
-		-mountpoint "$(ZREPL_PLATFORMTEST_MOUNTPOINT)" \
-		$(ZREPL_PLATFORMTEST_STOP_AND_KEEP) \
-		$(ZREPL_PLATFORMTEST_ARGS)
-
-cover-merge: $(ARTIFACTDIR)
-	$(GOCOVMERGE) $(ARTIFACTDIR)/platformtest.cover $(ARTIFACTDIR)/gotest.cover > $(ARTIFACTDIR)/merged.cover
-cover-html: cover-merge
-	$(GO) tool cover -html "$(ARTIFACTDIR)/merged.cover" -o "$(ARTIFACTDIR)/merged.cover.html"
-
-cover-full:
-	test "$$(id -u)" = "0" || echo "MUST RUN AS ROOT" 1>&2
-	$(MAKE) test-go COVER=1
-	$(MAKE) cover-platform-bin
-	$(MAKE) cover-platform
-	$(MAKE) cover-html
-
-##################### DEV TARGETS #####################
-# not part of the build, must do that manually
-.PHONY: generate format
-
-build/install:
-	rm -rf build/install.tmp
-	mkdir build/install.tmp
-
-	-echo "installing protoc"
-	mkdir build/install.tmp/protoc
-	bash -x build/get_protoc.bash build/install.tmp/protoc
-
-	-echo "installing golangci-lint"
-	mkdir -p build/install.tmp/golangci_lint
-	GOHOSTARCH=$(GOHOSTARCH) \
-		build/get_golangci_lint.bash build/install.tmp/golangci_lint
-
-	-echo "installing go tools"
-	build/go_install_tools.bash build/install.tmp/gobin
-
-	mv build/install.tmp build/install
-
-
-generate: build/install
-	# TODO: would be nice to run with a pure path here
-	PATH="$(CURDIR)/build/install/gobin:$(CURDIR)/build/install/protoc/bin:$$PATH" && \
-		build/install/protoc/bin/protoc -I=internal/replication/logic/pdu --go_out=internal/replication/logic/pdu --go-grpc_out=internal/replication/logic/pdu internal/replication/logic/pdu/pdu.proto && \
-		build/install/protoc/bin/protoc -I=internal/rpc/grpcclientidentity/example --go_out=internal/rpc/grpcclientidentity/example/pdu --go-grpc_out=internal/rpc/grpcclientidentity/example/pdu internal/rpc/grpcclientidentity/example/grpcauth.proto && \
-		$(GO) generate $(GO_BUILDFLAGS) -x ./... && \
-		true
-
-GOIMPORTS := goimports -srcdir . -local 'github.com/zrepl/zrepl'
-
-format: build/install
-	@ build/install/gobin/goimports -w -local 'github.com/zrepl/zrepl' $$(find . -type f -name '*.go' -not -path "./vendor/*" -not -name '*.pb.go' -not -name '*_enumer.go')
-
-##################### NOARCH #####################
-.PHONY: noarch $(ARTIFACTDIR)/bash_completion $(ARTIFACTDIR)/_zrepl.zsh_completion $(ARTIFACTDIR)/go_env.txt docs docs-clean
 
+cover: artifacts
+	@for pkg in $(_TESTPKGS); do \
+		profile="$(ARTIFACTDIR)/cover-$$(basename $$pkg).out"; \
+		go test -coverprofile "$$profile" $$pkg || exit 1; \
+		if [ -f "$$profile" ]; then \
+   			go tool cover -html="$$profile" -o "$${profile}.html" || exit 2; \
+		fi; \
+	done;
 
 $(ARTIFACTDIR):
 	mkdir -p "$@"
+
 $(ARTIFACTDIR)/docs: $(ARTIFACTDIR)
 	mkdir -p "$@"
 
-noarch: $(ARTIFACTDIR)/bash_completion $(ARTIFACTDIR)/_zrepl.zsh_completion $(ARTIFACTDIR)/go_env.txt docs
+$(ARTIFACTDIR)/bash_completion: $(RELEASE_BINS)
-	# pass
+	artifacts/$(THIS_PLATFORM_RELEASE_BIN) bashcomp "$@"
 
-$(ARTIFACTDIR)/bash_completion:
+$(ARTIFACTDIR)/go_version.txt:
-	$(MAKE) zrepl-bin GOOS=$(GOHOSTOS) GOARCH=$(GOHOSTARCH)
+	go version > $@
-	artifacts/zrepl-$(GOHOSTOS)-$(GOHOSTARCH) gencompletion bash "$@"
-
-$(ARTIFACTDIR)/_zrepl.zsh_completion:
-	$(MAKE) zrepl-bin GOOS=$(GOHOSTOS) GOARCH=$(GOHOSTARCH)
-	artifacts/zrepl-$(GOHOSTOS)-$(GOHOSTARCH) gencompletion zsh "$@"
-
-$(ARTIFACTDIR)/go_env.txt:
-	$(GO_ENV_VARS) $(GO) env > $@
-	$(GO) version >> $@
 
 docs: $(ARTIFACTDIR)/docs
-	# https://www.sphinx-doc.org/en/master/man/sphinx-build.html
+	make -C docs \
-	cd docs && uv sync --frozen
+		html \
-	cd docs && uv run sphinx-build -W --keep-going -n . ../artifacts/docs/html
+		BUILDDIR=../artifacts/docs \
 
 docs-clean:
-	rm -rf artifacts/docs
+	make -C docs \
-	rm -rf docs/.venv
+		clean \
+		BUILDDIR=../artifacts/docs
+
+
+.PHONY: $(RELEASE_BINS)
+# TODO: two wildcards possible
+$(RELEASE_BINS): $(ARTIFACTDIR)/zrepl-%-amd64: generate $(ARTIFACTDIR) vet test
+	@echo "INFO: In case of missing dependencies, run 'make vendordeps'"
+	GOOS=$* GOARCH=amd64   $(GO_BUILD) -o "$(ARTIFACTDIR)/zrepl-$*-amd64"
+
+$(RELEASE_NOARCH): docs $(ARTIFACTDIR)/bash_completion $(ARTIFACTDIR)/go_version.txt
+	tar --mtime='1970-01-01' --sort=name \
+		--transform 's/$(ARTIFACTDIR)/zrepl-$(ZREPL_VERSION)-noarch/' \
+		-acf $@ \
+		$(ARTIFACTDIR)/docs/html \
+		$(ARTIFACTDIR)/bash_completion \
+		$(ARTIFACTDIR)/go_version.txt
+
+release: $(RELEASE_BINS) $(RELEASE_NOARCH)
+	rm -rf "$(ARTIFACTDIR)/release"
+	mkdir -p "$(ARTIFACTDIR)/release"
+	cp $^ "$(ARTIFACTDIR)/release"
+	cd "$(ARTIFACTDIR)/release" && sha512sum $$(ls | sort) > sha512sum.txt
+	@if echo "$(ZREPL_VERSION)" | grep dirty > /dev/null; then\
+		echo '[WARN] Do not publish the artifacts, make variable ZREPL_VERSION=$(ZREPL_VERSION) indicates they are dirty!'; \
+		exit 1; \
+	fi
+
+clean: docs-clean
+	rm -rf "$(ARTIFACTDIR)"

README.md

@@ -1,15 +1,5 @@
-[![GitHub license](https://img.shields.io/github/license/zrepl/zrepl.svg)](https://github.com/zrepl/zrepl/blob/master/LICENSE)
-[![Language: Go](https://img.shields.io/badge/lang-Go-6ad7e5.svg)](https://golang.org/)
-[![User Docs](https://img.shields.io/badge/docs-web-blue.svg)](https://zrepl.github.io)
-[![Support me on Patreon](https://img.shields.io/badge/dynamic/json?color=yellow&label=Patreon&query=data.attributes.patron_count&url=https%3A%2F%2Fwww.patreon.com%2Fapi%2Fcampaigns%2F3095079)](https://patreon.com/zrepl)
-[![Donate via GitHub Sponsors](https://img.shields.io/static/v1?label=Sponsor&message=%E2%9D%A4&logo=GitHub&style=flat&color=yellow)](https://github.com/sponsors/problame)
-[![Donate via Liberapay](https://img.shields.io/liberapay/patrons/zrepl.svg?logo=liberapay)](https://liberapay.com/zrepl/donate)
-[![Donate via PayPal](https://img.shields.io/badge/donate-paypal-yellow.svg)](https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=R5QSXJVYHGX96)
-[![Twitter](https://img.shields.io/twitter/url/https/github.com/zrepl/zrepl.svg?style=social)](https://twitter.com/intent/tweet?text=Wow:&url=https%3A%2F%2Fgithub.com%2Fzrepl%2Fzrepl)
-[![Chat](https://img.shields.io/badge/chat-matrix-blue.svg)](https://matrix.to/#/#zrepl:matrix.org)
-
 # zrepl
-zrepl is a one-stop ZFS backup & replication solution.
+zrepl is a ZFS filesystem backup & replication solution written in Go.
 
 ## User Documentation
 
@@ -22,182 +12,78 @@ zrepl is a one-stop ZFS backup & replication solution.
 
 ## Feature Requests
 
-1. Does your feature request require default values / some kind of configuration?
+1. Does you feature request require default values / some kind of configuration?
    If so, think of an expressive configuration example.
 2. Think of at least one use case that generalizes from your concrete application.
 3. Open an issue on GitHub with example conf & use case attached.
-4. **Optional**: [Contact Christian Schwarz](https://cschwarz.com) for contract work.
 
 The above does not apply if you already implemented everything.
 Check out the *Coding Workflow* section below for details.
 
-## Development
+## Package Maintainer Information
 
-zrepl is written in [Go](https://golang.org) and uses [Go modules](https://github.com/golang/go/wiki/Modules) to manage dependencies.
+* Follow the steps in `docs/installation.rst -> Compiling from Source` and read the Makefile / shell scripts used in this process.
-The documentation is written in [ReStructured Text](http://docutils.sourceforge.net/rst.html) using the [Sphinx](https://www.sphinx-doc.org) framework.
+* Make sure your distro is compatible with the paths in `docs/installation.rst`.
-
-### Building
-
-#### Go Code
-
-Dependencies:
-
-* Go
-* GNU Make
-* Git
-* wget (`make generate`)
-* unzip (`make generate`)
-
-Some Go code is **generated**, and generated code is committed to the source tree.
-Therefore, building does not require having code generation tools set up.
-When making changes that require code to be (re-)generated, run `make generate`.
-I downloads and installs pinned versions of the code generation tools into `./build/install`.
-There is a CI check that ensures Git state is clean, i.e., code generation has been done by a PR and is deterministic.
-
-#### Docs
-
-Install [uv](https://docs.astral.sh/uv/getting-started/installation/), then run `make docs`.
-uv automatically manages Python and dependencies.
-
-### Testing
-
-The **test suite** is split into pure **Go tests** (`make test-go`) and **platform tests** that interact with ZFS and thus generally **require root privileges** (`sudo make test-platform`).
-Platform tests run on their own pool with the name `zreplplatformtest`, which is created using the file vdev in `/tmp`.
-
-For a full **code coverage** profile, run `make test-go COVER=1 && sudo make test-platform && make cover-merge`.
-An HTML report can be generated using `make cover-html`.
-
-### Circle CI
-
-We use CircleCI for automated build & test pre- and post-merge.
-
-There are two workflows:
-
-* `ci` runs for every commit / branch / tag pushed to GitHub.
-  It is supposed to run very fast (<5min and provides quick feedback to developers).
-  It runs formatting checks, lints and tests on the most important OSes / architectures.
-
-* `release` runs
-  * on manual triggers through the CircleCI API (in order to produce a release)
-  * periodically on `master`
-
-Artifacts are stored in CircleCI.
-
-### Releasing
-
-All zrepl releases are git-tagged and then published as a GitHub Release.
-There is a git tag for each zrepl release, usually `vMAJOR.MINOR.0`.
-We don't move git tags once the release has been published.
-
-The procedure to issue a release is as follows:
-
-* Prepare the release (as a PR to `master`):
-  * Finalize `docs/changelog.rst` for the release.
-  * Merge the PR. Docs are auto-published to zrepl.github.io on merge.
-* Tag the release:
-  * Git tag the release on the `master` branch (e.g., `vMAJOR.MINOR.0`).
-  * Push the tag.
-* Build and publish:
-  * Run the `release` pipeline (trigger via CircleCI UI).
-  * Download artifacts: `make download-circleci-release BUILD_NUM=<circleci-build-number>`
-  * Create GitHub release and upload artifacts:
-    ```bash
-    gh release create vX.Y.Z --title "vX.Y.Z" --notes "See changelog" --draft
-    gh release upload vX.Y.Z artifacts/release/*
-    ```
-  * Review the draft release, edit the changelog, then publish.
-  * Add the .rpm and .deb files to the official zrepl repos.
-    * Code for management of these repos: https://github.com/zrepl/package-repo-ops (private repo at this time)
-* Update docs version list:
-  * Update `docs/_templates/versions.html` with the new release.
-  * Verify the link to `zrepl-noarch.tar` in the GitHub release works.
-  * Merge to `master` (docs auto-publish).
-
-#### Patch releases, Go toolchain updates, APT/RPM Package rebuilds
-
-`vMAJOR.MINOR.0` is typically a tagged commit on `master`, because development velocity isn't high
-and thus release branches for stabilization aren't necessary.
-
-Occasionally though there is a need for patch changes to a release, e.g.
-- security issue in a dependency
-- Go toolchain update (e.g. security issue in standard library)
-
-The procedure for this is the following
-- create a branch off the release tag we need to patch, named `releases/MAJOR.MINOR.X`
-- that branch will never be merged into `master`, it'll be a dead-end for this specific patch
-- make changes in that branch
-- make the final commit that bumps version numbers
-- create the git tag
-- follow general procedure for publishing the release (previous sectino
-
-For Go toolchain updates and package rebuilds with no source code changes, leverage the APT/RPM package revision field.
-Control via the `ZREPL_PACKAGE_RELEASE` Makefile variable, see `origin/releases/0.6.1-2` for an example.
-
-
-### Updating Dependencies
-
-- Update the `go` directive and `toolchain` directive in `go.mod`
-  - `go` is the minimum supported version
-  - `toolchain` is the preferred toolchain version if `GOTOOLCHAIN` is not specified
-
-Run `go mod tidy` to ensure consistency.
-
-Update Go module dependencies:
-
-```bash
-# Update all other dependencies
-go get -u -t ./...
-go mod tidy
-
-# Above might fail if there are version selection conflicts.
-# Figure out what's going on by updating packages from error messages first.
-# Example:
-go get -u google.golang.org/genproto google.golang.org/grpc google.golang.org/protobuf
-```
-
-Update codegen & lint tools
-- `protoc` =>  `build/get_protoc.bash`
-  - GH releases publish sha256 sums
-- `golangci-lint` => `build/get_golangci_lint.bash`
-- bump versions in `build/tools.go`
-  - we use the tools.go trick:
-  - `go get -tags tools -u example.com/tool ; go mod tidy`
-  - review whether we're ready to switch to `go tool`: https://github.com/zrepl/zrepl/pull/909
-
-Now run `make generate`.
-
-Run `make lint` and `make vet`.
-
-Update the CI configuration `.circleci/config.yml`:
-- Update Go version references (we reference the minimum and max supported version)
-- Set `Makefile` `RELEASE_GOVERSION` to the new Go version
-- Update the pinned ZFS release tags in the `platformtest` matrix (`zfs_release` parameter) to the latest patch releases of each OpenZFS branch (currently 2.2, 2.3, 2.4)
-- Update the pinned Ubuntu machine image for `platformtest` (e.g. `ubuntu-2404:2025.09.1`) — this is pinned rather than `current` so the ZFS build cache stays valid across runs
-
-Update docs build tooling:
-- Update `uv` version in `.circleci/config.yml` (search for `astral.sh/uv/` and cache keys containing the version)
-- Check if there's now a CircleCI orb for uv that we could use
-- Update Python version in `docs/.python-version`
-
-Update docs dependencies (Sphinx, sphinx-rtd-theme):
-- Check current versions in `docs/pyproject.toml`
-- Review upstream changelogs for breaking changes
-- Update version constraints in `pyproject.toml` and the `uv` lockfile (see [uv docs on dependencies](https://docs.astral.sh/uv/concepts/projects/dependencies/)):
-- Test locally with `make docs`
-
-Kick a full CI pipeline run (`do_ci=true` and `do_release=true`).
-
-Merge PR with merge commit.
-
-## Notes to Distro Package Maintainers
-
-* The `Makefile` in this project is not suitable for builds in distros.
-* Run the platform tests (Docs -> Usage -> Platform Tests) **on a test system** to validate that zrepl's abstractions on top of ZFS work with the system ZFS.
 * Ship a default config that adheres to your distro's `hier` and logging system.
 * Ship a service manager file and _please_ try to upstream it to this repository.
-  * `dist/systemd` contains a Systemd unit template.
+* Use `make release ZREPL_VERSION='mydistro-1.2.3_1'`
-* Ship other material provided in `./dist`, e.g. in `/usr/share/zrepl/`.
+    * Your distro's name and any versioning supplemental to zrepl's (e.g. package revision) should be in this string
-* Have a look at the `Makefile`'s `ZREPL_VERSION` variable and how it passed to Go's `ldFlags`.
-  This is how `zrepl version` knows what version number to show.
-  Your build system should set the `ldFlags` flags appropriately and add a prefix or suffix that indicates that the given zrepl binary is a distro build, not an official one.
 * Make sure you are informed about new zrepl versions, e.g. by subscribing to GitHub's release RSS feed.
+
+## Developer Documentation
+
+First, use `./lazy.sh devsetup` to install build dependencies and read `docs/installation.rst -> Compiling from Source`.
+
+### Overall Architecture
+
+The application architecture is documented as part of the user docs in the *Implementation* section (`docs/content/impl`).
+Make sure to develop an understanding how zrepl is typically used by studying the user docs first.
+
+### Project Structure
+
+```
+├── cmd
+│   ├── sampleconf          # example configuration
+├── docs                    # sphinx-based documentation
+│   ├── **/*.rst            # documentation in reStructuredText
+│   ├── sphinxconf
+│   │   └── conf.py         # sphinx config (see commit 445a280 why its not in docs/)
+│   ├── requirements.txt    # pip3 requirements to build documentation
+│   ├── publish.sh          # shell script for automated rendering & deploy to zrepl.github.io repo
+│   ├── public_git          # checkout of zrepl.github.io managed by above shell script
+├── logger                  # logger package used by zrepl
+├── rpc                     # rpc protocol implementation
+├── util
+└── zfs                     # ZFS wrappers, filesystemm diffing
+```
+
+### Coding Workflow
+
+* Open an issue when starting to hack on a new feature
+* Commits should reference the issue they are related to
+* Docs improvements not documenting new features do not require an issue.
+
+### Breaking Changes
+
+Backward-incompatible changes must be documented in the git commit message and are listed in `docs/changelog.rst`.
+
+* Config-breaking changes must contain a line `BREAK CONFIG` in the commit message
+* Other breaking changes must contain a line `BREAK` in the commit message
+
+### Glossary & Naming Inconsistencies
+
+In ZFS, *dataset* refers to the objects *filesystem*, *ZVOL* and *snapshot*. <br />
+However, we need a word for *filesystem* & *ZVOL* but not a snapshot, bookmark, etc.
+
+Toward the user, the following terminology is used:
+
+* **filesystem**: a ZFS filesystem or a ZVOL
+* **filesystem version**: a ZFS snapshot or a bookmark
+
+Sadly, the zrepl implementation is inconsistent in its use of these words:
+variables and types are often named *dataset* when they in fact refer to a *filesystem*.
+
+There will not be a big refactoring (an attempt was made, but it's destroying too much history without much gain).
+
+However, new contributions & patches should fix naming without further notice in the commit message.
+

build.Dockerfile

@@ -0,0 +1,16 @@
+FROM golang:latest
+
+RUN apt-get update && apt-get install -y \
+    python3-pip
+
+ADD lazy.sh /tmp/lazy.sh
+ADD docs/requirements.txt /tmp/requirements.txt
+ENV ZREPL_LAZY_DOCS_REQPATH=/tmp/requirements.txt
+RUN /tmp/lazy.sh devsetup
+
+# prepare volume mount of git checkout to /zrepl
+RUN mkdir -p /go/src/github.com/zrepl/zrepl
+RUN chmod -R 0777 /go
+WORKDIR /go/src/github.com/zrepl/zrepl
+
+

build/.gitignore

@@ -1 +0,0 @@
-install/

build/build.Dockerfile

@@ -1,30 +0,0 @@
-FROM !SUBSTITUTED_BY_MAKEFILE
-
-ARG BUILD_UID=1000
-ARG BUILD_GID=1000
-
-RUN apt-get update && apt-get install -y \
-    python3 \
-    unzip \
-    gawk \
-    curl
-
-# Create build user with the host's UID/GID before installing uv
-RUN groupadd -g ${BUILD_GID} zrepl_build && \
-    useradd -u ${BUILD_UID} -g ${BUILD_GID} -m zrepl_build
-
-# Go toolchain uses xdg-cache
-RUN mkdir -p /.cache && chmod -R 0777 /.cache
-
-# Install uv as the build user - version from .uv-version file
-ADD .uv-version /tmp/.uv-version
-USER zrepl_build
-RUN UV_VERSION=$(cat /tmp/.uv-version) && \
-    curl -LsSf https://astral.sh/uv/${UV_VERSION}/install.sh | sh
-
-# Go devtools are managed by Makefile
-
-WORKDIR /src
-ENV PATH="/home/zrepl_build/.local/bin:$PATH" \
-    GOCACHE="/.cache/go-build"
-

build/get_golangci_lint.bash

@@ -1,33 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-set -x
-
-# golangci-lint recommends against using `go get` and friends to install it
-
-cd "$1"
-
-VERSION=2.8.0
-FILENAME=golangci-lint-"$VERSION"-linux-"$GOHOSTARCH".tar.gz
-
-if [ -e "$FILENAME" ]; then
-    echo "$FILENAME" already exists 1>&2
-    exit 1
-fi
-
-wget --continue https://github.com/golangci/golangci-lint/releases/download/v"$VERSION"/"$FILENAME"
-
-stat "$FILENAME"
-
-# Select the correct checksum for the downloaded architecture
-case "$GOHOSTARCH" in
-    arm64) EXPECTED_SHA256="2a58388db8af5ab9330791cea0ebdd4100723cd05ad7185d92febaaee272ec9a" ;;
-    amd64) EXPECTED_SHA256="7048bc6b25c9515ed092c83f9fa8709ca97937ead52d9ff317a143299ee97a50" ;;
-    *)     echo "Unknown architecture: $GOHOSTARCH" >&2; exit 1 ;;
-esac
-
-# Verify checksum explicitly - fails if hash doesn't match
-echo "$EXPECTED_SHA256  $FILENAME" | sha256sum -c -
-
-tar -x  --strip-components=1 -f "$FILENAME"
-
-stat ./golangci-lint

build/get_protoc.bash

@@ -1,32 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-set -x
-
-cd "$1"
-
-MACH=$(uname -m)
-MACH="${MACH/aarch64/aarch_64}"
-
-VERSION=33.5
-FILENAME=protoc-"$VERSION"-linux-"$MACH".zip
-
-if [ -e "$FILENAME" ]; then
-    echo "$FILENAME" already exists 1>&2
-    exit 1
-fi
-
-wget --continue https://github.com/protocolbuffers/protobuf/releases/download/v"$VERSION"/"$FILENAME"
-
-stat "$FILENAME"
-
-# Select the correct checksum for the downloaded architecture
-case "$MACH" in
-    aarch_64) EXPECTED_SHA256="2b0fcf9b2c32cbadccc0eb7a88b841fffecd4a06fc80acdba2b5be45e815c38a" ;;
-    x86_64)   EXPECTED_SHA256="24e58fb231d50306ee28491f33a170301e99540f7e29ca461e0e80fd1239f8d1" ;;
-    *)        echo "Unknown architecture: $MACH" >&2; exit 1 ;;
-esac
-
-# Verify checksum explicitly - fails if hash doesn't match
-echo "$EXPECTED_SHA256  $FILENAME" | sha256sum -c -
-
-unzip -d . "$FILENAME"

build/go.mod

@@ -1,24 +0,0 @@
-module github.com/zrepl/zrepl/build
-
-go 1.24.13
-
-toolchain go1.25.7
-
-require (
-	github.com/alvaroloes/enumer v1.1.2
-	github.com/wadey/gocovmerge v0.0.0-20160331181800-b5bfa59ec0ad
-	golang.org/x/tools v0.41.0
-	google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.6.1
-	google.golang.org/protobuf v1.36.11
-)
-
-require (
-	github.com/pascaldekloe/name v1.0.1 // indirect
-	golang.org/x/mod v0.32.0 // indirect
-	golang.org/x/sync v0.19.0 // indirect
-	golang.org/x/sys v0.41.0 // indirect
-	golang.org/x/telemetry v0.0.0-20260205145544-86a5c4bf3c8d // indirect
-	golang.org/x/text v0.31.0 // indirect
-	google.golang.org/genproto/googleapis/rpc v0.0.0-20250818200422-3122310a409c // indirect
-	google.golang.org/grpc v1.75.0 // indirect
-)

build/go.sum

@@ -1,37 +0,0 @@
-github.com/alvaroloes/enumer v1.1.2 h1:5khqHB33TZy1GWCO/lZwcroBFh7u+0j40T83VUbfAMY=
-github.com/alvaroloes/enumer v1.1.2/go.mod h1:FxrjvuXoDAx9isTJrv4c+T410zFi0DtXIT0m65DJ+Wo=
-github.com/google/go-cmp v0.7.0 h1:wk8382ETsv4JYUZwIsn6YpYiWiBsYLSJiTsyBybVuN8=
-github.com/google/go-cmp v0.7.0/go.mod h1:pXiqmnSA92OHEEa9HXL2W4E7lf9JzCmGVUdgjX3N/iU=
-github.com/pascaldekloe/name v0.0.0-20180628100202-0fd16699aae1/go.mod h1:eD5JxqMiuNYyFNmyY9rkJ/slN8y59oEu4Ei7F8OoKWQ=
-github.com/pascaldekloe/name v1.0.1 h1:9lnXOHeqeHHnWLbKfH6X98+4+ETVqFqxN09UXSjcMb0=
-github.com/pascaldekloe/name v1.0.1/go.mod h1:Z//MfYJnH4jVpQ9wkclwu2I2MkHmXTlT9wR5UZScttM=
-github.com/wadey/gocovmerge v0.0.0-20160331181800-b5bfa59ec0ad h1:W0LEBv82YCGEtcmPA3uNZBI33/qF//HAAs3MawDjRa0=
-github.com/wadey/gocovmerge v0.0.0-20160331181800-b5bfa59ec0ad/go.mod h1:Hy8o65+MXnS6EwGElrSRjUzQDLXreJlzYLlWiHtt8hM=
-golang.org/x/crypto v0.0.0-20190308221718-c2843e01d9a2/go.mod h1:djNgcEr1/C05ACkg1iLfiJU5Ep61QUkGW8qpdssI0+w=
-golang.org/x/mod v0.32.0 h1:9F4d3PHLljb6x//jOyokMv3eX+YDeepZSEo3mFJy93c=
-golang.org/x/mod v0.32.0/go.mod h1:SgipZ/3h2Ci89DlEtEXWUk/HteuRin+HHhN+WbNhguU=
-golang.org/x/net v0.0.0-20190311183353-d8887717615a/go.mod h1:t9HGtf8HONx5eT2rtn7q6eTqICYqUVnKs3thJo3Qplg=
-golang.org/x/net v0.49.0 h1:eeHFmOGUTtaaPSGNmjBKpbng9MulQsJURQUAfUwY++o=
-golang.org/x/net v0.49.0/go.mod h1:/ysNB2EvaqvesRkuLAyjI1ycPZlQHM3q01F02UY/MV8=
-golang.org/x/sync v0.0.0-20190423024810-112230192c58/go.mod h1:RxMgew5VJxzue5/jJTE5uejpjVlOe/izrB70Jof72aM=
-golang.org/x/sync v0.19.0 h1:vV+1eWNmZ5geRlYjzm2adRgW2/mcpevXNg50YZtPCE4=
-golang.org/x/sync v0.19.0/go.mod h1:9KTHXmSnoGruLpwFjVSX0lNNA75CykiMECbovNTZqGI=
-golang.org/x/sys v0.0.0-20190215142949-d0b11bdaac8a/go.mod h1:STP8DvDyc/dI5b8T5hshtkjS+E42TnysNCUPdjciGhY=
-golang.org/x/sys v0.41.0 h1:Ivj+2Cp/ylzLiEU89QhWblYnOE9zerudt9Ftecq2C6k=
-golang.org/x/sys v0.41.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
-golang.org/x/telemetry v0.0.0-20260205145544-86a5c4bf3c8d h1:xjbyaj6khjO0ocYYa7k9sQLbdi4R1KwHnsS1QGijbsM=
-golang.org/x/telemetry v0.0.0-20260205145544-86a5c4bf3c8d/go.mod h1:b7fPSJ0pKZ3ccUh8gnTONJxhn3c/PS6tyzQvyqw4iA8=
-golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=
-golang.org/x/text v0.31.0 h1:aC8ghyu4JhP8VojJ2lEHBnochRno1sgL6nEi9WGFGMM=
-golang.org/x/text v0.31.0/go.mod h1:tKRAlv61yKIjGGHX/4tP1LTbc13YSec1pxVEWXzfoeM=
-golang.org/x/tools v0.0.0-20190524210228-3d17549cdc6b/go.mod h1:RgjU9mgBXZiqYHBnxXauZ1Gv1EHHAz9KjViQ78xBX0Q=
-golang.org/x/tools v0.41.0 h1:a9b8iMweWG+S0OBnlU36rzLp20z1Rp10w+IY2czHTQc=
-golang.org/x/tools v0.41.0/go.mod h1:XSY6eDqxVNiYgezAVqqCeihT4j1U2CCsqvH3WhQpnlg=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20250818200422-3122310a409c h1:qXWI/sQtv5UKboZ/zUk7h+mrf/lXORyI+n9DKDAusdg=
-google.golang.org/genproto/googleapis/rpc v0.0.0-20250818200422-3122310a409c/go.mod h1:gw1tLEfykwDz2ET4a12jcXt4couGAm7IwsVaTy0Sflo=
-google.golang.org/grpc v1.75.0 h1:+TW+dqTd2Biwe6KKfhE5JpiYIBWq865PhKGSXiivqt4=
-google.golang.org/grpc v1.75.0/go.mod h1:JtPAzKiq4v1xcAB2hydNlWI2RnF85XXcV0mhKXr2ecQ=
-google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.6.1 h1:/WILD1UcXj/ujCxgoL/DvRgt2CP3txG8+FwkUbb9110=
-google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.6.1/go.mod h1:YNKnb2OAApgYn2oYY47Rn7alMr1zWjb2U8Q0aoGWiNc=
-google.golang.org/protobuf v1.36.11 h1:fV6ZwhNocDyBLK0dj+fg8ektcVegBBuEolpbTQyBNVE=
-google.golang.org/protobuf v1.36.11/go.mod h1:HTf+CrKn2C3g5S8VImy6tdcUvCska2kB7j23XfzDpco=

build/go_install_host_tool.source

@@ -1,6 +0,0 @@
-export GO111MODULE=on # otherwise, a checkout of this repo in GOPATH will disable modules on Go 1.12 and earlier
-source <(go env)
-# build tools for the host platform
-export GOOS="$GOHOSTOS"
-export GOARCH="$GOHOSTARCH"
-# TODO GOARM=$GOHOSTARM?

build/go_install_tools.bash

@@ -1,19 +0,0 @@
-#!/usr/bin/env bash
-
-set -euo pipefail
-set -x
-
-OUTDIR="$(readlink -f "$1")"
-if [ -e "$OUTDIR" ]; then
-    echo "$OUTDIR" already exists 1>&2
-    exit 1
-fi
-
-# go install command below will install tools to $GOBIN
-export GOBIN="$OUTDIR"
-
-cd "$(dirname "$0")"
-
-source ./go_install_host_tool.source
-
-cat tools.go | grep _ | awk -F'"' '{print $2}' | tee | xargs -tI '{}' go install '{}'

build/tools.go

@@ -1,14 +0,0 @@
-//go:build tools
-// +build tools
-
-package main
-
-// the lines are parsed by lazy.sh, do not edit
-import (
-	_ "github.com/alvaroloes/enumer"
-	_ "github.com/wadey/gocovmerge"
-	_ "golang.org/x/tools/cmd/goimports"
-	_ "golang.org/x/tools/cmd/stringer"
-	_ "google.golang.org/grpc/cmd/protoc-gen-go-grpc"
-	_ "google.golang.org/protobuf/cmd/protoc-gen-go"
-)

cmd/autosnap.go

@@ -0,0 +1,215 @@
+package cmd
+
+import (
+	"context"
+	"fmt"
+	"github.com/zrepl/zrepl/zfs"
+	"sort"
+	"sync"
+	"time"
+)
+
+type IntervalAutosnap struct {
+	task             *Task
+	DatasetFilter    zfs.DatasetFilter
+	Prefix           string
+	SnapshotInterval time.Duration
+}
+
+func (a *IntervalAutosnap) filterFilesystems() (fss []*zfs.DatasetPath, stop bool) {
+	a.task.Enter("filter_filesystems")
+	defer a.task.Finish()
+	fss, err := zfs.ZFSListMapping(a.DatasetFilter)
+	stop = err != nil
+	if err != nil {
+		a.task.Log().WithError(err).Error("cannot list datasets")
+	}
+	if len(fss) == 0 {
+		a.task.Log().Warn("no filesystem matching filesystem filter")
+	}
+	return fss, stop
+}
+
+func (a *IntervalAutosnap) findSyncPoint(fss []*zfs.DatasetPath) (syncPoint time.Time, err error) {
+	a.task.Enter("find_sync_point")
+	defer a.task.Finish()
+	type snapTime struct {
+		ds   *zfs.DatasetPath
+		time time.Time
+	}
+
+	if len(fss) == 0 {
+		return time.Now(), nil
+	}
+
+	snaptimes := make([]snapTime, 0, len(fss))
+
+	now := time.Now()
+
+	a.task.Log().Debug("examine filesystem state")
+	for _, d := range fss {
+
+		l := a.task.Log().WithField(logFSField, d.ToString())
+
+		fsvs, err := zfs.ZFSListFilesystemVersions(d, NewPrefixFilter(a.Prefix))
+		if err != nil {
+			l.WithError(err).Error("cannot list filesystem versions")
+			continue
+		}
+		if len(fsvs) <= 0 {
+			l.WithField("prefix", a.Prefix).Info("no filesystem versions with prefix")
+			continue
+		}
+
+		// Sort versions by creation
+		sort.SliceStable(fsvs, func(i, j int) bool {
+			return fsvs[i].CreateTXG < fsvs[j].CreateTXG
+		})
+
+		latest := fsvs[len(fsvs)-1]
+		l.WithField("creation", latest.Creation).
+			Debug("found latest snapshot")
+
+		since := now.Sub(latest.Creation)
+		if since < 0 {
+			l.WithField("snapshot", latest.Name).
+				WithField("creation", latest.Creation).
+				Error("snapshot is from the future")
+			continue
+		}
+		next := now
+		if since < a.SnapshotInterval {
+			next = latest.Creation.Add(a.SnapshotInterval)
+		}
+		snaptimes = append(snaptimes, snapTime{d, next})
+	}
+
+	if len(snaptimes) == 0 {
+		snaptimes = append(snaptimes, snapTime{nil, now})
+	}
+
+	sort.Slice(snaptimes, func(i, j int) bool {
+		return snaptimes[i].time.Before(snaptimes[j].time)
+	})
+
+	return snaptimes[0].time, nil
+
+}
+
+func (a *IntervalAutosnap) waitForSyncPoint(ctx context.Context, syncPoint time.Time) {
+	a.task.Enter("wait_sync_point")
+	defer a.task.Finish()
+
+	const LOG_TIME_FMT string = time.ANSIC
+
+	a.task.Log().WithField("sync_point", syncPoint.Format(LOG_TIME_FMT)).
+		Info("wait for sync point")
+
+	select {
+	case <-ctx.Done():
+		a.task.Log().WithError(ctx.Err()).Info("context done")
+		return
+	case <-time.After(syncPoint.Sub(time.Now())):
+	}
+}
+
+func (a *IntervalAutosnap) syncUpRun(ctx context.Context, didSnaps chan struct{}) (stop bool) {
+	a.task.Enter("sync_up")
+	defer a.task.Finish()
+
+	fss, stop := a.filterFilesystems()
+	if stop {
+		return true
+	}
+
+	syncPoint, err := a.findSyncPoint(fss)
+	if err != nil {
+		return true
+	}
+
+	a.waitForSyncPoint(ctx, syncPoint)
+
+	a.task.Log().Debug("snapshot all filesystems to enable further snaps in lockstep")
+	a.doSnapshots(didSnaps)
+	return false
+}
+
+func (a *IntervalAutosnap) Run(ctx context.Context, didSnaps chan struct{}) {
+
+	if a.syncUpRun(ctx, didSnaps) {
+		a.task.Log().Error("stoppping autosnap after error in sync up")
+		return
+	}
+
+	// task drops back to idle here
+
+	a.task.Log().Debug("setting up ticker in SnapshotInterval")
+	ticker := time.NewTicker(a.SnapshotInterval)
+	for {
+		select {
+		case <-ctx.Done():
+			ticker.Stop()
+			a.task.Log().WithError(ctx.Err()).Info("context done")
+			return
+
+		case <-ticker.C:
+			a.doSnapshots(didSnaps)
+		}
+	}
+
+}
+
+func (a *IntervalAutosnap) doSnapshots(didSnaps chan struct{}) {
+
+	a.task.Enter("do_snapshots")
+	defer a.task.Finish()
+
+	// don't cache the result from previous run in case the user added
+	// a new dataset in the meantime
+	fss, stop := a.filterFilesystems()
+	if stop {
+		return
+	}
+
+	a.task.Log().Info("beginning parallel snapshots")
+
+	// TODO channel programs -> allow a little jitter?
+	var wg sync.WaitGroup
+	for fsi := range fss {
+		wg.Add(1)
+		go func(fs *zfs.DatasetPath) {
+			defer wg.Done()
+
+			suffix := time.Now().In(time.UTC).Format("20060102_150405_000")
+			snapname := fmt.Sprintf("%s%s", a.Prefix, suffix)
+
+			l := a.task.Log().WithField(logFSField, fs.ToString()).
+				WithField("snapname", snapname)
+
+			l.Info("create snapshot")
+			err := zfs.ZFSSnapshot(fs, snapname, false)
+			if err != nil {
+				l.WithError(err).Error("cannot create snapshot")
+				return
+			}
+
+			l.Info("create corresponding bookmark")
+			err = zfs.ZFSBookmark(fs, snapname, snapname)
+			if err != nil {
+				l.WithError(err).Error("cannot create bookmark")
+			}
+
+		}(fss[fsi])
+	}
+
+	a.task.Log().Info("waiting for parallel snapshots to finish")
+	wg.Wait()
+	a.task.Log().Info("snapshots finished")
+
+	select {
+	case didSnaps <- struct{}{}:
+	default:
+		a.task.Log().Error("warning: callback channel is full, discarding")
+	}
+
+}

cmd/bashcomp.go

@@ -0,0 +1,29 @@
+package cmd
+
+import (
+	"fmt"
+
+	"github.com/spf13/cobra"
+	"os"
+)
+
+var bashcompCmd = &cobra.Command{
+	Use:   "bashcomp path/to/out/file",
+	Short: "generate bash completions",
+	Run: func(cmd *cobra.Command, args []string) {
+		if len(args) != 1 {
+			fmt.Fprintf(os.Stderr, "specify exactly one positional agument\n")
+			cmd.Usage()
+			os.Exit(1)
+		}
+		if err := RootCmd.GenBashCompletionFile(args[0]); err != nil {
+			fmt.Fprintf(os.Stderr, "error generating bash completion: %s", err)
+			os.Exit(1)
+		}
+	},
+	Hidden: true,
+}
+
+func init() {
+	RootCmd.AddCommand(bashcompCmd)
+}

cmd/config.go

@@ -0,0 +1,102 @@
+package cmd
+
+import (
+	"io"
+
+	"fmt"
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+type Config struct {
+	Global Global
+	Jobs   map[string]Job
+}
+
+func (c *Config) LookupJob(name string) (j Job, err error) {
+	j, ok := c.Jobs[name]
+	if !ok {
+		return nil, errors.Errorf("job '%s' is not defined", name)
+	}
+	return j, nil
+}
+
+type Global struct {
+	Serve struct {
+		Stdinserver struct {
+			SockDir string
+		}
+	}
+	Control struct {
+		Sockpath string
+	}
+	logging *LoggingConfig
+}
+
+type JobDebugSettings struct {
+	Conn struct {
+		ReadDump  string `mapstructure:"read_dump"`
+		WriteDump string `mapstructure:"write_dump"`
+	}
+	RPC struct {
+		Log bool
+	}
+}
+
+type RWCConnecter interface {
+	Connect() (io.ReadWriteCloser, error)
+}
+type AuthenticatedChannelListenerFactory interface {
+	Listen() (AuthenticatedChannelListener, error)
+}
+
+type AuthenticatedChannelListener interface {
+	Accept() (ch io.ReadWriteCloser, err error)
+	Close() (err error)
+}
+
+type SSHStdinServerConnectDescr struct {
+}
+
+type PrunePolicy interface {
+	// Prune filters versions and decide which to keep and which to remove.
+	// Prune **does not** implement the actual removal of the versions.
+	Prune(fs *zfs.DatasetPath, versions []zfs.FilesystemVersion) (keep, remove []zfs.FilesystemVersion, err error)
+}
+
+type PruningJob interface {
+	Pruner(task *Task, side PrunePolicySide, dryRun bool) (Pruner, error)
+}
+
+// A type for constants describing different prune policies of a PruningJob
+// This is mostly a special-case for LocalJob, which is the only job that has two prune policies
+// instead of one.
+// It implements github.com/spf13/pflag.Value to be used as CLI flag for the test subcommand
+type PrunePolicySide string
+
+const (
+	PrunePolicySideDefault PrunePolicySide = ""
+	PrunePolicySideLeft    PrunePolicySide = "left"
+	PrunePolicySideRight   PrunePolicySide = "right"
+)
+
+func (s *PrunePolicySide) String() string {
+	return string(*s)
+}
+
+func (s *PrunePolicySide) Set(news string) error {
+	p := PrunePolicySide(news)
+	switch p {
+	case PrunePolicySideRight:
+		fallthrough
+	case PrunePolicySideLeft:
+		*s = p
+	default:
+		return errors.Errorf("must be either %s or %s", PrunePolicySideLeft, PrunePolicySideRight)
+	}
+	return nil
+}
+
+func (s *PrunePolicySide) Type() string {
+	return fmt.Sprintf("%s | %s", PrunePolicySideLeft, PrunePolicySideRight)
+}

cmd/config_connect.go

@@ -0,0 +1,48 @@
+package cmd
+
+import (
+	"fmt"
+	"io"
+
+	"context"
+	"github.com/jinzhu/copier"
+	"github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"github.com/problame/go-netssh"
+)
+
+type SSHStdinserverConnecter struct {
+	Host                 string
+	User                 string
+	Port                 uint16
+	IdentityFile         string   `mapstructure:"identity_file"`
+	TransportOpenCommand []string `mapstructure:"transport_open_command"`
+	SSHCommand           string   `mapstructure:"ssh_command"`
+	Options              []string
+}
+
+func parseSSHStdinserverConnecter(i map[string]interface{}) (c *SSHStdinserverConnecter, err error) {
+
+	c = &SSHStdinserverConnecter{}
+	if err = mapstructure.Decode(i, c); err != nil {
+		err = errors.New(fmt.Sprintf("could not parse ssh transport: %s", err))
+		return nil, err
+	}
+
+	// TODO assert fields are filled
+	return
+
+}
+
+func (c *SSHStdinserverConnecter) Connect() (rwc io.ReadWriteCloser, err error) {
+
+	var endpoint netssh.Endpoint
+	if err = copier.Copy(&endpoint, c); err != nil {
+		return
+	}
+	if rwc, err = netssh.Dial(context.TODO(), endpoint); err != nil {
+		err = errors.WithStack(err)
+		return
+	}
+	return
+}

cmd/config_fsvfilter.go

@@ -0,0 +1,36 @@
+package cmd
+
+import (
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/zfs"
+	"strings"
+)
+
+type PrefixFilter struct {
+	prefix    string
+	fstype    zfs.VersionType
+	fstypeSet bool // optionals anyone?
+}
+
+func NewPrefixFilter(prefix string) *PrefixFilter {
+	return &PrefixFilter{prefix: prefix}
+}
+
+func NewTypedPrefixFilter(prefix string, versionType zfs.VersionType) *PrefixFilter {
+	return &PrefixFilter{prefix, versionType, true}
+}
+
+func parseSnapshotPrefix(i string) (p string, err error) {
+	if len(i) <= 0 {
+		err = errors.Errorf("snapshot prefix must not be empty string")
+		return
+	}
+	p = i
+	return
+}
+
+func (f *PrefixFilter) Filter(fsv zfs.FilesystemVersion) (accept bool, err error) {
+	fstypeMatches := (!f.fstypeSet || fsv.Type == f.fstype)
+	prefixMatches := strings.HasPrefix(fsv.Name, f.prefix)
+	return fstypeMatches && prefixMatches, nil
+}

cmd/config_job_control.go

@@ -0,0 +1,143 @@
+package cmd
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"github.com/pkg/errors"
+	"io"
+	"net"
+	"net/http"
+)
+
+type ControlJob struct {
+	Name     string
+	sockaddr *net.UnixAddr
+}
+
+func NewControlJob(name, sockpath string) (j *ControlJob, err error) {
+	j = &ControlJob{Name: name}
+
+	j.sockaddr, err = net.ResolveUnixAddr("unix", sockpath)
+	if err != nil {
+		err = errors.Wrap(err, "cannot resolve unix address")
+		return
+	}
+
+	return
+}
+
+func (j *ControlJob) JobName() string {
+	return j.Name
+}
+
+func (j *ControlJob) JobStatus(ctx context.Context) (*JobStatus, error) {
+	return &JobStatus{Tasks: nil}, nil
+}
+
+const (
+	ControlJobEndpointPProf   string = "/debug/pprof"
+	ControlJobEndpointVersion string = "/version"
+	ControlJobEndpointStatus  string = "/status"
+)
+
+func (j *ControlJob) JobStart(ctx context.Context) {
+
+	log := ctx.Value(contextKeyLog).(Logger)
+	defer log.Info("control job finished")
+
+	daemon := ctx.Value(contextKeyDaemon).(*Daemon)
+
+	l, err := ListenUnixPrivate(j.sockaddr)
+	if err != nil {
+		log.WithError(err).Error("error listening")
+		return
+	}
+
+	pprofServer := NewPProfServer(ctx)
+
+	mux := http.NewServeMux()
+	mux.Handle(ControlJobEndpointPProf, requestLogger{log: log, handlerFunc: func(w http.ResponseWriter, r *http.Request) {
+		var msg PprofServerControlMsg
+		err := json.NewDecoder(r.Body).Decode(&msg)
+		if err != nil {
+			log.WithError(err).Error("bad pprof request from client")
+			w.WriteHeader(http.StatusBadRequest)
+		}
+		pprofServer.Control(msg)
+		w.WriteHeader(200)
+	}})
+	mux.Handle(ControlJobEndpointVersion,
+		requestLogger{log: log, handler: jsonResponder{func() (interface{}, error) {
+			return NewZreplVersionInformation(), nil
+		}}})
+	mux.Handle(ControlJobEndpointStatus,
+		requestLogger{log: log, handler: jsonResponder{func() (interface{}, error) {
+			return daemon.Status(), nil
+		}}})
+	server := http.Server{Handler: mux}
+
+outer:
+	for {
+
+		served := make(chan error)
+		go func() {
+			served <- server.Serve(l)
+			close(served)
+		}()
+
+		select {
+		case <-ctx.Done():
+			log.WithError(ctx.Err()).Info("context done")
+			server.Shutdown(context.Background())
+			break outer
+		case err = <-served:
+			if err != nil {
+				log.WithError(err).Error("error serving")
+				break outer
+			}
+		}
+
+	}
+
+}
+
+type jsonResponder struct {
+	producer func() (interface{}, error)
+}
+
+func (j jsonResponder) ServeHTTP(w http.ResponseWriter, r *http.Request) {
+	res, err := j.producer()
+	if err != nil {
+		w.WriteHeader(http.StatusInternalServerError)
+		io.WriteString(w, err.Error())
+		return
+	}
+	var buf bytes.Buffer
+	err = json.NewEncoder(&buf).Encode(res)
+	if err != nil {
+		w.WriteHeader(http.StatusInternalServerError)
+		io.WriteString(w, err.Error())
+	} else {
+		io.Copy(w, &buf)
+	}
+}
+
+type requestLogger struct {
+	log         Logger
+	handler     http.Handler
+	handlerFunc http.HandlerFunc
+}
+
+func (l requestLogger) ServeHTTP(w http.ResponseWriter, r *http.Request) {
+	log := l.log.WithField("method", r.Method).WithField("url", r.URL)
+	log.Info("start")
+	if l.handlerFunc != nil {
+		l.handlerFunc(w, r)
+	} else if l.handler != nil {
+		l.handler.ServeHTTP(w, r)
+	} else {
+		log.Error("no handler or handlerFunc configured")
+	}
+	log.Info("finish")
+}

cmd/config_job_local.go

@@ -0,0 +1,214 @@
+package cmd
+
+import (
+	"time"
+
+	"context"
+	"github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/rpc"
+	"github.com/zrepl/zrepl/zfs"
+	"sync"
+)
+
+type LocalJob struct {
+	Name              string
+	Mapping           *DatasetMapFilter
+	SnapshotPrefix    string
+	Interval          time.Duration
+	InitialReplPolicy InitialReplPolicy
+	PruneLHS          PrunePolicy
+	PruneRHS          PrunePolicy
+	Debug             JobDebugSettings
+	snapperTask       *Task
+	mainTask          *Task
+	handlerTask       *Task
+	pruneRHSTask      *Task
+	pruneLHSTask      *Task
+}
+
+func parseLocalJob(c JobParsingContext, name string, i map[string]interface{}) (j *LocalJob, err error) {
+
+	var asMap struct {
+		Mapping           map[string]string
+		SnapshotPrefix    string `mapstructure:"snapshot_prefix"`
+		Interval          string
+		InitialReplPolicy string                 `mapstructure:"initial_repl_policy"`
+		PruneLHS          map[string]interface{} `mapstructure:"prune_lhs"`
+		PruneRHS          map[string]interface{} `mapstructure:"prune_rhs"`
+		Debug             map[string]interface{}
+	}
+
+	if err = mapstructure.Decode(i, &asMap); err != nil {
+		err = errors.Wrap(err, "mapstructure error")
+		return nil, err
+	}
+
+	j = &LocalJob{Name: name}
+
+	if j.Mapping, err = parseDatasetMapFilter(asMap.Mapping, false); err != nil {
+		return
+	}
+
+	if j.SnapshotPrefix, err = parseSnapshotPrefix(asMap.SnapshotPrefix); err != nil {
+		return
+	}
+
+	if j.Interval, err = parsePostitiveDuration(asMap.Interval); err != nil {
+		err = errors.Wrap(err, "cannot parse interval")
+		return
+	}
+
+	if j.InitialReplPolicy, err = parseInitialReplPolicy(asMap.InitialReplPolicy, DEFAULT_INITIAL_REPL_POLICY); err != nil {
+		return
+	}
+
+	if j.PruneLHS, err = parsePrunePolicy(asMap.PruneLHS, true); err != nil {
+		err = errors.Wrap(err, "cannot parse 'prune_lhs'")
+		return
+	}
+	if j.PruneRHS, err = parsePrunePolicy(asMap.PruneRHS, false); err != nil {
+		err = errors.Wrap(err, "cannot parse 'prune_rhs'")
+		return
+	}
+
+	if err = mapstructure.Decode(asMap.Debug, &j.Debug); err != nil {
+		err = errors.Wrap(err, "cannot parse 'debug'")
+		return
+	}
+
+	return
+}
+
+func (j *LocalJob) JobName() string {
+	return j.Name
+}
+
+func (j *LocalJob) JobStart(ctx context.Context) {
+
+	rootLog := ctx.Value(contextKeyLog).(Logger)
+
+	j.snapperTask = NewTask("snapshot", rootLog)
+	j.mainTask = NewTask("main", rootLog)
+	j.handlerTask = NewTask("handler", rootLog)
+	j.pruneRHSTask = NewTask("prune_rhs", rootLog)
+	j.pruneLHSTask = NewTask("prune_lhs", rootLog)
+
+	local := rpc.NewLocalRPC()
+	// Allow access to any dataset since we control what mapping
+	// is passed to the pull routine.
+	// All local datasets will be passed to its Map() function,
+	// but only those for which a mapping exists will actually be pulled.
+	// We can pay this small performance penalty for now.
+	handler := NewHandler(j.handlerTask.Log(), localPullACL{}, NewPrefixFilter(j.SnapshotPrefix))
+
+	registerEndpoints(local, handler)
+
+	snapper := IntervalAutosnap{
+		task:             j.snapperTask,
+		DatasetFilter:    j.Mapping.AsFilter(),
+		Prefix:           j.SnapshotPrefix,
+		SnapshotInterval: j.Interval,
+	}
+
+	plhs, err := j.Pruner(j.pruneLHSTask, PrunePolicySideLeft, false)
+	if err != nil {
+		rootLog.WithError(err).Error("error creating lhs pruner")
+		return
+	}
+	prhs, err := j.Pruner(j.pruneRHSTask, PrunePolicySideRight, false)
+	if err != nil {
+		rootLog.WithError(err).Error("error creating rhs pruner")
+		return
+	}
+
+	didSnaps := make(chan struct{})
+	go snapper.Run(ctx, didSnaps)
+
+outer:
+	for {
+
+		select {
+		case <-ctx.Done():
+			j.mainTask.Log().WithError(ctx.Err()).Info("context")
+			break outer
+		case <-didSnaps:
+			j.mainTask.Log().Debug("finished taking snapshots")
+			j.mainTask.Log().Info("starting replication procedure")
+		}
+
+		j.mainTask.Log().Debug("replicating from lhs to rhs")
+		j.mainTask.Enter("replicate")
+		puller := Puller{j.mainTask, local, j.Mapping, j.InitialReplPolicy}
+		puller.Pull()
+		j.mainTask.Finish()
+
+		// use a ctx as soon as Pull gains ctx support
+		select {
+		case <-ctx.Done():
+			break outer
+		default:
+		}
+
+		var wg sync.WaitGroup
+
+		j.mainTask.Log().Info("pruning lhs")
+		wg.Add(1)
+		go func() {
+			plhs.Run(ctx)
+			wg.Done()
+		}()
+
+		j.mainTask.Log().Info("pruning rhs")
+		wg.Add(1)
+		go func() {
+			prhs.Run(ctx)
+			wg.Done()
+		}()
+
+		wg.Wait()
+
+	}
+
+}
+
+func (j *LocalJob) JobStatus(ctxt context.Context) (*JobStatus, error) {
+	return &JobStatus{Tasks: []*TaskStatus{
+		j.snapperTask.Status(),
+		j.pruneLHSTask.Status(),
+		j.pruneRHSTask.Status(),
+		j.mainTask.Status(),
+	}}, nil
+}
+
+func (j *LocalJob) Pruner(task *Task, side PrunePolicySide, dryRun bool) (p Pruner, err error) {
+
+	var dsfilter zfs.DatasetFilter
+	var pp PrunePolicy
+	switch side {
+	case PrunePolicySideLeft:
+		pp = j.PruneLHS
+		dsfilter = j.Mapping.AsFilter()
+	case PrunePolicySideRight:
+		pp = j.PruneRHS
+		dsfilter, err = j.Mapping.InvertedFilter()
+		if err != nil {
+			err = errors.Wrap(err, "cannot invert mapping for prune_rhs")
+			return
+		}
+	default:
+		err = errors.Errorf("must be either left or right side")
+		return
+	}
+
+	p = Pruner{
+		task,
+		time.Now(),
+		dryRun,
+		dsfilter,
+		j.SnapshotPrefix,
+		pp,
+	}
+
+	return
+}

cmd/config_job_pull.go

@@ -0,0 +1,195 @@
+package cmd
+
+import (
+	"time"
+
+	"context"
+	"fmt"
+	"github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/rpc"
+	"github.com/zrepl/zrepl/util"
+)
+
+type PullJob struct {
+	Name     string
+	Connect  RWCConnecter
+	Interval time.Duration
+	Mapping  *DatasetMapFilter
+	// constructed from mapping during parsing
+	pruneFilter       *DatasetMapFilter
+	SnapshotPrefix    string
+	InitialReplPolicy InitialReplPolicy
+	Prune             PrunePolicy
+	Debug             JobDebugSettings
+
+	task *Task
+}
+
+func parsePullJob(c JobParsingContext, name string, i map[string]interface{}) (j *PullJob, err error) {
+
+	var asMap struct {
+		Connect           map[string]interface{}
+		Interval          string
+		Mapping           map[string]string
+		InitialReplPolicy string `mapstructure:"initial_repl_policy"`
+		Prune             map[string]interface{}
+		SnapshotPrefix    string `mapstructure:"snapshot_prefix"`
+		Debug             map[string]interface{}
+	}
+
+	if err = mapstructure.Decode(i, &asMap); err != nil {
+		err = errors.Wrap(err, "mapstructure error")
+		return nil, err
+	}
+
+	j = &PullJob{Name: name}
+
+	j.Connect, err = parseSSHStdinserverConnecter(asMap.Connect)
+	if err != nil {
+		err = errors.Wrap(err, "cannot parse 'connect'")
+		return nil, err
+	}
+
+	if j.Interval, err = parsePostitiveDuration(asMap.Interval); err != nil {
+		err = errors.Wrap(err, "cannot parse 'interval'")
+		return nil, err
+	}
+
+	j.Mapping, err = parseDatasetMapFilter(asMap.Mapping, false)
+	if err != nil {
+		err = errors.Wrap(err, "cannot parse 'mapping'")
+		return nil, err
+	}
+
+	if j.pruneFilter, err = j.Mapping.InvertedFilter(); err != nil {
+		err = errors.Wrap(err, "cannot automatically invert 'mapping' for prune job")
+		return nil, err
+	}
+
+	j.InitialReplPolicy, err = parseInitialReplPolicy(asMap.InitialReplPolicy, DEFAULT_INITIAL_REPL_POLICY)
+	if err != nil {
+		err = errors.Wrap(err, "cannot parse 'initial_repl_policy'")
+		return
+	}
+
+	if j.SnapshotPrefix, err = parseSnapshotPrefix(asMap.SnapshotPrefix); err != nil {
+		return
+	}
+
+	if j.Prune, err = parsePrunePolicy(asMap.Prune, false); err != nil {
+		err = errors.Wrap(err, "cannot parse prune policy")
+		return
+	}
+
+	if err = mapstructure.Decode(asMap.Debug, &j.Debug); err != nil {
+		err = errors.Wrap(err, "cannot parse 'debug'")
+		return
+	}
+
+	return
+}
+
+func (j *PullJob) JobName() string {
+	return j.Name
+}
+
+func (j *PullJob) JobStart(ctx context.Context) {
+
+	log := ctx.Value(contextKeyLog).(Logger)
+	defer log.Info("exiting")
+	j.task = NewTask("main", log)
+
+	// j.task is idle here idle here
+
+	ticker := time.NewTicker(j.Interval)
+	for {
+		j.doRun(ctx)
+		select {
+		case <-ctx.Done():
+			j.task.Log().WithError(ctx.Err()).Info("context")
+			return
+		case <-ticker.C:
+		}
+	}
+}
+
+func (j *PullJob) doRun(ctx context.Context) {
+
+	j.task.Enter("run")
+	defer j.task.Finish()
+
+	j.task.Log().Info("connecting")
+	rwc, err := j.Connect.Connect()
+	if err != nil {
+		j.task.Log().WithError(err).Error("error connecting")
+		return
+	}
+
+	rwc, err = util.NewReadWriteCloserLogger(rwc, j.Debug.Conn.ReadDump, j.Debug.Conn.WriteDump)
+	if err != nil {
+		return
+	}
+
+	client := rpc.NewClient(rwc)
+	if j.Debug.RPC.Log {
+		client.SetLogger(j.task.Log(), true)
+	}
+
+	j.task.Enter("pull")
+	puller := Puller{j.task, client, j.Mapping, j.InitialReplPolicy}
+	puller.Pull()
+	closeRPCWithTimeout(j.task, client, time.Second*1, "")
+	rwc.Close()
+	j.task.Finish()
+
+	j.task.Enter("prune")
+	pruner, err := j.Pruner(j.task, PrunePolicySideDefault, false)
+	if err != nil {
+		j.task.Log().WithError(err).Error("error creating pruner")
+	} else {
+		pruner.Run(ctx)
+	}
+	j.task.Finish()
+
+}
+
+func (j *PullJob) JobStatus(ctxt context.Context) (*JobStatus, error) {
+	return &JobStatus{Tasks: []*TaskStatus{j.task.Status()}}, nil
+}
+
+func (j *PullJob) Pruner(task *Task, side PrunePolicySide, dryRun bool) (p Pruner, err error) {
+	p = Pruner{
+		task,
+		time.Now(),
+		dryRun,
+		j.pruneFilter,
+		j.SnapshotPrefix,
+		j.Prune,
+	}
+	return
+}
+
+func closeRPCWithTimeout(task *Task, remote rpc.RPCClient, timeout time.Duration, goodbye string) {
+
+	task.Log().Info("closing rpc connection")
+
+	ch := make(chan error)
+	go func() {
+		ch <- remote.Close()
+		close(ch)
+	}()
+
+	var err error
+	select {
+	case <-time.After(timeout):
+		err = fmt.Errorf("timeout exceeded (%s)", timeout)
+	case closeRequestErr := <-ch:
+		err = closeRequestErr
+	}
+
+	if err != nil {
+		task.Log().WithError(err).Error("error closing connection")
+	}
+	return
+}

cmd/config_job_source.go

@@ -0,0 +1,250 @@
+package cmd
+
+import (
+	"context"
+	"io"
+	"time"
+
+	mapstructure "github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/rpc"
+	"github.com/zrepl/zrepl/util"
+)
+
+type SourceJob struct {
+	Name           string
+	Serve          AuthenticatedChannelListenerFactory
+	Filesystems    *DatasetMapFilter
+	SnapshotPrefix string
+	Interval       time.Duration
+	Prune          PrunePolicy
+	Debug          JobDebugSettings
+	serveTask      *Task
+	autosnapTask   *Task
+	pruneTask      *Task
+}
+
+func parseSourceJob(c JobParsingContext, name string, i map[string]interface{}) (j *SourceJob, err error) {
+
+	var asMap struct {
+		Serve          map[string]interface{}
+		Filesystems    map[string]string
+		SnapshotPrefix string `mapstructure:"snapshot_prefix"`
+		Interval       string
+		Prune          map[string]interface{}
+		Debug          map[string]interface{}
+	}
+
+	if err = mapstructure.Decode(i, &asMap); err != nil {
+		err = errors.Wrap(err, "mapstructure error")
+		return nil, err
+	}
+
+	j = &SourceJob{Name: name}
+
+	if j.Serve, err = parseAuthenticatedChannelListenerFactory(c, asMap.Serve); err != nil {
+		return
+	}
+
+	if j.Filesystems, err = parseDatasetMapFilter(asMap.Filesystems, true); err != nil {
+		return
+	}
+
+	if j.SnapshotPrefix, err = parseSnapshotPrefix(asMap.SnapshotPrefix); err != nil {
+		return
+	}
+
+	if j.Interval, err = parsePostitiveDuration(asMap.Interval); err != nil {
+		err = errors.Wrap(err, "cannot parse 'interval'")
+		return
+	}
+
+	if j.Prune, err = parsePrunePolicy(asMap.Prune, true); err != nil {
+		err = errors.Wrap(err, "cannot parse 'prune'")
+		return
+	}
+
+	if err = mapstructure.Decode(asMap.Debug, &j.Debug); err != nil {
+		err = errors.Wrap(err, "cannot parse 'debug'")
+		return
+	}
+
+	return
+}
+
+func (j *SourceJob) JobName() string {
+	return j.Name
+}
+
+func (j *SourceJob) JobStart(ctx context.Context) {
+
+	log := ctx.Value(contextKeyLog).(Logger)
+	defer log.Info("exiting")
+
+	j.autosnapTask = NewTask("autosnap", log)
+	j.pruneTask = NewTask("prune", log)
+	j.serveTask = NewTask("serve", log)
+
+	a := IntervalAutosnap{j.autosnapTask, j.Filesystems, j.SnapshotPrefix, j.Interval}
+	p, err := j.Pruner(j.pruneTask, PrunePolicySideDefault, false)
+
+	if err != nil {
+		log.WithError(err).Error("error creating pruner")
+		return
+	}
+
+	didSnaps := make(chan struct{})
+
+	go j.serve(ctx, j.serveTask)
+	go a.Run(ctx, didSnaps)
+
+outer:
+	for {
+		select {
+		case <-ctx.Done():
+			break outer
+		case <-didSnaps:
+			log.Info("starting pruner")
+			p.Run(ctx)
+			log.Info("pruner done")
+		}
+	}
+	log.WithError(ctx.Err()).Info("context")
+
+}
+
+func (j *SourceJob) JobStatus(ctxt context.Context) (*JobStatus, error) {
+	return &JobStatus{
+		Tasks: []*TaskStatus{
+			j.autosnapTask.Status(),
+			j.pruneTask.Status(),
+			j.serveTask.Status(),
+		}}, nil
+}
+
+func (j *SourceJob) Pruner(task *Task, side PrunePolicySide, dryRun bool) (p Pruner, err error) {
+	p = Pruner{
+		task,
+		time.Now(),
+		dryRun,
+		j.Filesystems,
+		j.SnapshotPrefix,
+		j.Prune,
+	}
+	return
+}
+
+func (j *SourceJob) serve(ctx context.Context, task *Task) {
+
+	listener, err := j.Serve.Listen()
+	if err != nil {
+		task.Log().WithError(err).Error("error listening")
+		return
+	}
+
+	type rwcChanMsg struct {
+		rwc io.ReadWriteCloser
+		err error
+	}
+	rwcChan := make(chan rwcChanMsg)
+
+	// Serve connections until interrupted or error
+outer:
+	for {
+
+		go func() {
+			rwc, err := listener.Accept()
+			if err != nil {
+				rwcChan <- rwcChanMsg{rwc, err}
+				close(rwcChan)
+				return
+			}
+			rwcChan <- rwcChanMsg{rwc, err}
+		}()
+
+		select {
+
+		case rwcMsg := <-rwcChan:
+
+			if rwcMsg.err != nil {
+				task.Log().WithError(err).Error("error accepting connection")
+				break outer
+			}
+
+			j.handleConnection(rwcMsg.rwc, task)
+
+		case <-ctx.Done():
+			task.Log().WithError(ctx.Err()).Info("context")
+			break outer
+
+		}
+
+	}
+
+	task.Enter("close_listener")
+	defer task.Finish()
+	err = listener.Close()
+	if err != nil {
+		task.Log().WithError(err).Error("error closing listener")
+	}
+
+	return
+
+}
+
+func (j *SourceJob) handleConnection(rwc io.ReadWriteCloser, task *Task) {
+
+	task.Enter("handle_connection")
+	defer task.Finish()
+
+	task.Log().Info("handling client connection")
+
+	rwc, err := util.NewReadWriteCloserLogger(rwc, j.Debug.Conn.ReadDump, j.Debug.Conn.WriteDump)
+	if err != nil {
+		panic(err)
+	}
+
+	// construct connection handler
+	handler := NewHandler(task.Log(), j.Filesystems, NewPrefixFilter(j.SnapshotPrefix))
+
+	// handle connection
+	rpcServer := rpc.NewServer(rwc)
+	if j.Debug.RPC.Log {
+		rpclog := task.Log().WithField("subsystem", "rpc")
+		rpcServer.SetLogger(rpclog, true)
+	}
+	registerEndpoints(rpcServer, handler)
+	if err = rpcServer.Serve(); err != nil {
+		task.Log().WithError(err).Error("error serving connection")
+	}
+
+	// wait for client to close connection
+	// FIXME: we cannot just close it like we would to with a TCP socket because
+	// FIXME: go-nettsh's Close() may overtake the remaining data in the pipe
+	const CLIENT_HANGUP_TIMEOUT = 1 * time.Second
+	task.Log().
+		WithField("timeout", CLIENT_HANGUP_TIMEOUT).
+		Debug("waiting for client to hang up")
+
+	wchan := make(chan error)
+	go func() {
+		var pseudo [1]byte
+		_, err := io.ReadFull(rwc, pseudo[:])
+		wchan <- err
+	}()
+	var werr error
+	select {
+	case werr = <-wchan:
+		// all right
+	case <-time.After(CLIENT_HANGUP_TIMEOUT):
+		werr = errors.New("client did not close connection within timeout")
+	}
+	if werr != nil && werr != io.EOF {
+		task.Log().WithError(werr).
+			Error("error waiting for client to hang up")
+	}
+	task.Log().Info("closing client connection")
+	if err = rwc.Close(); err != nil {
+		task.Log().WithError(err).Error("error force-closing connection")
+	}
+}

cmd/config_logging.go

@@ -0,0 +1,251 @@
+package cmd
+
+import (
+	"crypto/tls"
+	"crypto/x509"
+	"github.com/mattn/go-isatty"
+	"github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/logger"
+	"io/ioutil"
+	"os"
+	"time"
+)
+
+type LoggingConfig struct {
+	Outlets *logger.Outlets
+}
+
+type MetadataFlags int64
+
+const (
+	MetadataTime MetadataFlags = 1 << iota
+	MetadataLevel
+
+	MetadataNone MetadataFlags = 0
+	MetadataAll  MetadataFlags = ^0
+)
+
+func parseLogging(i interface{}) (c *LoggingConfig, err error) {
+
+	c = &LoggingConfig{}
+	c.Outlets = logger.NewOutlets()
+
+	var asList []interface{}
+	if err = mapstructure.Decode(i, &asList); err != nil {
+		return nil, errors.Wrap(err, "mapstructure error")
+	}
+	if len(asList) == 0 {
+		// Default config
+		out := WriterOutlet{&HumanFormatter{}, os.Stdout}
+		c.Outlets.Add(out, logger.Warn)
+		return
+	}
+
+	var syslogOutlets, stdoutOutlets int
+	for lei, le := range asList {
+
+		outlet, minLevel, err := parseOutlet(le)
+		if err != nil {
+			return nil, errors.Wrapf(err, "cannot parse outlet #%d", lei)
+		}
+		var _ logger.Outlet = WriterOutlet{}
+		var _ logger.Outlet = &SyslogOutlet{}
+		switch outlet.(type) {
+		case *SyslogOutlet:
+			syslogOutlets++
+		case WriterOutlet:
+			stdoutOutlets++
+		}
+
+		c.Outlets.Add(outlet, minLevel)
+
+	}
+
+	if syslogOutlets > 1 {
+		return nil, errors.Errorf("can only define one 'syslog' outlet")
+	}
+	if stdoutOutlets > 1 {
+		return nil, errors.Errorf("can only define one 'stdout' outlet")
+	}
+
+	return c, nil
+
+}
+
+func parseLogFormat(i interface{}) (f EntryFormatter, err error) {
+	var is string
+	switch j := i.(type) {
+	case string:
+		is = j
+	default:
+		return nil, errors.Errorf("invalid log format: wrong type: %T", i)
+	}
+
+	switch is {
+	case "human":
+		return &HumanFormatter{}, nil
+	case "logfmt":
+		return &LogfmtFormatter{}, nil
+	case "json":
+		return &JSONFormatter{}, nil
+	default:
+		return nil, errors.Errorf("invalid log format: '%s'", is)
+	}
+
+}
+
+func parseOutlet(i interface{}) (o logger.Outlet, level logger.Level, err error) {
+
+	var in struct {
+		Outlet string
+		Level  string
+		Format string
+	}
+	if err = mapstructure.Decode(i, &in); err != nil {
+		err = errors.Wrap(err, "mapstructure error")
+		return
+	}
+	if in.Outlet == "" || in.Level == "" || in.Format == "" {
+		err = errors.Errorf("must specify 'outlet', 'level' and 'format' field")
+		return
+	}
+
+	minLevel, err := logger.ParseLevel(in.Level)
+	if err != nil {
+		err = errors.Wrap(err, "cannot parse 'level' field")
+		return
+	}
+	formatter, err := parseLogFormat(in.Format)
+	if err != nil {
+		err = errors.Wrap(err, "cannot parse")
+		return
+	}
+
+	switch in.Outlet {
+	case "stdout":
+		o, err = parseStdoutOutlet(i, formatter)
+	case "tcp":
+		o, err = parseTCPOutlet(i, formatter)
+	case "syslog":
+		o, err = parseSyslogOutlet(i, formatter)
+	default:
+		err = errors.Errorf("unknown outlet type '%s'", in.Outlet)
+	}
+	return o, minLevel, err
+
+}
+
+func parseStdoutOutlet(i interface{}, formatter EntryFormatter) (WriterOutlet, error) {
+
+	var in struct {
+		Time bool
+	}
+	if err := mapstructure.Decode(i, &in); err != nil {
+		return WriterOutlet{}, errors.Wrap(err, "invalid structure for stdout outlet")
+	}
+
+	flags := MetadataAll
+	writer := os.Stdout
+	if !isatty.IsTerminal(writer.Fd()) && !in.Time {
+		flags &= ^MetadataTime
+	}
+
+	formatter.SetMetadataFlags(flags)
+	return WriterOutlet{
+		formatter,
+		os.Stdout,
+	}, nil
+}
+
+func parseTCPOutlet(i interface{}, formatter EntryFormatter) (out *TCPOutlet, err error) {
+
+	var in struct {
+		Net           string
+		Address       string
+		RetryInterval string `mapstructure:"retry_interval"`
+		TLS           *struct {
+			CA   string
+			Cert string
+			Key  string
+		}
+	}
+	if err = mapstructure.Decode(i, &in); err != nil {
+		return nil, errors.Wrap(err, "mapstructure error")
+	}
+
+	retryInterval, err := time.ParseDuration(in.RetryInterval)
+	if err != nil {
+		return nil, errors.Wrap(err, "cannot parse 'retry_interval'")
+	}
+
+	if len(in.Net) == 0 {
+		return nil, errors.New("field 'net' must not be empty")
+	}
+	if len(in.Address) == 0 {
+		return nil, errors.New("field 'address' must not be empty")
+	}
+
+	var tlsConfig *tls.Config
+	if in.TLS != nil {
+
+		cert, err := tls.LoadX509KeyPair(in.TLS.Cert, in.TLS.Key)
+		if err != nil {
+			return nil, errors.Wrap(err, "cannot load client cert")
+		}
+
+		var rootCAs *x509.CertPool
+		if in.TLS.CA == "" {
+			if rootCAs, err = x509.SystemCertPool(); err != nil {
+				return nil, errors.Wrap(err, "cannot open system cert pool")
+			}
+		} else {
+			rootCAs = x509.NewCertPool()
+			rootCAPEM, err := ioutil.ReadFile(in.TLS.CA)
+			if err != nil {
+				return nil, errors.Wrap(err, "cannot load CA cert")
+			}
+			if !rootCAs.AppendCertsFromPEM(rootCAPEM) {
+				return nil, errors.New("cannot parse CA cert")
+			}
+		}
+		if err != nil && in.TLS.CA == "" {
+			return nil, errors.Wrap(err, "cannot load root ca pool")
+		}
+
+		tlsConfig = &tls.Config{
+			Certificates: []tls.Certificate{cert},
+			RootCAs:      rootCAs,
+		}
+
+		tlsConfig.BuildNameToCertificate()
+	}
+
+	formatter.SetMetadataFlags(MetadataAll)
+	return NewTCPOutlet(formatter, in.Net, in.Address, tlsConfig, retryInterval), nil
+
+}
+
+func parseSyslogOutlet(i interface{}, formatter EntryFormatter) (out *SyslogOutlet, err error) {
+
+	var in struct {
+		RetryInterval string `mapstructure:"retry_interval"`
+	}
+	if err = mapstructure.Decode(i, &in); err != nil {
+		return nil, errors.Wrap(err, "mapstructure error")
+	}
+
+	out = &SyslogOutlet{}
+	out.Formatter = formatter
+	out.Formatter.SetMetadataFlags(MetadataNone)
+
+	out.RetryInterval = 0 // default to 0 as we assume local syslog will just work
+	if in.RetryInterval != "" {
+		out.RetryInterval, err = time.ParseDuration(in.RetryInterval)
+		if err != nil {
+			return nil, errors.Wrap(err, "cannot parse 'retry_interval'")
+		}
+	}
+
+	return
+}

cmd/config_mapfilter.go

@@ -1,13 +1,12 @@
-package filters
+package cmd
 
 import (
 	"fmt"
 	"strings"
 
+	"github.com/mitchellh/mapstructure"
 	"github.com/pkg/errors"
-
+	"github.com/zrepl/zrepl/zfs"
-	"github.com/zrepl/zrepl/internal/endpoint"
-	"github.com/zrepl/zrepl/internal/zfs"
 )
 
 type DatasetMapFilter struct {
@@ -102,7 +101,6 @@ func (m DatasetMapFilter) mostSpecificPrefixMapping(path *zfs.DatasetPath) (idx
 	return
 }
 
-// Returns target == nil if there is no mapping
 func (m DatasetMapFilter) Map(source *zfs.DatasetPath) (target *zfs.DatasetPath, err error) {
 
 	if m.filterMode {
@@ -116,18 +114,10 @@ func (m DatasetMapFilter) Map(source *zfs.DatasetPath) (target *zfs.DatasetPath,
 	}
 	me := m.entries[mi]
 
-	if me.mapping == "" {
-		// Special case treatment: 'foo/bar<' => ''
-		if !me.subtreeMatch {
-			return nil, fmt.Errorf("mapping to '' must be a subtree match")
-		}
-		// ok...
-	} else {
 	if strings.HasPrefix("!", me.mapping) {
 		// reject mapping
 		return nil, nil
 	}
-	}
 
 	target, err = zfs.NewDatasetPath(me.mapping)
 	if err != nil {
@@ -160,16 +150,8 @@ func (m DatasetMapFilter) Filter(p *zfs.DatasetPath) (pass bool, err error) {
 	return
 }
 
-func (m DatasetMapFilter) UserSpecifiedDatasets() (datasets zfs.UserSpecifiedDatasetsSet) {
-	datasets = make(zfs.UserSpecifiedDatasetsSet)
-	for i := range m.entries {
-		datasets[m.entries[i].path.ToString()] = true
-	}
-	return
-}
-
 // Construct a new filter-only DatasetMapFilter from a mapping
-// The new filter allows exactly those paths that were not forbidden by the mapping.
+// The new filter allows excactly those paths that were not forbidden by the mapping.
 func (m DatasetMapFilter) InvertedFilter() (inv *DatasetMapFilter, err error) {
 
 	if m.filterMode {
@@ -185,7 +167,7 @@ func (m DatasetMapFilter) InvertedFilter() (inv *DatasetMapFilter, err error) {
 	for i, e := range m.entries {
 		inv.entries[i].path, err = zfs.NewDatasetPath(e.mapping)
 		if err != nil {
-			err = errors.Wrapf(err, "mapping cannot be inverted: '%s' is not a dataset path", e.mapping)
+			err = errors.Wrapf(err, "mapping cannot be inverted: '%s' is not a dataset path: %s", e.mapping)
 			return
 		}
 		inv.entries[i].mapping = MapFilterResultOk
@@ -195,43 +177,12 @@ func (m DatasetMapFilter) InvertedFilter() (inv *DatasetMapFilter, err error) {
 	return inv, nil
 }
 
-// FIXME investigate whether we can support more...
-func (m DatasetMapFilter) Invert() (endpoint.FSMap, error) {
-
-	if m.filterMode {
-		return nil, errors.Errorf("can only invert mappings")
-	}
-
-	if len(m.entries) != 1 {
-		return nil, errors.Errorf("inversion of complicated mappings is not implemented") // FIXME
-	}
-
-	e := m.entries[0]
-
-	inv := &DatasetMapFilter{
-		make([]datasetMapFilterEntry, len(m.entries)),
-		false,
-	}
-	mp, err := zfs.NewDatasetPath(e.mapping)
-	if err != nil {
-		return nil, err
-	}
-
-	inv.entries[0] = datasetMapFilterEntry{
-		path:         mp,
-		mapping:      e.path.ToString(),
-		subtreeMatch: e.subtreeMatch,
-	}
-
-	return inv, nil
-}
-
 // Creates a new DatasetMapFilter in filter mode from a mapping
 // All accepting mapping results are mapped to accepting filter results
 // All rejecting mapping results are mapped to rejecting filter results
-func (m DatasetMapFilter) AsFilter() endpoint.FSFilter {
+func (m DatasetMapFilter) AsFilter() (f *DatasetMapFilter) {
 
-	f := &DatasetMapFilter{
+	f = &DatasetMapFilter{
 		make([]datasetMapFilterEntry, len(m.entries)),
 		true,
 	}
@@ -266,14 +217,16 @@ func (m DatasetMapFilter) parseDatasetFilterResult(result string) (pass bool, er
 	return false, fmt.Errorf("'%s' is not a valid filter result", result)
 }
 
-func DatasetMapFilterFromConfig(in map[string]bool) (f *DatasetMapFilter, err error) {
+func parseDatasetMapFilter(mi interface{}, filterMode bool) (f *DatasetMapFilter, err error) {
 
-	f = NewDatasetMapFilter(len(in), true)
+	var m map[string]string
-	for pathPattern, accept := range in {
+	if err = mapstructure.Decode(mi, &m); err != nil {
-		mapping := MapFilterResultOmit
+		err = fmt.Errorf("maps / filters must be specified as map[string]string: %s", err)
-		if accept {
+		return
-			mapping = MapFilterResultOk
 	}
+
+	f = NewDatasetMapFilter(len(m), filterMode)
+	for pathPattern, mapping := range m {
 		if err = f.Add(pathPattern, mapping); err != nil {
 			err = fmt.Errorf("invalid mapping entry ['%s':'%s']: %s", pathPattern, mapping, err)
 			return

cmd/config_parse.go

@@ -0,0 +1,294 @@
+package cmd
+
+import (
+	"io/ioutil"
+
+	"fmt"
+	yaml "github.com/go-yaml/yaml"
+	"github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"os"
+	"regexp"
+	"strconv"
+	"time"
+)
+
+var ConfigFileDefaultLocations []string = []string{
+	"/etc/zrepl/zrepl.yml",
+	"/usr/local/etc/zrepl/zrepl.yml",
+}
+
+const (
+	JobNameControl string = "control"
+)
+
+var ReservedJobNames []string = []string{
+	JobNameControl,
+}
+
+type ConfigParsingContext struct {
+	Global *Global
+}
+
+func ParseConfig(path string) (config *Config, err error) {
+
+	if path == "" {
+		// Try default locations
+		for _, l := range ConfigFileDefaultLocations {
+			stat, err := os.Stat(l)
+			if err != nil {
+				continue
+			}
+			if !stat.Mode().IsRegular() {
+				err = errors.Errorf("file at default location is not a regular file: %s", l)
+				continue
+			}
+			path = l
+			break
+		}
+	}
+
+	var i interface{}
+
+	var bytes []byte
+
+	if bytes, err = ioutil.ReadFile(path); err != nil {
+		err = errors.WithStack(err)
+		return
+	}
+
+	if err = yaml.Unmarshal(bytes, &i); err != nil {
+		err = errors.WithStack(err)
+		return
+	}
+
+	return parseConfig(i)
+}
+
+func parseConfig(i interface{}) (c *Config, err error) {
+
+	var asMap struct {
+		Global map[string]interface{}
+		Jobs   []map[string]interface{}
+	}
+	if err := mapstructure.Decode(i, &asMap); err != nil {
+		return nil, errors.Wrap(err, "config root must be a dict")
+	}
+
+	c = &Config{}
+
+	// Parse global with defaults
+	c.Global.Serve.Stdinserver.SockDir = "/var/run/zrepl/stdinserver"
+	c.Global.Control.Sockpath = "/var/run/zrepl/control"
+
+	err = mapstructure.Decode(asMap.Global, &c.Global)
+	if err != nil {
+		err = errors.Wrap(err, "mapstructure error on 'global' section: %s")
+		return
+	}
+
+	if c.Global.logging, err = parseLogging(asMap.Global["logging"]); err != nil {
+		return nil, errors.Wrap(err, "cannot parse logging section")
+	}
+
+	cpc := ConfigParsingContext{&c.Global}
+	jpc := JobParsingContext{cpc}
+
+	// Jobs
+	c.Jobs = make(map[string]Job, len(asMap.Jobs))
+	for i := range asMap.Jobs {
+		job, err := parseJob(jpc, asMap.Jobs[i])
+		if err != nil {
+			// Try to find its name
+			namei, ok := asMap.Jobs[i]["name"]
+			if !ok {
+				namei = fmt.Sprintf("<no name, entry #%d in list>", i)
+			}
+			err = errors.Wrapf(err, "cannot parse job '%v'", namei)
+			return nil, err
+		}
+		jn := job.JobName()
+		if _, ok := c.Jobs[jn]; ok {
+			err = errors.Errorf("duplicate job name: %s", jn)
+			return nil, err
+		}
+		c.Jobs[job.JobName()] = job
+	}
+
+	cj, err := NewControlJob(JobNameControl, jpc.Global.Control.Sockpath)
+	if err != nil {
+		err = errors.Wrap(err, "cannot create control job")
+		return
+	}
+	c.Jobs[JobNameControl] = cj
+
+	return c, nil
+
+}
+
+func extractStringField(i map[string]interface{}, key string, notempty bool) (field string, err error) {
+	vi, ok := i[key]
+	if !ok {
+		err = errors.Errorf("must have field '%s'", key)
+		return "", err
+	}
+	field, ok = vi.(string)
+	if !ok {
+		err = errors.Errorf("'%s' field must have type string", key)
+		return "", err
+	}
+	if notempty && len(field) <= 0 {
+		err = errors.Errorf("'%s' field must not be empty", key)
+		return "", err
+	}
+	return
+}
+
+type JobParsingContext struct {
+	ConfigParsingContext
+}
+
+func parseJob(c JobParsingContext, i map[string]interface{}) (j Job, err error) {
+
+	name, err := extractStringField(i, "name", true)
+	if err != nil {
+		return
+	}
+
+	for _, r := range ReservedJobNames {
+		if name == r {
+			err = errors.Errorf("job name '%s' is reserved", name)
+			return
+		}
+	}
+
+	jobtype, err := extractStringField(i, "type", true)
+	if err != nil {
+		return
+	}
+
+	switch jobtype {
+	case "pull":
+		return parsePullJob(c, name, i)
+	case "source":
+		return parseSourceJob(c, name, i)
+	case "local":
+		return parseLocalJob(c, name, i)
+	default:
+		return nil, errors.Errorf("unknown job type '%s'", jobtype)
+	}
+
+}
+
+func parseConnect(i map[string]interface{}) (c RWCConnecter, err error) {
+
+	t, err := extractStringField(i, "type", true)
+	if err != nil {
+		return nil, err
+	}
+
+	switch t {
+	case "ssh+stdinserver":
+		return parseSSHStdinserverConnecter(i)
+	default:
+		return nil, errors.Errorf("unknown connection type '%s'", t)
+	}
+
+}
+
+func parseInitialReplPolicy(v interface{}, defaultPolicy InitialReplPolicy) (p InitialReplPolicy, err error) {
+	s, ok := v.(string)
+	if !ok {
+		goto err
+	}
+
+	switch {
+	case s == "":
+		p = defaultPolicy
+	case s == "most_recent":
+		p = InitialReplPolicyMostRecent
+	case s == "all":
+		p = InitialReplPolicyAll
+	default:
+		goto err
+	}
+
+	return
+
+err:
+	err = errors.New(fmt.Sprintf("expected InitialReplPolicy, got %#v", v))
+	return
+}
+
+func parsePrunePolicy(v map[string]interface{}, willSeeBookmarks bool) (p PrunePolicy, err error) {
+
+	policyName, err := extractStringField(v, "policy", true)
+	if err != nil {
+		return
+	}
+
+	switch policyName {
+	case "grid":
+		return parseGridPrunePolicy(v, willSeeBookmarks)
+	case "noprune":
+		return NoPrunePolicy{}, nil
+	default:
+		err = errors.Errorf("unknown policy '%s'", policyName)
+		return
+	}
+}
+
+func parseAuthenticatedChannelListenerFactory(c JobParsingContext, v map[string]interface{}) (p AuthenticatedChannelListenerFactory, err error) {
+
+	t, err := extractStringField(v, "type", true)
+	if err != nil {
+		return nil, err
+	}
+
+	switch t {
+	case "stdinserver":
+		return parseStdinserverListenerFactory(c, v)
+	default:
+		err = errors.Errorf("unknown type '%s'", t)
+		return
+	}
+
+}
+
+var durationStringRegex *regexp.Regexp = regexp.MustCompile(`^\s*(\d+)\s*(s|m|h|d|w)\s*$`)
+
+func parsePostitiveDuration(e string) (d time.Duration, err error) {
+	comps := durationStringRegex.FindStringSubmatch(e)
+	if len(comps) != 3 {
+		err = fmt.Errorf("does not match regex: %s %#v", e, comps)
+		return
+	}
+
+	durationFactor, err := strconv.ParseInt(comps[1], 10, 64)
+	if err != nil {
+		return 0, err
+	}
+	if durationFactor <= 0 {
+		return 0, errors.New("duration must be positive integer")
+	}
+
+	var durationUnit time.Duration
+	switch comps[2] {
+	case "s":
+		durationUnit = time.Second
+	case "m":
+		durationUnit = time.Minute
+	case "h":
+		durationUnit = time.Hour
+	case "d":
+		durationUnit = 24 * time.Hour
+	case "w":
+		durationUnit = 24 * 7 * time.Hour
+	default:
+		err = fmt.Errorf("contains unknown time unit '%s'", comps[2])
+		return
+	}
+
+	d = time.Duration(durationFactor) * durationUnit
+	return
+}

cmd/config_prune_grid.go

@@ -0,0 +1,245 @@
+package cmd
+
+import (
+	"fmt"
+	"github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/util"
+	"github.com/zrepl/zrepl/zfs"
+	"math"
+	"regexp"
+	"sort"
+	"strconv"
+	"strings"
+	"time"
+)
+
+type GridPrunePolicy struct {
+	RetentionGrid *util.RetentionGrid
+	MaxBookmarks  int
+}
+
+const GridPrunePolicyMaxBookmarksKeepAll = -1
+
+type retentionGridAdaptor struct {
+	zfs.FilesystemVersion
+}
+
+func (a retentionGridAdaptor) Date() time.Time {
+	return a.Creation
+}
+
+func (a retentionGridAdaptor) LessThan(b util.RetentionGridEntry) bool {
+	return a.CreateTXG < b.(retentionGridAdaptor).CreateTXG
+}
+
+// Prune filters snapshots with the retention grid.
+// Bookmarks are deleted such that KeepBookmarks are kept in the end.
+// The oldest bookmarks are removed first.
+func (p *GridPrunePolicy) Prune(_ *zfs.DatasetPath, versions []zfs.FilesystemVersion) (keep, remove []zfs.FilesystemVersion, err error) {
+	skeep, sremove := p.pruneSnapshots(versions)
+	keep, remove = p.pruneBookmarks(skeep)
+	remove = append(remove, sremove...)
+	return keep, remove, nil
+}
+
+func (p *GridPrunePolicy) pruneSnapshots(versions []zfs.FilesystemVersion) (keep, remove []zfs.FilesystemVersion) {
+
+	// Build adaptors for retention grid
+	keep = []zfs.FilesystemVersion{}
+	adaptors := make([]util.RetentionGridEntry, 0)
+	for fsv := range versions {
+		if versions[fsv].Type != zfs.Snapshot {
+			keep = append(keep, versions[fsv])
+			continue
+		}
+		adaptors = append(adaptors, retentionGridAdaptor{versions[fsv]})
+	}
+
+	sort.SliceStable(adaptors, func(i, j int) bool {
+		return adaptors[i].LessThan(adaptors[j])
+	})
+	now := adaptors[len(adaptors)-1].Date()
+
+	// Evaluate retention grid
+	keepa, removea := p.RetentionGrid.FitEntries(now, adaptors)
+
+	// Revert adaptors
+	for i := range keepa {
+		keep = append(keep, keepa[i].(retentionGridAdaptor).FilesystemVersion)
+	}
+	remove = make([]zfs.FilesystemVersion, len(removea))
+	for i := range removea {
+		remove[i] = removea[i].(retentionGridAdaptor).FilesystemVersion
+	}
+	return
+
+}
+
+func (p *GridPrunePolicy) pruneBookmarks(versions []zfs.FilesystemVersion) (keep, remove []zfs.FilesystemVersion) {
+
+	if p.MaxBookmarks == GridPrunePolicyMaxBookmarksKeepAll {
+		return versions, []zfs.FilesystemVersion{}
+	}
+
+	keep = []zfs.FilesystemVersion{}
+	bookmarks := make([]zfs.FilesystemVersion, 0)
+	for fsv := range versions {
+		if versions[fsv].Type != zfs.Bookmark {
+			keep = append(keep, versions[fsv])
+			continue
+		}
+		bookmarks = append(bookmarks, versions[fsv])
+	}
+
+	if len(bookmarks) == 0 {
+		return keep, []zfs.FilesystemVersion{}
+	}
+	if len(bookmarks) < p.MaxBookmarks {
+		keep = append(keep, bookmarks...)
+		return keep, []zfs.FilesystemVersion{}
+	}
+
+	// NOTE: sorting descending by descending by createtxg <=> sorting ascending wrt creation time
+	sort.SliceStable(bookmarks, func(i, j int) bool {
+		return (bookmarks[i].CreateTXG > bookmarks[j].CreateTXG)
+	})
+
+	keep = append(keep, bookmarks[:p.MaxBookmarks]...)
+	remove = bookmarks[p.MaxBookmarks:]
+
+	return keep, remove
+}
+
+func parseGridPrunePolicy(e map[string]interface{}, willSeeBookmarks bool) (p *GridPrunePolicy, err error) {
+
+	const KeepBookmarksAllString = "all"
+	var i struct {
+		Grid          string
+		KeepBookmarks string `mapstructure:"keep_bookmarks"`
+	}
+
+	dec, err := mapstructure.NewDecoder(&mapstructure.DecoderConfig{Result: &i, WeaklyTypedInput: true})
+	if err != nil {
+		err = errors.Wrap(err, "mapstructure error")
+		return
+	}
+	if err = dec.Decode(e); err != nil {
+		err = errors.Wrapf(err, "mapstructure error")
+		return
+	}
+
+	// Parse grid
+	intervals, err := parseRetentionGridIntervalsString(i.Grid)
+	if err != nil {
+		err = fmt.Errorf("cannot parse retention grid: %s", err)
+		return
+	}
+	// Assert intervals are of increasing length (not necessarily required, but indicates config mistake)
+	lastDuration := time.Duration(0)
+	for i := range intervals {
+
+		if intervals[i].Length < lastDuration {
+			// If all intervals before were keep=all, this is ok
+			allPrevKeepCountAll := true
+			for j := i - 1; allPrevKeepCountAll && j >= 0; j-- {
+				allPrevKeepCountAll = intervals[j].KeepCount == util.RetentionGridKeepCountAll
+			}
+			if allPrevKeepCountAll {
+				goto isMonotonicIncrease
+			}
+			err = errors.New("retention grid interval length must be monotonically increasing")
+			return
+		}
+	isMonotonicIncrease:
+		lastDuration = intervals[i].Length
+
+	}
+
+	// Parse KeepBookmarks
+	keepBookmarks := 0
+	if i.KeepBookmarks == KeepBookmarksAllString || (i.KeepBookmarks == "" && !willSeeBookmarks) {
+		keepBookmarks = GridPrunePolicyMaxBookmarksKeepAll
+	} else {
+		i, err := strconv.ParseInt(i.KeepBookmarks, 10, 32)
+		if err != nil || i <= 0 || i > math.MaxInt32 {
+			return nil, errors.Errorf("keep_bookmarks must be positive integer or 'all'")
+		}
+		keepBookmarks = int(i)
+	}
+	return &GridPrunePolicy{
+		util.NewRetentionGrid(intervals),
+		keepBookmarks,
+	}, nil
+}
+
+var retentionStringIntervalRegex *regexp.Regexp = regexp.MustCompile(`^\s*(\d+)\s*x\s*([^\(]+)\s*(\((.*)\))?\s*$`)
+
+func parseRetentionGridIntervalString(e string) (intervals []util.RetentionInterval, err error) {
+
+	comps := retentionStringIntervalRegex.FindStringSubmatch(e)
+	if comps == nil {
+		err = fmt.Errorf("retention string does not match expected format")
+		return
+	}
+
+	times, err := strconv.Atoi(comps[1])
+	if err != nil {
+		return nil, err
+	} else if times <= 0 {
+		return nil, fmt.Errorf("contains factor <= 0")
+	}
+
+	duration, err := parsePostitiveDuration(comps[2])
+	if err != nil {
+		return nil, err
+	}
+
+	keepCount := 1
+	if comps[3] != "" {
+		// Decompose key=value, comma separated
+		// For now, only keep_count is supported
+		re := regexp.MustCompile(`^\s*keep=(.+)\s*$`)
+		res := re.FindStringSubmatch(comps[4])
+		if res == nil || len(res) != 2 {
+			err = fmt.Errorf("interval parameter contains unknown parameters")
+			return
+		}
+		if res[1] == "all" {
+			keepCount = util.RetentionGridKeepCountAll
+		} else {
+			keepCount, err = strconv.Atoi(res[1])
+			if err != nil {
+				err = fmt.Errorf("cannot parse keep_count value")
+				return
+			}
+		}
+	}
+
+	intervals = make([]util.RetentionInterval, times)
+	for i := range intervals {
+		intervals[i] = util.RetentionInterval{
+			Length:    duration,
+			KeepCount: keepCount,
+		}
+	}
+
+	return
+
+}
+
+func parseRetentionGridIntervalsString(s string) (intervals []util.RetentionInterval, err error) {
+
+	ges := strings.Split(s, "|")
+	intervals = make([]util.RetentionInterval, 0, 7*len(ges))
+
+	for intervalIdx, e := range ges {
+		parsed, err := parseRetentionGridIntervalString(e)
+		if err != nil {
+			return nil, fmt.Errorf("cannot parse interval %d of %d: %s: %s", intervalIdx+1, len(ges), err, strings.TrimSpace(e))
+		}
+		intervals = append(intervals, parsed...)
+	}
+
+	return
+}

cmd/config_prune_none.go

@@ -0,0 +1,11 @@
+package cmd
+
+import "github.com/zrepl/zrepl/zfs"
+
+type NoPrunePolicy struct{}
+
+func (p NoPrunePolicy) Prune(fs *zfs.DatasetPath, versions []zfs.FilesystemVersion) (keep, remove []zfs.FilesystemVersion, err error) {
+	keep = versions
+	remove = []zfs.FilesystemVersion{}
+	return
+}

cmd/config_serve_stdinserver.go

@@ -0,0 +1,56 @@
+package cmd
+
+import (
+	"github.com/mitchellh/mapstructure"
+	"github.com/pkg/errors"
+	"github.com/problame/go-netssh"
+	"io"
+	"path"
+)
+
+type StdinserverListenerFactory struct {
+	ClientIdentity string `mapstructure:"client_identity"`
+	sockpath       string
+}
+
+func parseStdinserverListenerFactory(c JobParsingContext, i map[string]interface{}) (f *StdinserverListenerFactory, err error) {
+
+	f = &StdinserverListenerFactory{}
+
+	if err = mapstructure.Decode(i, f); err != nil {
+		return nil, errors.Wrap(err, "mapstructure error")
+	}
+	if !(len(f.ClientIdentity) > 0) {
+		err = errors.Errorf("must specify 'client_identity'")
+		return
+	}
+
+	f.sockpath = path.Join(c.Global.Serve.Stdinserver.SockDir, f.ClientIdentity)
+
+	return
+}
+
+func (f *StdinserverListenerFactory) Listen() (al AuthenticatedChannelListener, err error) {
+
+	if err = PreparePrivateSockpath(f.sockpath); err != nil {
+		return nil, err
+	}
+
+	l, err := netssh.Listen(f.sockpath)
+	if err != nil {
+		return nil, err
+	}
+	return StdinserverListener{l}, nil
+}
+
+type StdinserverListener struct {
+	l *netssh.Listener
+}
+
+func (l StdinserverListener) Accept() (ch io.ReadWriteCloser, err error) {
+	return l.l.Accept()
+}
+
+func (l StdinserverListener) Close() (err error) {
+	return l.l.Close()
+}

cmd/config_test.go

@@ -0,0 +1,222 @@
+package cmd
+
+import (
+	"testing"
+	"time"
+
+	"github.com/kr/pretty"
+	"github.com/stretchr/testify/assert"
+	"github.com/zrepl/zrepl/util"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+func TestSampleConfigsAreParsedWithoutErrors(t *testing.T) {
+
+	paths := []string{
+		"./sampleconf/localbackup/host1.yml",
+		"./sampleconf/pullbackup/backuphost.yml",
+		"./sampleconf/pullbackup/productionhost.yml",
+		"./sampleconf/random/debugging.yml",
+		"./sampleconf/random/logging.yml",
+	}
+
+	for _, p := range paths {
+
+		c, err := ParseConfig(p)
+		if err != nil {
+			t.Errorf("error parsing %s:\n%+v", p, err)
+		}
+
+		t.Logf("file: %s", p)
+		t.Log(pretty.Sprint(c))
+
+	}
+
+}
+
+func TestParseRetentionGridStringParsing(t *testing.T) {
+
+	intervals, err := parseRetentionGridIntervalsString("2x10m(keep=2) | 1x1h | 3x1w")
+
+	assert.Nil(t, err)
+	assert.Len(t, intervals, 6)
+	proto := util.RetentionInterval{
+		KeepCount: 2,
+		Length:    10 * time.Minute,
+	}
+	assert.EqualValues(t, proto, intervals[0])
+	assert.EqualValues(t, proto, intervals[1])
+
+	proto.KeepCount = 1
+	proto.Length = 1 * time.Hour
+	assert.EqualValues(t, proto, intervals[2])
+
+	proto.Length = 7 * 24 * time.Hour
+	assert.EqualValues(t, proto, intervals[3])
+	assert.EqualValues(t, proto, intervals[4])
+	assert.EqualValues(t, proto, intervals[5])
+
+	intervals, err = parseRetentionGridIntervalsString("|")
+	assert.Error(t, err)
+	intervals, err = parseRetentionGridIntervalsString("2x10m")
+	assert.NoError(t, err)
+
+	intervals, err = parseRetentionGridIntervalsString("1x10m(keep=all)")
+	assert.NoError(t, err)
+	assert.Len(t, intervals, 1)
+	assert.EqualValues(t, util.RetentionGridKeepCountAll, intervals[0].KeepCount)
+
+}
+
+func TestDatasetMapFilter(t *testing.T) {
+
+	expectMapping := func(m map[string]string, from, to string) {
+		dmf, err := parseDatasetMapFilter(m, false)
+		if err != nil {
+			t.Logf("expect test map to be valid: %s", err)
+			t.FailNow()
+		}
+		fromPath, err := zfs.NewDatasetPath(from)
+		if err != nil {
+			t.Logf("expect test from path to be valid: %s", err)
+			t.FailNow()
+		}
+
+		res, err := dmf.Map(fromPath)
+		if to == "" {
+			assert.Nil(t, res)
+			assert.Nil(t, err)
+			t.Logf("%s => NOT MAPPED", fromPath.ToString())
+			return
+		}
+
+		assert.Nil(t, err)
+		toPath, err := zfs.NewDatasetPath(to)
+		if err != nil {
+			t.Logf("expect test to path to be valid: %s", err)
+			t.FailNow()
+		}
+		assert.True(t, res.Equal(toPath))
+	}
+
+	expectFilter := func(m map[string]string, path string, pass bool) {
+		dmf, err := parseDatasetMapFilter(m, true)
+		if err != nil {
+			t.Logf("expect test filter to be valid: %s", err)
+			t.FailNow()
+		}
+		p, err := zfs.NewDatasetPath(path)
+		if err != nil {
+			t.Logf("expect test path to be valid: %s", err)
+			t.FailNow()
+		}
+		res, err := dmf.Filter(p)
+		assert.Nil(t, err)
+		assert.Equal(t, pass, res)
+	}
+
+	map1 := map[string]string{
+		"a/b/c<":     "root1",
+		"a/b<":       "root2",
+		"<":          "root3/b/c",
+		"b":          "!",
+		"a/b/c/d/e<": "!",
+		"q<":         "root4/1/2",
+	}
+
+	expectMapping(map1, "a/b/c", "root1")
+	expectMapping(map1, "a/b/c/d", "root1/d")
+	expectMapping(map1, "a/b/c/d/e", "")
+	expectMapping(map1, "a/b/e", "root2/e")
+	expectMapping(map1, "a/b", "root2")
+	expectMapping(map1, "x", "root3/b/c/x")
+	expectMapping(map1, "x/y", "root3/b/c/x/y")
+	expectMapping(map1, "q", "root4/1/2")
+	expectMapping(map1, "b", "")
+	expectMapping(map1, "q/r", "root4/1/2/r")
+
+	filter1 := map[string]string{
+		"<":    "!",
+		"a<":   "ok",
+		"a/b<": "!",
+	}
+
+	expectFilter(filter1, "b", false)
+	expectFilter(filter1, "a", true)
+	expectFilter(filter1, "a/d", true)
+	expectFilter(filter1, "a/b", false)
+	expectFilter(filter1, "a/b/c", false)
+
+	filter2 := map[string]string{}
+	expectFilter(filter2, "foo", false) // default to omit
+
+}
+
+func TestDatasetMapFilter_AsFilter(t *testing.T) {
+
+	mapspec := map[string]string{
+		"a/b/c<":     "root1",
+		"a/b<":       "root2",
+		"<":          "root3/b/c",
+		"b":          "!",
+		"a/b/c/d/e<": "!",
+		"q<":         "root4/1/2",
+	}
+
+	m, err := parseDatasetMapFilter(mapspec, false)
+	assert.Nil(t, err)
+
+	f := m.AsFilter()
+
+	t.Logf("Mapping:\n%s\nFilter:\n%s", pretty.Sprint(m), pretty.Sprint(f))
+
+	tf := func(f zfs.DatasetFilter, path string, pass bool) {
+		p, err := zfs.NewDatasetPath(path)
+		assert.Nil(t, err)
+		r, err := f.Filter(p)
+		assert.Nil(t, err)
+		assert.Equal(t, pass, r)
+	}
+
+	tf(f, "a/b/c", true)
+	tf(f, "a/b", true)
+	tf(f, "b", false)
+	tf(f, "a/b/c/d/e", false)
+	tf(f, "a/b/c/d/e/f", false)
+	tf(f, "a", true)
+
+}
+
+func TestDatasetMapFilter_InvertedFilter(t *testing.T) {
+	mapspec := map[string]string{
+		"a/b":      "1/2",
+		"a/b/c<":   "3",
+		"a/b/c/d<": "1/2/a",
+		"a/b/d":    "!",
+	}
+
+	m, err := parseDatasetMapFilter(mapspec, false)
+	assert.Nil(t, err)
+
+	inv, err := m.InvertedFilter()
+	assert.Nil(t, err)
+
+	t.Log(pretty.Sprint(inv))
+
+	expectMapping := func(m *DatasetMapFilter, ps string, expRes bool) {
+		p, err := zfs.NewDatasetPath(ps)
+		assert.Nil(t, err)
+		r, err := m.Filter(p)
+		assert.Nil(t, err)
+		assert.Equal(t, expRes, r)
+	}
+
+	expectMapping(inv, "4", false)
+	expectMapping(inv, "3", true)
+	expectMapping(inv, "3/x", true)
+	expectMapping(inv, "1", false)
+	expectMapping(inv, "1/2", true)
+	expectMapping(inv, "1/2/3", false)
+	expectMapping(inv, "1/2/a/b", true)
+
+}

cmd/control.go

@@ -0,0 +1,305 @@
+package cmd
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"github.com/dustin/go-humanize"
+	"github.com/pkg/errors"
+	"github.com/spf13/cobra"
+	"github.com/zrepl/zrepl/logger"
+	"io"
+	golog "log"
+	"net"
+	"net/http"
+	"os"
+	"sort"
+	"strings"
+	"time"
+)
+
+var controlCmd = &cobra.Command{
+	Use:   "control",
+	Short: "control zrepl daemon",
+}
+
+var pprofCmd = &cobra.Command{
+	Use:   "pprof off | [on TCP_LISTEN_ADDRESS]",
+	Short: "start a http server exposing go-tool-compatible profiling endpoints at TCP_LISTEN_ADDRESS",
+	Run:   doControlPProf,
+	PreRunE: func(cmd *cobra.Command, args []string) error {
+		if cmd.Flags().NArg() < 1 {
+			goto enargs
+		}
+		switch cmd.Flags().Arg(0) {
+		case "on":
+			pprofCmdArgs.msg.Run = true
+			if cmd.Flags().NArg() != 2 {
+				return errors.New("must specify TCP_LISTEN_ADDRESS as second positional argument")
+			}
+			pprofCmdArgs.msg.HttpListenAddress = cmd.Flags().Arg(1)
+		case "off":
+			if cmd.Flags().NArg() != 1 {
+				goto enargs
+			}
+			pprofCmdArgs.msg.Run = false
+		}
+		return nil
+	enargs:
+		return errors.New("invalid number of positional arguments")
+
+	},
+}
+var pprofCmdArgs struct {
+	msg PprofServerControlMsg
+}
+
+var controlVersionCmd = &cobra.Command{
+	Use:   "version",
+	Short: "print version of running zrepl daemon",
+	Run:   doControLVersionCmd,
+}
+
+var controlStatusCmdArgs struct {
+	format      string
+	level       logger.Level
+	onlyShowJob string
+}
+
+var controlStatusCmd = &cobra.Command{
+	Use:   "status [JOB_NAME]",
+	Short: "get current status",
+	Run:   doControlStatusCmd,
+}
+
+func init() {
+	RootCmd.AddCommand(controlCmd)
+	controlCmd.AddCommand(pprofCmd)
+	controlCmd.AddCommand(controlVersionCmd)
+	controlCmd.AddCommand(controlStatusCmd)
+	controlStatusCmd.Flags().StringVar(&controlStatusCmdArgs.format, "format", "human", "output format (human|raw)")
+	controlStatusCmdArgs.level = logger.Warn
+	controlStatusCmd.Flags().Var(&controlStatusCmdArgs.level, "level", "minimum log level to show")
+}
+
+func controlHttpClient() (client http.Client, err error) {
+
+	conf, err := ParseConfig(rootArgs.configFile)
+	if err != nil {
+		return http.Client{}, err
+	}
+
+	return http.Client{
+		Transport: &http.Transport{
+			DialContext: func(_ context.Context, _, _ string) (net.Conn, error) {
+				return net.Dial("unix", conf.Global.Control.Sockpath)
+			},
+		},
+	}, nil
+}
+
+func doControlPProf(cmd *cobra.Command, args []string) {
+
+	log := golog.New(os.Stderr, "", 0)
+
+	die := func() {
+		log.Printf("exiting after error")
+		os.Exit(1)
+	}
+
+	log.Printf("connecting to zrepl daemon")
+	httpc, err := controlHttpClient()
+	if err != nil {
+		log.Printf("error parsing config: %s", err)
+		die()
+	}
+
+	var buf bytes.Buffer
+	if err := json.NewEncoder(&buf).Encode(&pprofCmdArgs.msg); err != nil {
+		log.Printf("error marshaling request: %s", err)
+		die()
+	}
+	_, err = httpc.Post("http://unix"+ControlJobEndpointPProf, "application/json", &buf)
+	if err != nil {
+		log.Printf("error: %s", err)
+		die()
+	}
+
+	log.Printf("finished")
+}
+
+func doControLVersionCmd(cmd *cobra.Command, args []string) {
+
+	log := golog.New(os.Stderr, "", 0)
+
+	die := func() {
+		log.Printf("exiting after error")
+		os.Exit(1)
+	}
+
+	httpc, err := controlHttpClient()
+	if err != nil {
+		log.Printf("could not connect to daemon: %s", err)
+		die()
+	}
+
+	resp, err := httpc.Get("http://unix" + ControlJobEndpointVersion)
+	if err != nil {
+		log.Printf("error: %s", err)
+		die()
+	} else if resp.StatusCode != http.StatusOK {
+		var msg bytes.Buffer
+		io.CopyN(&msg, resp.Body, 4096)
+		log.Printf("error: %s", msg.String())
+		die()
+	}
+
+	var info ZreplVersionInformation
+	err = json.NewDecoder(resp.Body).Decode(&info)
+	if err != nil {
+		log.Printf("error unmarshaling response: %s", err)
+		die()
+	}
+
+	fmt.Println(info.String())
+
+}
+
+func doControlStatusCmd(cmd *cobra.Command, args []string) {
+
+	log := golog.New(os.Stderr, "", 0)
+
+	die := func() {
+		log.Print("exiting after error")
+		os.Exit(1)
+	}
+
+	if len(args) == 1 {
+		controlStatusCmdArgs.onlyShowJob = args[0]
+	} else if len(args) > 1 {
+		log.Print("can only specify one job as positional argument")
+		cmd.Usage()
+		die()
+	}
+
+	httpc, err := controlHttpClient()
+	if err != nil {
+		log.Printf("could not connect to daemon: %s", err)
+		die()
+	}
+
+	resp, err := httpc.Get("http://unix" + ControlJobEndpointStatus)
+	if err != nil {
+		log.Printf("error: %s", err)
+		die()
+	} else if resp.StatusCode != http.StatusOK {
+		var msg bytes.Buffer
+		io.CopyN(&msg, resp.Body, 4096)
+		log.Printf("error: %s", msg.String())
+		die()
+	}
+
+	var status DaemonStatus
+	err = json.NewDecoder(resp.Body).Decode(&status)
+	if err != nil {
+		log.Printf("error unmarshaling response: %s", err)
+		die()
+	}
+
+	switch controlStatusCmdArgs.format {
+	case "raw":
+		enc := json.NewEncoder(os.Stdout)
+		enc.SetIndent("", "  ")
+		if err := enc.Encode(status); err != nil {
+			log.Panic(err)
+		}
+	case "human":
+
+		formatter := HumanFormatter{}
+		formatter.SetMetadataFlags(MetadataAll)
+		formatter.SetIgnoreFields([]string{
+			logJobField,
+		})
+		jobNames := make([]string, 0, len(status.Jobs))
+		for name, _ := range status.Jobs {
+			jobNames = append(jobNames, name)
+		}
+		sort.Slice(jobNames, func(i, j int) bool {
+			return strings.Compare(jobNames[i], jobNames[j]) == -1
+		})
+		now := time.Now()
+		for _, name := range jobNames {
+
+			if controlStatusCmdArgs.onlyShowJob != "" && name != controlStatusCmdArgs.onlyShowJob {
+				continue
+			}
+
+			job := status.Jobs[name]
+			jobLogEntries := make([]logger.Entry, 0)
+			informAboutError := false
+
+			fmt.Printf("Job '%s':\n", name)
+			for _, task := range job.Tasks {
+
+				var header bytes.Buffer
+				fmt.Fprintf(&header, "  Task '%s': ", task.Name)
+				if !task.Idle {
+					fmt.Fprint(&header, strings.Join(task.ActivityStack, "."))
+				} else {
+					fmt.Fprint(&header, "<idle>")
+				}
+				fmt.Fprint(&header, " ")
+				const TASK_STALLED_HOLDOFF_DURATION = 10 * time.Second
+				sinceLastUpdate := now.Sub(task.LastUpdate)
+				if !task.Idle || task.ProgressRx != 0 || task.ProgressTx != 0 {
+					fmt.Fprintf(&header, "(%s / %s , Rx/Tx",
+						humanize.Bytes(uint64(task.ProgressRx)),
+						humanize.Bytes(uint64(task.ProgressTx)))
+					if task.Idle {
+						fmt.Fprint(&header, ", values from last run")
+					}
+					fmt.Fprint(&header, ")")
+				}
+				fmt.Fprint(&header, "\n")
+				if !task.Idle && !task.LastUpdate.IsZero() && sinceLastUpdate >= TASK_STALLED_HOLDOFF_DURATION {
+					informAboutError = true
+					fmt.Fprintf(&header, "    WARNING: last update %s ago at %s)",
+						sinceLastUpdate.String(),
+						task.LastUpdate.Format(HumanFormatterDateFormat))
+					fmt.Fprint(&header, "\n")
+				}
+				io.Copy(os.Stdout, &header)
+
+				jobLogEntries = append(jobLogEntries, task.LogEntries...)
+				informAboutError = informAboutError || task.MaxLogLevel >= logger.Warn
+			}
+
+			sort.Slice(jobLogEntries, func(i, j int) bool {
+				return jobLogEntries[i].Time.Before(jobLogEntries[j].Time)
+			})
+			if informAboutError {
+				fmt.Println("    WARNING: Some tasks encountered problems since the last time they left idle state:")
+				fmt.Println("             check the logs below or your log file for more information.")
+				fmt.Println("             Use the --level flag if you need debug information.")
+				fmt.Println()
+			}
+			for _, e := range jobLogEntries {
+				if e.Level < controlStatusCmdArgs.level {
+					continue
+				}
+				formatted, err := formatter.Format(&e)
+				if err != nil {
+					panic(err)
+				}
+				fmt.Printf("      %s\n", string(formatted))
+			}
+			fmt.Println()
+
+		}
+	default:
+		log.Printf("invalid output format '%s'", controlStatusCmdArgs.format)
+		die()
+	}
+
+}

cmd/control_pprof.go

@@ -1,4 +1,4 @@
-package daemon
+package cmd
 
 import (
 	"net/http"
@@ -7,16 +7,11 @@ import (
 	"context"
 	"net"
 	"net/http/pprof"
-
-	"github.com/prometheus/client_golang/prometheus/promhttp"
-	"golang.org/x/net/websocket"
-
-	"github.com/zrepl/zrepl/internal/daemon/job"
-	"github.com/zrepl/zrepl/internal/daemon/logging/trace"
 )
 
-type pprofServer struct {
+type PProfServer struct {
 	cc       chan PprofServerControlMsg
+	state    PprofServerControlMsg
 	listener net.Listener
 }
 
@@ -27,9 +22,9 @@ type PprofServerControlMsg struct {
 	HttpListenAddress string
 }
 
-func NewPProfServer(ctx context.Context) *pprofServer {
+func NewPProfServer(ctx context.Context) *PProfServer {
 
-	s := &pprofServer{
+	s := &PProfServer{
 		cc: make(chan PprofServerControlMsg),
 	}
 
@@ -37,7 +32,7 @@ func NewPProfServer(ctx context.Context) *pprofServer {
 	return s
 }
 
-func (s *pprofServer) controlLoop(ctx context.Context) {
+func (s *PProfServer) controlLoop(ctx context.Context) {
 outer:
 	for {
 
@@ -68,16 +63,7 @@ outer:
 			mux.Handle("/debug/pprof/profile", http.HandlerFunc(pprof.Profile))
 			mux.Handle("/debug/pprof/symbol", http.HandlerFunc(pprof.Symbol))
 			mux.Handle("/debug/pprof/trace", http.HandlerFunc(pprof.Trace))
-			mux.Handle("/metrics", promhttp.Handler())
+			go http.Serve(s.listener, mux)
-			mux.Handle("/debug/zrepl/activity-trace", websocket.Handler(trace.ChrometraceClientWebsocketHandler))
-			go func() {
-				err := http.Serve(s.listener, mux)
-				if ctx.Err() != nil {
-					return
-				} else if err != nil {
-					job.GetLogger(ctx).WithError(err).Error("pprof server serve error")
-				}
-			}()
 			continue
 		}
 
@@ -89,6 +75,6 @@ outer:
 	}
 }
 
-func (s *pprofServer) Control(msg PprofServerControlMsg) {
+func (s *PProfServer) Control(msg PprofServerControlMsg) {
 	s.cc <- msg
 }

cmd/daemon.go

@@ -0,0 +1,394 @@
+package cmd
+
+import (
+	"container/list"
+	"context"
+	"fmt"
+	"github.com/spf13/cobra"
+	"github.com/zrepl/zrepl/logger"
+	"io"
+	"os"
+	"os/signal"
+	"strings"
+	"sync"
+	"syscall"
+	"time"
+)
+
+// daemonCmd represents the daemon command
+var daemonCmd = &cobra.Command{
+	Use:   "daemon",
+	Short: "start daemon",
+	Run:   doDaemon,
+}
+
+func init() {
+	RootCmd.AddCommand(daemonCmd)
+}
+
+type Job interface {
+	JobName() string
+	JobStart(ctxt context.Context)
+	JobStatus(ctxt context.Context) (*JobStatus, error)
+}
+
+func doDaemon(cmd *cobra.Command, args []string) {
+
+	conf, err := ParseConfig(rootArgs.configFile)
+	if err != nil {
+		fmt.Fprintf(os.Stderr, "error parsing config: %s\n", err)
+		os.Exit(1)
+	}
+
+	log := logger.NewLogger(conf.Global.logging.Outlets, 1*time.Second)
+
+	log.Info(NewZreplVersionInformation().String())
+	log.Debug("starting daemon")
+	ctx := context.WithValue(context.Background(), contextKeyLog, log)
+	ctx = context.WithValue(ctx, contextKeyLog, log)
+
+	d := NewDaemon(conf)
+	d.Loop(ctx)
+
+}
+
+type contextKey string
+
+const (
+	contextKeyLog    contextKey = contextKey("log")
+	contextKeyDaemon contextKey = contextKey("daemon")
+)
+
+type Daemon struct {
+	conf      *Config
+	startedAt time.Time
+}
+
+func NewDaemon(initialConf *Config) *Daemon {
+	return &Daemon{conf: initialConf}
+}
+
+func (d *Daemon) Loop(ctx context.Context) {
+
+	d.startedAt = time.Now()
+
+	log := ctx.Value(contextKeyLog).(Logger)
+
+	ctx, cancel := context.WithCancel(ctx)
+	ctx = context.WithValue(ctx, contextKeyDaemon, d)
+
+	sigChan := make(chan os.Signal, 1)
+	finishs := make(chan Job)
+
+	signal.Notify(sigChan, syscall.SIGINT, syscall.SIGTERM)
+
+	log.Info("starting jobs from config")
+	i := 0
+	for _, job := range d.conf.Jobs {
+		logger := log.WithField(logJobField, job.JobName())
+		logger.Info("starting")
+		i++
+		jobCtx := context.WithValue(ctx, contextKeyLog, logger)
+		go func(j Job) {
+			j.JobStart(jobCtx)
+			finishs <- j
+		}(job)
+	}
+
+	finishCount := 0
+outer:
+	for {
+		select {
+		case <-finishs:
+			finishCount++
+			if finishCount == len(d.conf.Jobs) {
+				log.Info("all jobs finished")
+				break outer
+			}
+
+		case sig := <-sigChan:
+			log.WithField("signal", sig).Info("received signal")
+			log.Info("cancelling all jobs")
+			cancel()
+		}
+	}
+
+	signal.Stop(sigChan)
+	cancel() // make go vet happy
+
+	log.Info("exiting")
+
+}
+
+// Representation of a Job's status that is composed of Tasks
+type JobStatus struct {
+	// Statuses of all tasks of this job
+	Tasks []*TaskStatus
+	// Error != "" if JobStatus() returned an error
+	JobStatusError string
+}
+
+// Representation of a Daemon's status that is composed of Jobs
+type DaemonStatus struct {
+	StartedAt time.Time
+	Jobs      map[string]*JobStatus
+}
+
+func (d *Daemon) Status() (s *DaemonStatus) {
+
+	s = &DaemonStatus{}
+	s.StartedAt = d.startedAt
+
+	s.Jobs = make(map[string]*JobStatus, len(d.conf.Jobs))
+
+	for name, j := range d.conf.Jobs {
+		status, err := j.JobStatus(context.TODO())
+		if err != nil {
+			s.Jobs[name] = &JobStatus{nil, err.Error()}
+			continue
+		}
+		s.Jobs[name] = status
+	}
+
+	return
+}
+
+// Representation of a Task's status
+type TaskStatus struct {
+	Name string
+	// Whether the task is idle.
+	Idle bool
+	// The stack of activities the task is currently executing.
+	// The first element is the root activity and equal to Name.
+	ActivityStack []string
+	// Number of bytes received by the task since it last left idle state.
+	ProgressRx int64
+	// Number of bytes sent by the task since it last left idle state.
+	ProgressTx int64
+	// Log entries emitted by the task since it last left idle state.
+	// Only contains the log entries emitted through the task's logger
+	// (provided by Task.Log()).
+	LogEntries []logger.Entry
+	// The maximum log level of LogEntries.
+	// Only valid if len(LogEntries) > 0.
+	MaxLogLevel logger.Level
+	// Last time something about the Task changed
+	LastUpdate time.Time
+}
+
+// An instance of Task tracks  a single thread of activity that is part of a Job.
+type Task struct {
+	// Stack of activities the task is currently in
+	// Members are instances of taskActivity
+	activities *list.List
+	// Last time activities was changed (not the activities inside, the list)
+	activitiesLastUpdate time.Time
+	// Protects Task members from modification
+	rwl sync.RWMutex
+}
+
+// Structure that describes the progress a Task has made
+type taskProgress struct {
+	rx         int64
+	tx         int64
+	lastUpdate time.Time
+	logEntries []logger.Entry
+	mtx        sync.RWMutex
+}
+
+func newTaskProgress() (p *taskProgress) {
+	return &taskProgress{
+		logEntries: make([]logger.Entry, 0),
+	}
+}
+
+func (p *taskProgress) UpdateIO(drx, dtx int64) {
+	p.mtx.Lock()
+	defer p.mtx.Unlock()
+	p.rx += drx
+	p.tx += dtx
+	p.lastUpdate = time.Now()
+}
+
+func (p *taskProgress) UpdateLogEntry(entry logger.Entry) {
+	p.mtx.Lock()
+	defer p.mtx.Unlock()
+	// FIXME: ensure maximum size (issue #48)
+	p.logEntries = append(p.logEntries, entry)
+	p.lastUpdate = time.Now()
+}
+
+func (p *taskProgress) DeepCopy() (out taskProgress) {
+	p.mtx.RLock()
+	defer p.mtx.RUnlock()
+	out.rx, out.tx = p.rx, p.tx
+	out.lastUpdate = p.lastUpdate
+	out.logEntries = make([]logger.Entry, len(p.logEntries))
+	for i := range p.logEntries {
+		out.logEntries[i] = p.logEntries[i]
+	}
+	return
+}
+
+// returns a copy of this taskProgress, the mutex carries no semantic value
+func (p *taskProgress) Read() (out taskProgress) {
+	p.mtx.RLock()
+	defer p.mtx.RUnlock()
+	return p.DeepCopy()
+}
+
+// Element of a Task's activity stack
+type taskActivity struct {
+	name   string
+	idle   bool
+	logger *logger.Logger
+	// The progress of the task that is updated by UpdateIO() and UpdateLogEntry()
+	//
+	// Progress happens on a task-level and is thus global to the task.
+	// That's why progress is just a pointer to the current taskProgress:
+	// we reset progress when leaving the idle root activity
+	progress *taskProgress
+}
+
+func NewTask(name string, lg *logger.Logger) *Task {
+	t := &Task{
+		activities: list.New(),
+	}
+	rootLogger := lg.ReplaceField(logTaskField, name).
+		WithOutlet(t, logger.Debug)
+	rootAct := &taskActivity{name, true, rootLogger, newTaskProgress()}
+	t.activities.PushFront(rootAct)
+	return t
+}
+
+// callers must hold t.rwl
+func (t *Task) cur() *taskActivity {
+	return t.activities.Front().Value.(*taskActivity)
+}
+
+// buildActivityStack returns the stack of activity names
+// t.rwl must be held, but the slice can be returned since strings are immutable
+func (t *Task) buildActivityStack() []string {
+	comps := make([]string, 0, t.activities.Len())
+	for e := t.activities.Back(); e != nil; e = e.Prev() {
+		act := e.Value.(*taskActivity)
+		comps = append(comps, act.name)
+	}
+	return comps
+}
+
+// Start a sub-activity.
+// Must always be matched with a call to t.Finish()
+// --- consider using defer for this purpose.
+func (t *Task) Enter(activity string) {
+	t.rwl.Lock()
+	defer t.rwl.Unlock()
+
+	prev := t.cur()
+	if prev.idle {
+		// reset progress when leaving idle task
+		// we leave the old progress dangling to have the user not worry about
+		prev.progress = newTaskProgress()
+	}
+	act := &taskActivity{activity, false, nil, prev.progress}
+	t.activities.PushFront(act)
+	stack := t.buildActivityStack()
+	activityField := strings.Join(stack, ".")
+	act.logger = prev.logger.ReplaceField(logTaskField, activityField)
+
+	t.activitiesLastUpdate = time.Now()
+}
+
+func (t *Task) UpdateProgress(dtx, drx int64) {
+	t.rwl.RLock()
+	p := t.cur().progress // protected by own rwlock
+	t.rwl.RUnlock()
+	p.UpdateIO(dtx, drx)
+}
+
+// Returns a wrapper io.Reader that updates this task's _current_ progress value.
+// Progress updates after this task resets its progress value are discarded.
+func (t *Task) ProgressUpdater(r io.Reader) *IOProgressUpdater {
+	t.rwl.RLock()
+	defer t.rwl.RUnlock()
+	return &IOProgressUpdater{r, t.cur().progress}
+}
+
+func (t *Task) Status() *TaskStatus {
+	t.rwl.RLock()
+	defer t.rwl.RUnlock()
+	// NOTE
+	// do not return any state in TaskStatus that is protected by t.rwl
+
+	cur := t.cur()
+	stack := t.buildActivityStack()
+	prog := cur.progress.Read()
+
+	var maxLevel logger.Level
+	for _, entry := range prog.logEntries {
+		if maxLevel < entry.Level {
+			maxLevel = entry.Level
+		}
+	}
+
+	lastUpdate := prog.lastUpdate
+	if lastUpdate.Before(t.activitiesLastUpdate) {
+		lastUpdate = t.activitiesLastUpdate
+	}
+
+	s := &TaskStatus{
+		Name:          stack[0],
+		ActivityStack: stack,
+		Idle:          cur.idle,
+		ProgressRx:    prog.rx,
+		ProgressTx:    prog.tx,
+		LogEntries:    prog.logEntries,
+		MaxLogLevel:   maxLevel,
+		LastUpdate:    lastUpdate,
+	}
+
+	return s
+}
+
+// Finish a sub-activity.
+// Corresponds to a preceding call to t.Enter()
+func (t *Task) Finish() {
+	t.rwl.Lock()
+	defer t.rwl.Unlock()
+	top := t.activities.Front()
+	if top.Next() == nil {
+		return // cannot remove root activity
+	}
+	t.activities.Remove(top)
+	t.activitiesLastUpdate = time.Now()
+
+}
+
+// Returns a logger derived from the logger passed to the constructor function.
+// The logger's task field contains the current activity stack joined by '.'.
+func (t *Task) Log() *logger.Logger {
+	t.rwl.RLock()
+	defer t.rwl.RUnlock()
+	// FIXME should influence TaskStatus's LastUpdate field
+	return t.cur().logger
+}
+
+// implement logger.Outlet interface
+func (t *Task) WriteEntry(entry logger.Entry) error {
+	t.rwl.RLock()
+	defer t.rwl.RUnlock()
+	t.cur().progress.UpdateLogEntry(entry)
+	return nil
+}
+
+type IOProgressUpdater struct {
+	r io.Reader
+	p *taskProgress
+}
+
+func (u *IOProgressUpdater) Read(p []byte) (n int, err error) {
+	n, err = u.r.Read(p)
+	u.p.UpdateIO(int64(n), 0)
+	return
+
+}

cmd/handler.go

@@ -0,0 +1,181 @@
+package cmd
+
+import (
+	"fmt"
+	"io"
+
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/rpc"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+type DatasetMapping interface {
+	Map(source *zfs.DatasetPath) (target *zfs.DatasetPath, err error)
+}
+
+type FilesystemRequest struct {
+	Roots []string // may be nil, indicating interest in all filesystems
+}
+
+type FilesystemVersionsRequest struct {
+	Filesystem *zfs.DatasetPath
+}
+
+type InitialTransferRequest struct {
+	Filesystem        *zfs.DatasetPath
+	FilesystemVersion zfs.FilesystemVersion
+}
+
+type IncrementalTransferRequest struct {
+	Filesystem *zfs.DatasetPath
+	From       zfs.FilesystemVersion
+	To         zfs.FilesystemVersion
+}
+
+type Handler struct {
+	logger Logger
+	dsf    zfs.DatasetFilter
+	fsvf   zfs.FilesystemVersionFilter
+}
+
+func NewHandler(logger Logger, dsfilter zfs.DatasetFilter, snapfilter zfs.FilesystemVersionFilter) (h Handler) {
+	return Handler{logger, dsfilter, snapfilter}
+}
+
+func registerEndpoints(server rpc.RPCServer, handler Handler) (err error) {
+	err = server.RegisterEndpoint("FilesystemRequest", handler.HandleFilesystemRequest)
+	if err != nil {
+		panic(err)
+	}
+	err = server.RegisterEndpoint("FilesystemVersionsRequest", handler.HandleFilesystemVersionsRequest)
+	if err != nil {
+		panic(err)
+	}
+	err = server.RegisterEndpoint("InitialTransferRequest", handler.HandleInitialTransferRequest)
+	if err != nil {
+		panic(err)
+	}
+	err = server.RegisterEndpoint("IncrementalTransferRequest", handler.HandleIncrementalTransferRequest)
+	if err != nil {
+		panic(err)
+	}
+	return nil
+}
+
+func (h Handler) HandleFilesystemRequest(r *FilesystemRequest, roots *[]*zfs.DatasetPath) (err error) {
+
+	log := h.logger.WithField("endpoint", "FilesystemRequest")
+
+	log.WithField("request", r).Debug("request")
+	log.WithField("dataset_filter", h.dsf).Debug("dsf")
+
+	allowed, err := zfs.ZFSListMapping(h.dsf)
+	if err != nil {
+		log.WithError(err).Error("error listing filesystems")
+		return
+	}
+
+	log.WithField("response", allowed).Debug("response")
+	*roots = allowed
+	return
+}
+
+func (h Handler) HandleFilesystemVersionsRequest(r *FilesystemVersionsRequest, versions *[]zfs.FilesystemVersion) (err error) {
+
+	log := h.logger.WithField("endpoint", "FilesystemVersionsRequest")
+
+	log.WithField("request", r).Debug("request")
+
+	// allowed to request that?
+	if h.pullACLCheck(r.Filesystem, nil); err != nil {
+		log.WithError(err).Warn("pull ACL check failed")
+		return
+	}
+
+	// find our versions
+	vs, err := zfs.ZFSListFilesystemVersions(r.Filesystem, h.fsvf)
+	if err != nil {
+		log.WithError(err).Error("cannot list filesystem versions")
+		return
+	}
+
+	log.WithField("response", vs).Debug("response")
+
+	*versions = vs
+	return
+
+}
+
+func (h Handler) HandleInitialTransferRequest(r *InitialTransferRequest, stream *io.Reader) (err error) {
+
+	log := h.logger.WithField("endpoint", "InitialTransferRequest")
+
+	log.WithField("request", r).Debug("request")
+	if err = h.pullACLCheck(r.Filesystem, &r.FilesystemVersion); err != nil {
+		log.WithError(err).Warn("pull ACL check failed")
+		return
+	}
+
+	log.Debug("invoking zfs send")
+
+	s, err := zfs.ZFSSend(r.Filesystem, &r.FilesystemVersion, nil)
+	if err != nil {
+		log.WithError(err).Error("cannot send filesystem")
+	}
+	*stream = s
+
+	return
+
+}
+
+func (h Handler) HandleIncrementalTransferRequest(r *IncrementalTransferRequest, stream *io.Reader) (err error) {
+
+	log := h.logger.WithField("endpoint", "IncrementalTransferRequest")
+	log.WithField("request", r).Debug("request")
+	if err = h.pullACLCheck(r.Filesystem, &r.From); err != nil {
+		log.WithError(err).Warn("pull ACL check failed")
+		return
+	}
+	if err = h.pullACLCheck(r.Filesystem, &r.To); err != nil {
+		log.WithError(err).Warn("pull ACL check failed")
+		return
+	}
+
+	log.Debug("invoking zfs send")
+
+	s, err := zfs.ZFSSend(r.Filesystem, &r.From, &r.To)
+	if err != nil {
+		log.WithError(err).Error("cannot send filesystem")
+	}
+
+	*stream = s
+	return
+
+}
+
+func (h Handler) pullACLCheck(p *zfs.DatasetPath, v *zfs.FilesystemVersion) (err error) {
+	var fsAllowed, vAllowed bool
+	fsAllowed, err = h.dsf.Filter(p)
+	if err != nil {
+		err = fmt.Errorf("error evaluating ACL: %s", err)
+		return
+	}
+	if !fsAllowed {
+		err = fmt.Errorf("ACL prohibits access to %s", p.ToString())
+		return
+	}
+	if v == nil {
+		return
+	}
+
+	vAllowed, err = h.fsvf.Filter(*v)
+	if err != nil {
+		err = errors.Wrap(err, "error evaluating version filter")
+		return
+	}
+	if !vAllowed {
+		err = fmt.Errorf("ACL prohibits access to %s", v.ToAbsPath(p))
+		return
+	}
+	return
+}

cmd/helpers.go

@@ -1,11 +1,10 @@
-package nethelpers
+package cmd
 
 import (
+	"github.com/pkg/errors"
 	"net"
 	"os"
 	"path/filepath"
-
-	"github.com/pkg/errors"
 )
 
 func PreparePrivateSockpath(sockpath string) error {

cmd/logging_formatters.go

@@ -1,18 +1,20 @@
-package logging
+package cmd
 
 import (
 	"bytes"
 	"encoding/json"
 	"fmt"
-	"time"
-
-	"github.com/fatih/color"
 	"github.com/go-logfmt/logfmt"
 	"github.com/pkg/errors"
-
+	"github.com/zrepl/zrepl/logger"
-	"github.com/zrepl/zrepl/internal/logger"
+	"time"
 )
 
+type EntryFormatter interface {
+	SetMetadataFlags(flags MetadataFlags)
+	Format(e *logger.Entry) ([]byte, error)
+}
+
 const (
 	FieldLevel   = "level"
 	FieldMessage = "msg"
@@ -20,20 +22,13 @@ const (
 )
 
 const (
-	JobField    string = "job"
+	logJobField     string = "job"
-	SubsysField string = "subsystem"
+	logTaskField    string = "task"
-	SpanField   string = "span"
+	logFSField      string = "filesystem"
-)
+	logMapFromField string = "map_from"
-
+	logMapToField   string = "map_to"
-type MetadataFlags int64
+	logIncFromField string = "inc_from"
-
+	logIncToField   string = "inc_to"
-const (
-	MetadataTime MetadataFlags = 1 << iota
-	MetadataLevel
-	MetadataColor
-
-	MetadataNone MetadataFlags = 0
-	MetadataAll  MetadataFlags = ^0
 )
 
 type NoFormatter struct{}
@@ -74,29 +69,39 @@ func (f *HumanFormatter) ignored(field string) bool {
 func (f *HumanFormatter) Format(e *logger.Entry) (out []byte, err error) {
 
 	var line bytes.Buffer
-	col := color.New()
-	if f.metadataFlags&MetadataColor != 0 {
-		col = e.Color()
-	}
 
 	if f.metadataFlags&MetadataTime != 0 {
 		fmt.Fprintf(&line, "%s ", e.Time.Format(HumanFormatterDateFormat))
 	}
 	if f.metadataFlags&MetadataLevel != 0 {
-		fmt.Fprintf(&line, "[%s]", col.Sprint(e.Level.Short()))
+		fmt.Fprintf(&line, "[%s]", e.Level.Short())
 	}
 
-	prefixFields := []string{JobField, SubsysField, SpanField}
+	prefixFields := []string{logJobField, logTaskField, logFSField}
 	prefixed := make(map[string]bool, len(prefixFields)+2)
 	for _, field := range prefixFields {
-		val, ok := e.Fields[field]
+		val, ok := e.Fields[field].(string)
-		if !ok {
+		if ok {
-			continue
-		}
 			if !f.ignored(field) {
-			fmt.Fprintf(&line, "[%s]", col.Sprint(val))
+				fmt.Fprintf(&line, "[%s]", val)
 				prefixed[field] = true
 			}
+		} else {
+			break
+		}
+	}
+	// even more prefix fields
+	mapFrom, mapFromOk := e.Fields[logMapFromField].(string)
+	mapTo, mapToOk := e.Fields[logMapToField].(string)
+	if mapFromOk && mapToOk && !f.ignored(logMapFromField) && !f.ignored(logMapToField) {
+		fmt.Fprintf(&line, "[%s => %s]", mapFrom, mapTo)
+		prefixed[logMapFromField], prefixed[logMapToField] = true, true
+	}
+	incFrom, incFromOk := e.Fields[logIncFromField].(string)
+	incTo, incToOk := e.Fields[logIncToField].(string)
+	if incFromOk && incToOk && !f.ignored(logIncFromField) && !f.ignored(logMapToField) {
+		fmt.Fprintf(&line, "[%s => %s]", incFrom, incTo)
+		prefixed[logIncFromField], prefixed[logIncToField] = true, true
 	}
 
 	if line.Len() > 0 {
@@ -105,11 +110,15 @@ func (f *HumanFormatter) Format(e *logger.Entry) (out []byte, err error) {
 	fmt.Fprint(&line, e.Message)
 
 	if len(e.Fields)-len(prefixed) > 0 {
+		fmt.Fprint(&line, " ")
+		enc := logfmt.NewEncoder(&line)
 		for field, value := range e.Fields {
 			if prefixed[field] || f.ignored(field) {
 				continue
 			}
-			fmt.Fprintf(&line, " %s=%q", col.Sprint(field), fmt.Sprint(value))
+			if err := logfmtTryEncodeKeyval(enc, field, value); err != nil {
+				return nil, err
+			}
 		}
 	}
 
@@ -162,21 +171,15 @@ func (f *LogfmtFormatter) Format(e *logger.Entry) ([]byte, error) {
 	enc := logfmt.NewEncoder(&buf)
 
 	if f.metadataFlags&MetadataTime != 0 {
-		err := enc.EncodeKeyval(FieldTime, e.Time)
+		enc.EncodeKeyval(FieldTime, e.Time)
-		if err != nil {
-			return nil, errors.Wrap(err, "logfmt: encode time")
-		}
 	}
 	if f.metadataFlags&MetadataLevel != 0 {
-		err := enc.EncodeKeyval(FieldLevel, e.Level)
+		enc.EncodeKeyval(FieldLevel, e.Level)
-		if err != nil {
-			return nil, errors.Wrap(err, "logfmt: encode level")
-		}
 	}
 
 	// at least try and put job and task in front
-	prefixed := make(map[string]bool, 3)
+	prefixed := make(map[string]bool, 2)
-	prefix := []string{JobField, SubsysField, SpanField}
+	prefix := []string{logJobField, logTaskField}
 	for _, pf := range prefix {
 		v, ok := e.Fields[pf]
 		if !ok {
@@ -188,10 +191,8 @@ func (f *LogfmtFormatter) Format(e *logger.Entry) ([]byte, error) {
 		prefixed[pf] = true
 	}
 
-	err := enc.EncodeKeyval(FieldMessage, e.Message)
+	enc.EncodeKeyval(FieldMessage, e.Message)
-	if err != nil {
+
-		return nil, errors.Wrap(err, "logfmt: encode message")
-	}
 	for k, v := range e.Fields {
 		if !prefixed[k] {
 			if err := logfmtTryEncodeKeyval(enc, k, v); err != nil {
@@ -210,10 +211,7 @@ func logfmtTryEncodeKeyval(enc *logfmt.Encoder, field, value interface{}) error
 	case nil: // ok
 		return nil
 	case logfmt.ErrUnsupportedValueType:
-		err := enc.EncodeKeyval(field, fmt.Sprintf("<%T>", value))
+		enc.EncodeKeyval(field, fmt.Sprintf("<%T>", value))
-		if err != nil {
-			return errors.Wrap(err, "cannot encode unsupported value type Go type")
-		}
 		return nil
 	}
 	return errors.Wrapf(err, "cannot encode field '%s'", field)

cmd/logging_outlets.go

@@ -1,39 +1,29 @@
-package logging
+package cmd
 
 import (
 	"bytes"
 	"context"
 	"crypto/tls"
+	"github.com/pkg/errors"
+	"github.com/zrepl/zrepl/logger"
 	"io"
 	"log/syslog"
 	"net"
 	"time"
-
-	"github.com/pkg/errors"
-
-	"github.com/zrepl/zrepl/internal/logger"
 )
 
-type EntryFormatter interface {
-	SetMetadataFlags(flags MetadataFlags)
-	Format(e *logger.Entry) ([]byte, error)
-}
-
 type WriterOutlet struct {
-	formatter EntryFormatter
+	Formatter EntryFormatter
-	writer    io.Writer
+	Writer    io.Writer
 }
 
 func (h WriterOutlet) WriteEntry(entry logger.Entry) error {
-	bytes, err := h.formatter.Format(&entry)
+	bytes, err := h.Formatter.Format(&entry)
 	if err != nil {
 		return err
 	}
-	_, err = h.writer.Write(bytes)
+	_, err = h.Writer.Write(bytes)
-	if err != nil {
+	h.Writer.Write([]byte("\n"))
-		return err
-	}
-	_, err = h.writer.Write([]byte("\n"))
 	return err
 }
 
@@ -63,7 +53,7 @@ func NewTCPOutlet(formatter EntryFormatter, network, address string, tlsConfig *
 		return
 	}
 
-	entryChan := make(chan *bytes.Buffer, 1) // allow one message in flight while previous is in io.Copy()
+	entryChan := make(chan *bytes.Buffer, 1) // allow one message in flight while previos is in io.Copy()
 
 	o := &TCPOutlet{
 		formatter: formatter,
@@ -97,10 +87,8 @@ func (h *TCPOutlet) outLoop(retryInterval time.Duration) {
 				conn = nil
 			}
 		}
-		err = conn.SetWriteDeadline(time.Now().Add(retryInterval))
+		conn.SetWriteDeadline(time.Now().Add(retryInterval))
-		if err == nil {
 		_, err = io.Copy(conn, msg)
-		}
 		if err != nil {
 			retry = time.Now().Add(retryInterval)
 			conn.Close()
@@ -131,7 +119,6 @@ func (h *TCPOutlet) WriteEntry(e logger.Entry) error {
 type SyslogOutlet struct {
 	Formatter          EntryFormatter
 	RetryInterval      time.Duration
-	Facility           syslog.Priority
 	writer             *syslog.Writer
 	lastConnectAttempt time.Time
 }
@@ -150,7 +137,7 @@ func (o *SyslogOutlet) WriteEntry(entry logger.Entry) error {
 		if now.Sub(o.lastConnectAttempt) < o.RetryInterval {
 			return nil // not an error toward logger
 		}
-		o.writer, err = syslog.New(o.Facility, "zrepl")
+		o.writer, err = syslog.New(syslog.LOG_LOCAL0, "zrepl")
 		o.lastConnectAttempt = time.Now()
 		if err != nil {
 			o.writer = nil

cmd/main.go

@@ -0,0 +1,47 @@
+// zrepl replicates ZFS filesystems & volumes between pools
+//
+// Code Organization
+//
+// The cmd package uses github.com/spf13/cobra for its CLI.
+//
+// It combines the other packages in the zrepl project to implement zrepl functionality.
+//
+// Each subcommand's code is in the corresponding *.go file.
+// All other *.go files contain code shared by the subcommands.
+package cmd
+
+import (
+	"github.com/spf13/cobra"
+	"github.com/zrepl/zrepl/logger"
+)
+
+//
+//type Logger interface {
+//	Printf(format string, v ...interface{})
+//}
+
+var (
+	zreplVersion string // set by build infrastructure
+)
+
+type Logger = *logger.Logger
+
+var RootCmd = &cobra.Command{
+	Use:   "zrepl",
+	Short: "ZFS dataset replication",
+	Long: `Replicate ZFS filesystems & volumes between pools:
+
+  - push & pull mode
+  - automatic snapshot creation & pruning
+  - local / over the network
+  - ACLs instead of blank SSH access`,
+}
+
+var rootArgs struct {
+	configFile string
+}
+
+func init() {
+	//cobra.OnInitialize(initConfig)
+	RootCmd.PersistentFlags().StringVar(&rootArgs.configFile, "config", "", "config file path")
+}

cmd/prune.go

@@ -0,0 +1,157 @@
+package cmd
+
+import (
+	"context"
+	"fmt"
+	"github.com/zrepl/zrepl/util"
+	"github.com/zrepl/zrepl/zfs"
+	"sync"
+	"time"
+)
+
+type Pruner struct {
+	task           *Task
+	Now            time.Time
+	DryRun         bool
+	DatasetFilter  zfs.DatasetFilter
+	SnapshotPrefix string
+	PrunePolicy    PrunePolicy
+}
+
+type PruneResult struct {
+	Filesystem *zfs.DatasetPath
+	All        []zfs.FilesystemVersion
+	Keep       []zfs.FilesystemVersion
+	Remove     []zfs.FilesystemVersion
+}
+
+// FIXME must not call p.task.Enter because it runs in parallel
+func (p *Pruner) filterFilesystems() (filesystems []*zfs.DatasetPath, stop bool) {
+	filesystems, err := zfs.ZFSListMapping(p.DatasetFilter)
+	if err != nil {
+		p.task.Log().WithError(err).Error("error applying filesystem filter")
+		return nil, true
+	}
+	if len(filesystems) <= 0 {
+		p.task.Log().Info("no filesystems matching filter")
+		return nil, true
+	}
+	return filesystems, false
+}
+
+// FIXME must not call p.task.Enter because it runs in parallel
+func (p *Pruner) filterVersions(fs *zfs.DatasetPath) (fsversions []zfs.FilesystemVersion, stop bool) {
+	log := p.task.Log().WithField(logFSField, fs.ToString())
+
+	filter := NewPrefixFilter(p.SnapshotPrefix)
+	fsversions, err := zfs.ZFSListFilesystemVersions(fs, filter)
+	if err != nil {
+		log.WithError(err).Error("error listing filesytem versions")
+		return nil, true
+	}
+	if len(fsversions) == 0 {
+		log.WithField("prefix", p.SnapshotPrefix).Info("no filesystem versions matching prefix")
+		return nil, true
+	}
+	return fsversions, false
+}
+
+// FIXME must not call p.task.Enter because it runs in parallel
+func (p *Pruner) pruneFilesystem(fs *zfs.DatasetPath, destroySemaphore util.Semaphore) (r PruneResult, valid bool) {
+	log := p.task.Log().WithField(logFSField, fs.ToString())
+
+	fsversions, stop := p.filterVersions(fs)
+	if stop {
+		return
+	}
+
+	keep, remove, err := p.PrunePolicy.Prune(fs, fsversions)
+	if err != nil {
+		log.WithError(err).Error("error evaluating prune policy")
+		return
+	}
+
+	log.WithField("fsversions", fsversions).
+		WithField("keep", keep).
+		WithField("remove", remove).
+		Debug("prune policy debug dump")
+
+	r = PruneResult{fs, fsversions, keep, remove}
+
+	var wg sync.WaitGroup
+	for v := range remove {
+		wg.Add(1)
+		go func(v zfs.FilesystemVersion) {
+			defer wg.Done()
+			// log fields
+			fields := make(map[string]interface{})
+			fields["version"] = v.ToAbsPath(fs)
+			timeSince := v.Creation.Sub(p.Now)
+			fields["age_ns"] = timeSince
+			const day time.Duration = 24 * time.Hour
+			days := timeSince / day
+			remainder := timeSince % day
+			fields["age_str"] = fmt.Sprintf("%dd%s", days, remainder)
+
+			log.WithFields(fields).Info("destroying version")
+			// echo what we'll do and exec zfs destroy if not dry run
+			// TODO special handling for EBUSY (zfs hold)
+			// TODO error handling for clones? just echo to cli, skip over, and exit with non-zero status code (we're idempotent)
+			if !p.DryRun {
+				destroySemaphore.Down()
+				err := zfs.ZFSDestroyFilesystemVersion(fs, v)
+				destroySemaphore.Up()
+				if err != nil {
+					log.WithFields(fields).WithError(err).Error("error destroying version")
+				}
+			}
+		}(remove[v])
+	}
+	wg.Wait()
+
+	return r, true
+}
+
+func (p *Pruner) Run(ctx context.Context) (r []PruneResult, err error) {
+	p.task.Enter("run")
+	defer p.task.Finish()
+
+	if p.DryRun {
+		p.task.Log().Info("doing dry run")
+	}
+
+	filesystems, stop := p.filterFilesystems()
+	if stop {
+		return
+	}
+
+	maxConcurrentDestroy := len(filesystems)
+	p.task.Log().WithField("max_concurrent_destroy", maxConcurrentDestroy).Info("begin concurrent destroy")
+	destroySem := util.NewSemaphore(maxConcurrentDestroy)
+
+	resChan := make(chan PruneResult, len(filesystems))
+	var wg sync.WaitGroup
+	for _, fs := range filesystems {
+		wg.Add(1)
+		go func(fs *zfs.DatasetPath) {
+			defer wg.Done()
+			res, ok := p.pruneFilesystem(fs, destroySem)
+			if ok {
+				resChan <- res
+			}
+
+		}(fs)
+	}
+	wg.Wait()
+	close(resChan)
+
+	p.task.Log().Info("destroys done")
+
+	r = make([]PruneResult, 0, len(filesystems))
+	for res := range resChan {
+		r = append(r, res)
+	}
+
+	return
+
+}

cmd/replication.go

@@ -0,0 +1,323 @@
+package cmd
+
+import (
+	"fmt"
+	"io"
+
+	"bytes"
+	"encoding/json"
+	"github.com/zrepl/zrepl/rpc"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+type localPullACL struct{}
+
+func (a localPullACL) Filter(p *zfs.DatasetPath) (pass bool, err error) {
+	return true, nil
+}
+
+const DEFAULT_INITIAL_REPL_POLICY = InitialReplPolicyMostRecent
+
+type InitialReplPolicy string
+
+const (
+	InitialReplPolicyMostRecent InitialReplPolicy = "most_recent"
+	InitialReplPolicyAll        InitialReplPolicy = "all"
+)
+
+type Puller struct {
+	task              *Task
+	Remote            rpc.RPCClient
+	Mapping           DatasetMapping
+	InitialReplPolicy InitialReplPolicy
+}
+
+type remoteLocalMapping struct {
+	Remote *zfs.DatasetPath
+	Local  *zfs.DatasetPath
+}
+
+func (p *Puller) getRemoteFilesystems() (rfs []*zfs.DatasetPath, ok bool) {
+	p.task.Enter("fetch_remote_fs_list")
+	defer p.task.Finish()
+
+	fsr := FilesystemRequest{}
+	if err := p.Remote.Call("FilesystemRequest", &fsr, &rfs); err != nil {
+		p.task.Log().WithError(err).Error("cannot fetch remote filesystem list")
+		return nil, false
+	}
+	return rfs, true
+}
+
+func (p *Puller) buildReplMapping(remoteFilesystems []*zfs.DatasetPath) (replMapping map[string]remoteLocalMapping, ok bool) {
+	p.task.Enter("build_repl_mapping")
+	defer p.task.Finish()
+
+	replMapping = make(map[string]remoteLocalMapping, len(remoteFilesystems))
+	for fs := range remoteFilesystems {
+		var err error
+		var localFs *zfs.DatasetPath
+		localFs, err = p.Mapping.Map(remoteFilesystems[fs])
+		if err != nil {
+			err := fmt.Errorf("error mapping %s: %s", remoteFilesystems[fs], err)
+			p.task.Log().WithError(err).WithField(logMapFromField, remoteFilesystems[fs]).Error("cannot map")
+			return nil, false
+		}
+		if localFs == nil {
+			continue
+		}
+		p.task.Log().WithField(logMapFromField, remoteFilesystems[fs].ToString()).
+			WithField(logMapToField, localFs.ToString()).Debug("mapping")
+		m := remoteLocalMapping{remoteFilesystems[fs], localFs}
+		replMapping[m.Local.ToString()] = m
+	}
+	return replMapping, true
+}
+
+// returns true if the receiving filesystem (local side) exists and can have child filesystems
+func (p *Puller) replFilesystem(m remoteLocalMapping, localFilesystemState map[string]zfs.FilesystemState) (localExists bool) {
+
+	p.task.Enter("repl_fs")
+	defer p.task.Finish()
+	var err error
+	remote := p.Remote
+
+	log := p.task.Log().
+		WithField(logMapFromField, m.Remote.ToString()).
+		WithField(logMapToField, m.Local.ToString())
+
+	log.Debug("examining local filesystem state")
+	localState, localExists := localFilesystemState[m.Local.ToString()]
+	var versions []zfs.FilesystemVersion
+	switch {
+	case !localExists:
+		log.Info("local filesystem does not exist")
+	case localState.Placeholder:
+		log.Info("local filesystem is marked as placeholder")
+	default:
+		log.Debug("local filesystem exists")
+		log.Debug("requesting local filesystem versions")
+		if versions, err = zfs.ZFSListFilesystemVersions(m.Local, nil); err != nil {
+			log.WithError(err).Error("cannot get local filesystem versions")
+			return false
+		}
+	}
+
+	log.Info("requesting remote filesystem versions")
+	r := FilesystemVersionsRequest{
+		Filesystem: m.Remote,
+	}
+	var theirVersions []zfs.FilesystemVersion
+	if err = remote.Call("FilesystemVersionsRequest", &r, &theirVersions); err != nil {
+		log.WithError(err).Error("cannot get remote filesystem versions")
+		log.Warn("stopping replication for all filesystems mapped as children of receiving filesystem")
+		return false
+	}
+
+	log.Debug("computing diff between remote and local filesystem versions")
+	diff := zfs.MakeFilesystemDiff(versions, theirVersions)
+	log.WithField("diff", diff).Debug("diff between local and remote filesystem")
+
+	if localState.Placeholder && diff.Conflict != zfs.ConflictAllRight {
+		panic("internal inconsistency: local placeholder implies ConflictAllRight")
+	}
+
+	switch diff.Conflict {
+	case zfs.ConflictAllRight:
+
+		log.WithField("replication_policy", p.InitialReplPolicy).Info("performing initial sync, following policy")
+
+		if p.InitialReplPolicy != InitialReplPolicyMostRecent {
+			panic(fmt.Sprintf("policy '%s' not implemented", p.InitialReplPolicy))
+		}
+
+		snapsOnly := make([]zfs.FilesystemVersion, 0, len(diff.MRCAPathRight))
+		for s := range diff.MRCAPathRight {
+			if diff.MRCAPathRight[s].Type == zfs.Snapshot {
+				snapsOnly = append(snapsOnly, diff.MRCAPathRight[s])
+			}
+		}
+
+		if len(snapsOnly) < 1 {
+			log.Warn("cannot perform initial sync: no remote snapshots")
+			return false
+		}
+
+		r := InitialTransferRequest{
+			Filesystem:        m.Remote,
+			FilesystemVersion: snapsOnly[len(snapsOnly)-1],
+		}
+
+		log.WithField("version", r.FilesystemVersion).Debug("requesting snapshot stream")
+
+		var stream io.Reader
+
+		if err = remote.Call("InitialTransferRequest", &r, &stream); err != nil {
+			log.WithError(err).Error("cannot request initial transfer")
+			return false
+		}
+		log.Debug("received initial transfer request response")
+
+		log.Debug("invoke zfs receive")
+		recvArgs := []string{"-u"}
+		if localState.Placeholder {
+			log.Info("receive with forced rollback to replace placeholder filesystem")
+			recvArgs = append(recvArgs, "-F")
+		}
+		progressStream := p.task.ProgressUpdater(stream)
+		if err = zfs.ZFSRecv(m.Local, progressStream, recvArgs...); err != nil {
+			log.WithError(err).Error("cannot receive stream")
+			return false
+		}
+		log.Info("finished receiving stream") // TODO rx delta
+
+		// TODO unify with recv path of ConflictIncremental
+		log.Debug("configuring properties of received filesystem")
+		props := zfs.NewZFSProperties()
+		props.Set("readonly", "on")
+		if err = zfs.ZFSSet(m.Local, props); err != nil {
+			log.WithError(err).Error("cannot set readonly property")
+		}
+
+		log.Info("finished initial transfer")
+		return true
+
+	case zfs.ConflictIncremental:
+
+		if len(diff.IncrementalPath) < 2 {
+			log.Info("remote and local are in sync")
+			return true
+		}
+
+		log.Info("following incremental path from diff")
+		for i := 0; i < len(diff.IncrementalPath)-1; i++ {
+
+			from, to := diff.IncrementalPath[i], diff.IncrementalPath[i+1]
+
+			log, _ := log.WithField(logIncFromField, from.Name).WithField(logIncToField, to.Name), 0
+
+			log.Debug("requesting incremental snapshot stream")
+			r := IncrementalTransferRequest{
+				Filesystem: m.Remote,
+				From:       from,
+				To:         to,
+			}
+			var stream io.Reader
+			if err = remote.Call("IncrementalTransferRequest", &r, &stream); err != nil {
+				log.WithError(err).Error("cannot request incremental snapshot stream")
+				return false
+			}
+
+			log.Debug("invoking zfs receive")
+			progressStream := p.task.ProgressUpdater(stream)
+			// TODO protect against malicious incremental stream
+			if err = zfs.ZFSRecv(m.Local, progressStream); err != nil {
+				log.WithError(err).Error("cannot receive stream")
+				return false
+			}
+			log.Info("finished incremental transfer") // TODO increment rx
+
+		}
+		log.Info("finished following incremental path") // TODO path rx
+		return true
+
+	case zfs.ConflictNoCommonAncestor:
+		fallthrough
+	case zfs.ConflictDiverged:
+
+		var jsonDiff bytes.Buffer
+		if err := json.NewEncoder(&jsonDiff).Encode(diff); err != nil {
+			log.WithError(err).Error("cannot JSON-encode diff")
+			return false
+		}
+
+		var problem, resolution string
+
+		switch diff.Conflict {
+		case zfs.ConflictNoCommonAncestor:
+			problem = "remote and local filesystem have snapshots, but no common one"
+			resolution = "perform manual establish a common snapshot history"
+		case zfs.ConflictDiverged:
+			problem = "remote and local filesystem share a history but have diverged"
+			resolution = "perform manual replication or delete snapshots on the receiving" +
+				"side  to establish an incremental replication parse"
+		}
+
+		log.WithField("diff", jsonDiff.String()).
+			WithField("problem", problem).
+			WithField("resolution", resolution).
+			Error("manual conflict resolution required")
+
+		return false
+
+	}
+
+	panic("should not be reached")
+}
+
+func (p *Puller) Pull() {
+	p.task.Enter("run")
+	defer p.task.Finish()
+
+	p.task.Log().Info("request remote filesystem list")
+	remoteFilesystems, ok := p.getRemoteFilesystems()
+	if !ok {
+		return
+	}
+
+	p.task.Log().Debug("map remote filesystems to local paths and determine order for per-filesystem sync")
+	replMapping, ok := p.buildReplMapping(remoteFilesystems)
+	if !ok {
+
+	}
+
+	p.task.Log().Debug("build cache for already present local filesystem state")
+	p.task.Enter("cache_local_fs_state")
+	localFilesystemState, err := zfs.ZFSListFilesystemState()
+	p.task.Finish()
+	if err != nil {
+		p.task.Log().WithError(err).Error("cannot request local filesystem state")
+		return
+	}
+
+	localTraversal := zfs.NewDatasetPathForest()
+	for _, m := range replMapping {
+		localTraversal.Add(m.Local)
+	}
+
+	p.task.Log().Info("start per-filesystem sync")
+	localTraversal.WalkTopDown(func(v zfs.DatasetPathVisit) bool {
+
+		p.task.Enter("tree_walk")
+		defer p.task.Finish()
+
+		log := p.task.Log().WithField(logFSField, v.Path.ToString())
+
+		if v.FilledIn {
+			if _, exists := localFilesystemState[v.Path.ToString()]; exists {
+				// No need to verify if this is a placeholder or not. It is sufficient
+				// to know we can add child filesystems to it
+				return true
+			}
+			log.Debug("create placeholder filesystem")
+			p.task.Enter("create_placeholder")
+			err = zfs.ZFSCreatePlaceholderFilesystem(v.Path)
+			p.task.Finish()
+			if err != nil {
+				log.Error("cannot create placeholder filesystem")
+				return false
+			}
+			return true
+		}
+
+		m, ok := replMapping[v.Path.ToString()]
+		if !ok {
+			panic("internal inconsistency: replMapping should contain mapping for any path that was not filled in by WalkTopDown()")
+		}
+
+		return p.replFilesystem(m, localFilesystemState)
+	})
+
+	return
+
+}

cmd/sampleconf/localbackup/host1.yml

@@ -0,0 +1,29 @@
+jobs:
+- name: mirror_local
+  type: local
+
+  # snapshot the filesystems matched by the left-hand-side of the mapping
+  # every 10m with zrepl_ as prefix
+  mapping: {
+    "zroot/var/db<":    "storage/backups/local/zroot/var/db",
+    "zroot/usr/home<":  "storage/backups/local/zroot/usr/home",
+    "zroot/usr/home/paranoid":  "!", #don't backup paranoid user
+    "zroot/poudriere/ports<": "!", #don't backup the ports trees
+  }
+  snapshot_prefix: zrepl_
+  interval: 10m
+  initial_repl_policy: most_recent
+
+  # keep one hour of 10m interval snapshots of filesystems matched by
+  # the left-hand-side of the mapping
+  prune_lhs:
+    policy: grid
+    grid: 1x1h(keep=all)
+    keep_bookmarks: all
+
+  # follow a grandfathering scheme for filesystems on the right-hand-side of the mapping
+  prune_rhs:
+    policy: grid
+    grid: 1x1h(keep=all) | 24x1h | 35x1d | 6x30d
+
+

cmd/sampleconf/pullbackup/backuphost.yml

@@ -0,0 +1,27 @@
+jobs:
+- name: fullbackup_prod1
+  type: pull
+  # connect to remote using ssh / stdinserver command
+  connect:
+    type: ssh+stdinserver
+    host: prod1.example.com
+    user: root
+    port: 22
+    identity_file: /root/.ssh/id_ed25519
+
+  # pull (=ask for new snapshots) every 10m, prune afterwards
+  # this will leave us at most 10m behind production
+  interval: 10m
+
+  # pull all offered filesystems to storage/backups/zrepl/pull/prod1.example.com
+  mapping: {
+    "<":"storage/backups/zrepl/pull/prod1.example.com"
+  }
+  initial_repl_policy: most_recent
+
+  # follow a grandfathering scheme for filesystems on the right-hand-side of the mapping
+  snapshot_prefix: zrepl_
+  prune:
+    policy: grid
+    grid: 1x1h(keep=all) | 24x1h | 35x1d | 6x30d
+

cmd/sampleconf/pullbackup/productionhost.yml

@@ -0,0 +1,47 @@
+global:
+  serve:
+    stdinserver:
+      # Directory where AF_UNIX sockets for stdinserver command are placed.
+      #
+      # `zrepl stdinserver CLIENT_IDENTITY`
+      #   * connects to the socket in $sockdir/CLIENT_IDENTITY
+      #   * sends its stdin / stdout file descriptors to the `zrepl daemon` process (see cmsg(3))
+      #   * does nothing more
+      #
+      # This enables a setup where `zrepl daemon` is not directly exposed to the internet
+      # but instead all traffic is tunnelled through SSH.
+      # The server with the source job has an authorized_keys file entry for the public key
+      # used by the corresponding pull job
+      #
+      #   command="/mnt/zrepl stdinserver CLIENT_IDENTITY" ssh-ed25519 AAAAC3NzaC1E... zrepl@pullingserver
+      #
+      # Below is the default value.
+      sockdir: /var/run/zrepl/stdinserver
+
+jobs:
+
+- name: fullbackup_prod1
+  # expect remote to connect via ssh+stdinserver with fullbackup_prod1 as client_identity
+  type: source
+  serve:
+    type: stdinserver # see global.serve.stdinserver for explanation
+    client_identity: fullbackup_prod1
+
+  # snapshot these filesystems every 10m with zrepl_ as prefix
+  filesystems: {
+    "zroot/var/db<": "ok",
+    "zroot/usr/home<": "ok",
+    "zroot/var/tmp": "!", #don't backup /tmp
+  }
+  snapshot_prefix: zrepl_
+  interval: 10m
+
+
+  # keep 1 hour of snapshots (6 at 10m interval)
+  # and  one day of bookmarks in case pull doesn't work (link down, etc)
+  # => keep_bookmarks = 24h / interval = 24h / 10m = 144
+  prune:
+    policy: grid
+    grid: 1x1h(keep=all)
+    keep_bookmarks: 144
+

cmd/sampleconf/pushbackup/backuphost.yml

@@ -0,0 +1,20 @@
+jobs:
+- name: fullbackup_prod1
+
+  # expect remote to connect via ssh+stdinserver with fullbackup_prod1 as client_identity
+  type: push-sink
+  serve:
+    type: stdinserver
+    client_identity: fullbackup_prod1
+
+  # map all pushed datasets to storage/backups/zrepl/sink/prod1.example.com
+  mapping: {
+    "<":"storage/backups/zrepl/sink/prod1.example.com"
+  }
+
+  # follow a grandfathering scheme for filesystems on the right-hand-side of the mapping
+  prune:
+    policy: grid
+    grid: 1x1h(keep=all) | 24x1h | 35x1d | 6x30d
+
+

cmd/sampleconf/pushbackup/productionhost.yml

@@ -0,0 +1,26 @@
+jobs:
+- name: fullbackup_prod1
+
+  # connect to remote using ssh / stdinserver command
+  type: push
+  connect:
+      type: ssh+stdinserver
+      host: prod1.example.com
+      user: root
+      port: 22
+      identity_file: /root/.ssh/id_ed25519
+
+  # snapshot these datsets every 10m with zrepl_ as prefix
+  filesystems: {
+    "zroot/var/db<": "ok",
+    "zroot/usr/home<": "!",
+  }
+  snapshot_prefix: zrepl_
+  interval: 10m
+
+  # keep a one day window 10m interval snapshots in case push doesn't work (link down, etc)
+  # (we cannot keep more than one day because this host will run out of disk space)
+  prune:
+    policy: grid
+    grid: 1x1d(keep=all)
+

cmd/sampleconf/random/debugging.yml

@@ -0,0 +1,33 @@
+global:
+  serve:
+    stdinserver:
+      sockdir: /var/run/zrepl/stdinserver
+
+jobs:
+
+- name: debian2_pull
+  # JOB DEBUGGING OPTIONS
+  # should be equal for all job types, but each job implements the debugging itself
+  # => consult job documentation for supported options
+  debug:
+    conn: # debug the io.ReadWriteCloser connection
+      read_dump: /tmp/connlog_read   # dump results of Read() invocations to this file
+      write_dump: /tmp/connlog_write # dump results of Write() invocations to this file
+    rpc: # debug the RPC protocol implementation
+      log: true # log output from rpc layer to the job log
+
+  # ... just to make the unit tests pass.
+  # check other examples, e.g. localbackup or pullbackup for what the sutff below means
+  type: source
+  serve:
+    type: stdinserver
+    client_identity: debian2
+  filesystems: {
+    "pool1/db<": ok
+  }
+  snapshot_prefix: zrepl_
+  interval: 1s
+  prune:
+    policy: grid
+    grid: 1x10s(keep=all)
+    keep_bookmarks: all

cmd/sampleconf/random/logging.yml

@@ -0,0 +1,24 @@
+global:
+  logging:
+
+    - outlet: stdout
+      level: warn
+      format: human
+
+    - outlet: tcp
+      level: debug
+      format: json
+      net: tcp
+      address: 127.0.0.1:8080
+      retry_interval: 1s
+      tls: # if not specified, use plain TCP
+        ca: sampleconf/random/logging/logserver.crt
+        cert: sampleconf/random/logging/client.crt
+        key: sampleconf/random/logging/client.key
+
+    - outlet: syslog
+      level: debug
+      format: logfmt
+
+
+jobs: []

cmd/sampleconf/random/logging/client.crt

@@ -0,0 +1,19 @@
+-----BEGIN CERTIFICATE-----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+-----END CERTIFICATE-----

cmd/sampleconf/random/logging/client.csr

@@ -0,0 +1,16 @@
+-----BEGIN CERTIFICATE REQUEST-----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=
+-----END CERTIFICATE REQUEST-----

cmd/sampleconf/random/logging/client.key

@@ -0,0 +1,28 @@
+-----BEGIN PRIVATE KEY-----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+-----END PRIVATE KEY-----

cmd/sampleconf/random/logging/logserver.crt

@@ -0,0 +1,21 @@
+-----BEGIN CERTIFICATE-----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=
+-----END CERTIFICATE-----

cmd/sampleconf/random/logging/logserver.key

@@ -0,0 +1,28 @@
+-----BEGIN PRIVATE KEY-----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+-----END PRIVATE KEY-----

cmd/stdinserver.go

@@ -1,51 +1,55 @@
-package client
+package cmd
 
 import (
 	"os"
 
-	"github.com/problame/go-netssh"
-
-	"github.com/zrepl/zrepl/internal/cli"
-	"github.com/zrepl/zrepl/internal/config"
-
 	"context"
-	"errors"
+	"github.com/problame/go-netssh"
+	"github.com/spf13/cobra"
 	"log"
 	"path"
 )
 
-var StdinserverCmd = &cli.Subcommand{
+var StdinserverCmd = &cobra.Command{
 	Use:   "stdinserver CLIENT_IDENTITY",
-	Short: "stdinserver transport mode (started from authorized_keys file as forced command)",
+	Short: "start in stdinserver mode (from authorized_keys file)",
-	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
+	Run:   cmdStdinServer,
-		return runStdinserver(subcommand.Config(), args)
-	},
 }
 
-func runStdinserver(config *config.Config, args []string) error {
+func init() {
+	RootCmd.AddCommand(StdinserverCmd)
+}
+
+func cmdStdinServer(cmd *cobra.Command, args []string) {
 
 	// NOTE: the netssh proxying protocol requires exiting with non-zero status if anything goes wrong
 	defer os.Exit(1)
 
 	log := log.New(os.Stderr, "", log.LUTC|log.Ldate|log.Ltime)
 
+	conf, err := ParseConfig(rootArgs.configFile)
+	if err != nil {
+		log.Printf("error parsing config: %s", err)
+		return
+	}
+
 	if len(args) != 1 || args[0] == "" {
-		err := errors.New("must specify client_identity as positional argument")
+		log.Print("must specify client_identity as positional argument")
-		return err
+		return
 	}
 
 	identity := args[0]
-	unixaddr := path.Join(config.Global.Serve.StdinServer.SockDir, identity)
+	unixaddr := path.Join(conf.Global.Serve.Stdinserver.SockDir, identity)
 
 	log.Printf("proxying client identity '%s' to zrepl daemon '%s'", identity, unixaddr)
 
 	ctx := netssh.ContextWithLog(context.TODO(), log)
 
-	err := netssh.Proxy(ctx, unixaddr)
+	err = netssh.Proxy(ctx, unixaddr)
 	if err == nil {
 		log.Print("proxying finished successfully, exiting with status 0")
 		os.Exit(0)
 	}
 	log.Printf("error proxying: %s", err)
-	return nil
+
 }

cmd/test.go

@@ -0,0 +1,215 @@
+package cmd
+
+import (
+	"os"
+
+	"bytes"
+	"context"
+	"fmt"
+	"sort"
+	"strings"
+
+	"github.com/kr/pretty"
+	"github.com/spf13/cobra"
+	"github.com/zrepl/zrepl/logger"
+	"github.com/zrepl/zrepl/zfs"
+	"time"
+)
+
+var testCmd = &cobra.Command{
+	Use:              "test",
+	Short:            "test configuration",
+	PersistentPreRun: testCmdGlobalInit,
+}
+
+var testCmdGlobal struct {
+	log  Logger
+	conf *Config
+}
+
+var testConfigSyntaxCmd = &cobra.Command{
+	Use:   "config",
+	Short: "parse config file and dump parsed datastructure",
+	Run:   doTestConfig,
+}
+
+var testDatasetMapFilter = &cobra.Command{
+	Use:     "pattern jobname test/zfs/dataset/path",
+	Short:   "test dataset mapping / filter specified in config",
+	Example: ` zrepl test pattern my_pull_job tank/tmp`,
+	Run:     doTestDatasetMapFilter,
+}
+
+var testPrunePolicyArgs struct {
+	side        PrunePolicySide
+	showKept    bool
+	showRemoved bool
+}
+
+var testPrunePolicyCmd = &cobra.Command{
+	Use:   "prune jobname",
+	Short: "do a dry-run of the pruning part of a job",
+	Run:   doTestPrunePolicy,
+}
+
+func init() {
+	RootCmd.AddCommand(testCmd)
+	testCmd.AddCommand(testConfigSyntaxCmd)
+	testCmd.AddCommand(testDatasetMapFilter)
+
+	testPrunePolicyCmd.Flags().VarP(&testPrunePolicyArgs.side, "side", "s", "prune_lhs (left) or prune_rhs (right)")
+	testPrunePolicyCmd.Flags().BoolVar(&testPrunePolicyArgs.showKept, "kept", false, "show kept snapshots")
+	testPrunePolicyCmd.Flags().BoolVar(&testPrunePolicyArgs.showRemoved, "removed", true, "show removed snapshots")
+	testCmd.AddCommand(testPrunePolicyCmd)
+}
+
+func testCmdGlobalInit(cmd *cobra.Command, args []string) {
+
+	out := logger.NewOutlets()
+	out.Add(WriterOutlet{&NoFormatter{}, os.Stdout}, logger.Info)
+	log := logger.NewLogger(out, 1*time.Second)
+	testCmdGlobal.log = log
+
+	var err error
+	if testCmdGlobal.conf, err = ParseConfig(rootArgs.configFile); err != nil {
+		testCmdGlobal.log.Printf("error parsing config file: %s", err)
+		os.Exit(1)
+	}
+
+}
+
+func doTestConfig(cmd *cobra.Command, args []string) {
+
+	log, conf := testCmdGlobal.log, testCmdGlobal.conf
+
+	log.Printf("config ok")
+	log.Printf("%# v", pretty.Formatter(conf))
+	return
+}
+
+func doTestDatasetMapFilter(cmd *cobra.Command, args []string) {
+
+	log, conf := testCmdGlobal.log, testCmdGlobal.conf
+
+	if len(args) != 2 {
+		log.Printf("specify job name as first postitional argument, test input as second")
+		log.Printf(cmd.UsageString())
+		os.Exit(1)
+	}
+	n, i := args[0], args[1]
+
+	jobi, err := conf.LookupJob(n)
+	if err != nil {
+		log.Printf("%s", err)
+		os.Exit(1)
+	}
+
+	var mf *DatasetMapFilter
+	switch j := jobi.(type) {
+	case *PullJob:
+		mf = j.Mapping
+	case *SourceJob:
+		mf = j.Filesystems
+	case *LocalJob:
+		mf = j.Mapping
+	default:
+		panic("incomplete implementation")
+	}
+
+	ip, err := zfs.NewDatasetPath(i)
+	if err != nil {
+		log.Printf("cannot parse test input as ZFS dataset path: %s", err)
+		os.Exit(1)
+	}
+
+	if mf.filterMode {
+		pass, err := mf.Filter(ip)
+		if err != nil {
+			log.Printf("error evaluating filter: %s", err)
+			os.Exit(1)
+		}
+		log.Printf("filter result: %v", pass)
+	} else {
+		res, err := mf.Map(ip)
+		if err != nil {
+			log.Printf("error evaluating mapping: %s", err)
+			os.Exit(1)
+		}
+		toStr := "NO MAPPING"
+		if res != nil {
+			toStr = res.ToString()
+		}
+		log.Printf("%s => %s", ip.ToString(), toStr)
+
+	}
+
+}
+
+func doTestPrunePolicy(cmd *cobra.Command, args []string) {
+
+	log, conf := testCmdGlobal.log, testCmdGlobal.conf
+
+	if cmd.Flags().NArg() != 1 {
+		log.Printf("specify job name as first positional argument")
+		log.Printf(cmd.UsageString())
+		os.Exit(1)
+	}
+
+	jobname := cmd.Flags().Arg(0)
+	jobi, err := conf.LookupJob(jobname)
+	if err != nil {
+		log.Printf("%s", err)
+		os.Exit(1)
+	}
+
+	jobp, ok := jobi.(PruningJob)
+	if !ok {
+		log.Printf("job doesn't do any prunes")
+		os.Exit(0)
+	}
+
+	log.Printf("job dump:\n%s", pretty.Sprint(jobp))
+
+	task := NewTask("", log)
+	pruner, err := jobp.Pruner(task, testPrunePolicyArgs.side, true)
+	if err != nil {
+		log.Printf("cannot create test pruner: %s", err)
+		os.Exit(1)
+	}
+
+	log.Printf("start pruning")
+
+	ctx := context.WithValue(context.Background(), contextKeyLog, log)
+	result, err := pruner.Run(ctx)
+	if err != nil {
+		log.Printf("error running pruner: %s", err)
+		os.Exit(1)
+	}
+
+	sort.Slice(result, func(i, j int) bool {
+		return strings.Compare(result[i].Filesystem.ToString(), result[j].Filesystem.ToString()) == -1
+	})
+
+	var b bytes.Buffer
+	for _, r := range result {
+		fmt.Fprintf(&b, "%s\n", r.Filesystem.ToString())
+
+		if testPrunePolicyArgs.showKept {
+			fmt.Fprintf(&b, "\tkept:\n")
+			for _, v := range r.Keep {
+				fmt.Fprintf(&b, "\t- %s\n", v.Name)
+			}
+		}
+
+		if testPrunePolicyArgs.showRemoved {
+			fmt.Fprintf(&b, "\tremoved:\n")
+			for _, v := range r.Remove {
+				fmt.Fprintf(&b, "\t- %s\n", v.Name)
+			}
+		}
+
+	}
+
+	log.Printf("pruning result:\n%s", b.String())
+
+}

cmd/version.go

@@ -0,0 +1,42 @@
+package cmd
+
+import (
+	"fmt"
+	"github.com/spf13/cobra"
+	"runtime"
+)
+
+var versionCmd = &cobra.Command{
+	Use:   "version",
+	Short: "print version of zrepl binary (for running daemon 'zrepl control version' command)",
+	Run:   doVersion,
+}
+
+func init() {
+	RootCmd.AddCommand(versionCmd)
+}
+
+func doVersion(cmd *cobra.Command, args []string) {
+	fmt.Println(NewZreplVersionInformation().String())
+}
+
+type ZreplVersionInformation struct {
+	Version         string
+	RuntimeGOOS     string
+	RuntimeGOARCH   string
+	RUNTIMECompiler string
+}
+
+func NewZreplVersionInformation() *ZreplVersionInformation {
+	return &ZreplVersionInformation{
+		Version:         zreplVersion,
+		RuntimeGOOS:     runtime.GOOS,
+		RuntimeGOARCH:   runtime.GOARCH,
+		RUNTIMECompiler: runtime.Compiler,
+	}
+}
+
+func (i *ZreplVersionInformation) String() string {
+	return fmt.Sprintf("zrepl version=%s GOOS=%s GOARCH=%s Compiler=%s",
+		i.Version, i.RuntimeGOOS, i.RuntimeGOARCH, i.RUNTIMECompiler)
+}

debian

@@ -1 +0,0 @@
-packaging/deb/debian

dist/openrc/zrepl

@@ -1,23 +0,0 @@
-#!/sbin/openrc-run
-
-command='/usr/local/bin/zrepl'
-command_args='daemon'
-command_background='true'
-pidfile="/run/${RC_SVCNAME}.pid"
-output_log="/var/log/${RC_SVCNAME}.log"
-error_log="/var/log/${RC_SVCNAME}.log"
-
-zrepl_runtime_dir='/var/run/zrepl'
-
-start() {
-	mkdir -p "$zrepl_runtime_dir/stdinserver"
-	chmod -R 0700 "$zrepl_runtime_dir"
-	default_start
-}
-
-stop() {
-	rm -rf "$zrepl_runtime_dir"
-	default_stop
-}
-
-# vi: noet sw=8 sts=0

dist/systemd/zrepl.service

@@ -1,16 +0,0 @@
-[Unit]
-Description=zrepl daemon
-Documentation=https://zrepl.github.io
-
-[Service]
-Type=simple
-ExecStartPre=/usr/local/bin/zrepl --config /etc/zrepl/zrepl.yml configcheck
-ExecStart=/usr/local/bin/zrepl --config /etc/zrepl/zrepl.yml daemon
-RuntimeDirectory=zrepl zrepl/stdinserver
-RuntimeDirectoryMode=0700
-
-# Make Go produce coredumps
-Environment=GOTRACEBACK='crash'
-
-[Install]
-WantedBy=multi-user.target

docs/.python-version

@@ -1 +0,0 @@
-3.14

docs/Makefile

@@ -10,11 +10,11 @@ BUILDDIR      = _build
 
 # Put it first so that "make" without argument is like "make help".
 help:
-	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" -c sphinxconf $(SPHINXOPTS) $(O)
 
 .PHONY: help Makefile
 
 # Catch-all target: route all unknown targets to Sphinx using the new
 # "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
 %: Makefile
-	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" -c sphinxconf $(SPHINXOPTS) $(O)

docs/_static/banner.css

@@ -1,43 +0,0 @@
-/* https://github.com/sphinx-contrib/sphinxcontrib-versioning/blob/0b56210959c6b21bbb730072e81876491b4e1371/sphinxcontrib/versioning/_static/banner.css */
-
-.scv-banner {
-    padding: 3px;
-    border-radius: 2px;
-    font-size: 80%;
-    text-align: center;
-    color: white;
-    background: #d40 linear-gradient(-45deg,
-        rgba(255, 255, 255, 0.2) 0%,
-        rgba(255, 255, 255, 0.2) 25%,
-        transparent 25%,
-        transparent 50%,
-        rgba(255, 255, 255, 0.2) 50%,
-        rgba(255, 255, 255, 0.2) 75%,
-        transparent 75%,
-        transparent
-    );
-    background-size: 28px 28px;
-}
-.scv-banner > a {
-    color: white;
-}
-
-
-.scv-sphinx_rtd_theme {
-    background-color: #2980B9;
-}
-
-
-.scv-bizstyle {
-    background-color: #336699;
-}
-
-
-.scv-classic {
-    text-align: center !important;
-}
-
-
-.scv-traditional {
-    text-align: center !important;
-}

docs/_templates/page.html

@@ -1 +0,0 @@
-{% extends "!page.html" %}

docs/_templates/versions.html

@@ -1,24 +0,0 @@
-<div class="rst-versions" data-toggle="rst-versions" role="note" aria-label="versions">
-  <span class="rst-current-version" data-toggle="rst-current-version">
-    <span class="fa fa-book"> Other Versions</span>
-    v: latest
-    <span class="fa fa-caret-down"></span>
-  </span>
-  <div class="rst-other-versions">
-    <dl>
-      <dt>Releases</dt>
-      <dd><a href="https://github.com/zrepl/zrepl/releases/tag/v0.6.1">v0.6.1</a></dd>
-      <dd><a href="https://github.com/zrepl/zrepl/releases/tag/v0.5.0">v0.5.0</a></dd>
-      <dd><a href="https://github.com/zrepl/zrepl/releases/tag/v0.4.0">v0.4.0</a></dd>
-      <dd><a href="https://github.com/zrepl/zrepl/releases">All releases...</a></dd>
-    </dl>
-    <dl>
-      <dt>Note</dt>
-      <dd style="white-space: normal;">
-        Release-specific docs are in the
-        <span style="font-family: monospace;">zrepl-noarch.tar</span>
-        asset on each GitHub release.
-      </dd>
-    </dl>
-  </div>
-</div>

docs/changelog.rst

@@ -1,389 +1,68 @@
-.. include:: global.rst.inc
+.. |break_config| replace:: **[BREAK]**
-
-.. |break_config| replace:: **[CONFIG]**
 .. |break| replace:: **[BREAK]**
 .. |bugfix| replace:: [BUG]
-.. |docs| replace:: [DOCS]
 .. |feature| replace:: [FEATURE]
-.. |mig| replace:: **[MIGRATION]**
-.. |maint| replace:: [MAINT]
-
-.. _changelog:
 
 Changelog
 =========
 
-0.7.0
+The changelog summarized bugfixes that are deemed relevant for users.
+Developers should consult the git commit log or GitHub issue tracker.
+
+0.0.3
 -----
 
-* |feature| Config file inclusion using ``include`` directive.
+* |break_config| |feature| :issue:`34`: automatic bookmarking of snapshots
-  This allows distributing zrepl job definitions across multiple YAML files in a ``conf.d`` style directory.
-  (:commit:`4d6583e`, thanks, `@ZeyadTamimi <https://github.com/zeyadtamimi>`_).
-* |feature| Add configurable timezone for snapshot name timestamps. The default remains UTC. (:commit:`b9b9ad1`, thanks, `@mjasnik <https://github.com/mjasnik>`_).
-* |feature| Improve Grafana dashboard with additional panels and refinements (:commit:`3f75127`, thanks, `@deajan <https://github.com/deajan>`_).
-* |bugfix| Fix replication of placeholder filesystems (:commit:`5615f49`).
-* |bugfix| Avoid using the word "error" in info-level log lines about ZFS command execution (:commit:`d3b2295`, thanks, `@bakhtiyarneyman <https://github.com/bakhtiyarneyman>`_).
-* |bugfix| Detect duplicate and internal job names at config parse time.
-  zrepl will now refuse to load a config with duplicate or internally reserved job names, instead of failing later at daemon startup.
-  (:commit:`860a9be`).
-* |docs| Add installation instructions for openSUSE (:commit:`40c4827`, thanks, `@findesgh <https://github.com/findesgh>`_).
-* |docs| Improve documentation on ``send_properties`` configuration and NFS/SMB considerations (:commit:`b5d8538`, thanks, `@Malvineous <https://github.com/Malvineous>`_).
-* |docs| Warn more prominently about the risks of the ``not_replicated`` keep rule (:commit:`affe00a`, thanks, `@wxiaoguang <https://github.com/wxiaoguang>`_).
-* |docs| Improve TLS/EasyRSA setup instructions (:commit:`e524b60`, thanks, `@alorimer <https://github.com/alorimer>`_).
-* |docs| ``zrepl.github.io``: auto-publish from ``master`` branch, retire ``stable`` branch (:commit:`4f950bb`).
-* |maint| ``zrepl status``: switch from the unmaintained ``cview`` fork back to the actively maintained ``tview`` library (:commit:`d7ede3f`).
-* |maint| Update to Go 1.25 toolchain with Go 1.24 language level & Go dependencies
-* |maint| Debian packages: proper bash completion placement (:commit:`2923009`)
-* |maint| RPM packages: AlmaLinux 8 for RPM builds (:commit:`27f4ad1`).
-* |maint| Improve build system & CI setup
 
-.. NOTE::
+  * Snapshots are automatically bookmarked and pruning of bookmarks **must** be configured.
-  |  zrepl is a spare-time project primarily developed by `Christian Schwarz <https://cschwarz.com>`_.
+  * This breaks existing configuration: ``grid`` :ref:`prune policy <prune-retention-grid>`  specifications require the new ``keep_bookmarks`` parameter.
-  |  You can support maintenance and feature development through one of the following services:
+  * Make sure to understand the meaning bookmarks have for :ref:`maximum replication downtime <replication-downtime>`.
-  |  |Donate via Patreon| |Donate via GitHub Sponsors| |Donate via Liberapay| |Donate via PayPal|
+  * Example: :sampleconf:`pullbackup/productionhost.yml`
-  |  Note that PayPal processing fees are relatively high for small donations.
-  |  For SEPA wire transfer and **commercial support**, please `contact Christian directly <https://cschwarz.com>`_.
 
-0.6.1
+* |break| :commit:`ccd062e`: both sides of a replication setup must be updated and restarted. Otherwise the connecting side will hang and not time out.
+* |break_config| :commit:`2bfcfa5`: first outlet in ``global.logging`` is now used for logging meta-errors, for example problems encountered when writing to other outlets.
+* |feature| :issue:`10`: ``zrepl control status`` subcommand
+
+  * Allows inspection of job activity per task and their log output at runtime.
+  * Supports ``--format raw`` option for JSON output, usable for monitoring from scripts.
+
+* |feature| :commit:`d7f3fb9`: subcommand bash completions
+
+  * Package maintainers should install this as appropriate.
+
+* |bugfix| :issue:`61`: fix excessive memory usage
+* |bugfix| :issue:`8` and :issue:`56`: ``ssh+stdinserver`` transport properly reaps SSH child processes
+* |bugfix| :commit:`cef63ac`: ``human`` format now prints non-string values correctly
+* |bugfix| :issue:`26`: slow TCP outlets no longer block the daemon
+
+0.0.2
 -----
 
-* |feature| add metric to detect filesystems rules that don't match any local dataset (thanks, `@gmekicaxcient <https://github.com/gmekicaxcient>`_).
+* |break_config| :commit:`b95260f`: ``global.logging`` is no longer a dictionary but a list
-* |bugfix| ``zrepl status``: hide progress bar once all filesystems reach terminal state (thanks, `@0x3333 <https://github.com/0x3333>`_).
-* |bugfix| handling of tenative cursor presence if protection strategy doesn't use it (:issue:`714`).
-* |docs| address setup with two or more external disks (thanks, `@se-jaeger <https://github.com/se-jaeger>`_).
-* |docs| document ``replication`` and ``conflict_resolution`` options (thanks, `@InsanePrawn <https://github.com/InsanePrawn>`_).
-* |docs| docs: talks: add note on keep_bookmarks option (thanks, `@skirmess <https://github.com/skirmess>`_).
-* |maint| dist: add openrc service file (thanks, `@gramosg <https://github.com/gramosg>`_).
-* |maint| grafana: update dashboard to Grafana 9.3.6.
-* |maint| run platform tests as part of CI.
-* |maint| build: upgrade to Go 1.21 and update golangci-lint; minimum Go version for builds is now 1.20
 
-0.6
+* |break_config| :commit:`3e647c1`: ``source`` job field ``datasets`` renamed to ``filesystems``
----
 
-* |feature| :ref:`Schedule-based snapshotting<job-snapshotting--cron>` using ``cron`` syntax instead of an interval.
+  * **NOTE**: zrepl will parse missing ``filesystems`` field as an empty filter,
-* |feature| Configurable initial replication policy.
+    i.e. no filesystems are presented to the other side.
-  When a filesystem is first replicated to a receiver, this control whether just the newest
-  snapshot will be replicated vs. all existing snapshots. Learn more :ref:`in the docs <conflict_resolution-initial_replication>`.
-* |feature| Configurable timestamp format for snapshot names via :ref:`timestamp_format<job-snapshotting-timestamp_format>`
-  (Thanks, `@ydylla <https://github.com/ydylla>`_).
-* |feature| Add ``ZREPL_DESTROY_MAX_BATCH_SIZE`` env var (default 0=unlimited)
-  (Thanks, `@3nprob <https://github.com/3nprob>`_).
-* |feature| Add ``zrepl configcheck --skip-cert-check`` flag (Thanks, `@cole-h <https://github.com/cole-h>`_).
-* |bugfix| Fix resuming from interrupted replications that use ``send.raw`` on unencrypted datasets.
 
-  * The send options introduced in zrepl 0.4 allow users to specify additional zfs send flags for zrepl to use.
+* |bugfix| :commit:`72d2885` fix aliasing bug with root `<` subtree wildcard
-    Before this fix, when setting ``send.raw=true`` on a job that replicates unencrypted datasets,
-    zrepl would not allow an interrupted replication to resume.
-    The reason were overly cautious checks to support the ``send.encrypted`` option.
-  * This bugfix removes these checks from the replication planner.
-    This makes ``send.encrypted`` a sender-side-only concern, much like all other ``send.*`` flags.
-  * However, this means that the ``zrepl status`` UI no longer indicates whether a replication step uses encrypted sends or not.
-    The setting is still effective though.
 
-* |break| convert Prometheus metric ``zrepl_version_daemon`` to ``zrepl_start_time`` metric
+  * Filesystems paths with final match at blank `s` subtree wildcard are now appended to the target path
+  * Non-root subtree wildcards, e.g. `zroot/foo/bar<` still map directrly onto the target path
 
-  * The metric still reports the zrepl version in a label.
+* Support days (``d``) and weeks (``w``) in durations
-    But the metric *value* is now the Unix timestamp at the time the daemon was started.
-    The Grafana dashboard in :repomasterlink:`dist/grafana` has been updated.
 
-* |bugfix| transient zrepl status error: ``Post "http://unix/status": EOF``
+* Docs
-* |bugfix| don't treat receive-side bookmarks as a replication conflict.
-  This facilitates chaining of replication jobs. See :issue:`490`.
-* |bugfix| workaround for Go/gRPC problem on Illumos where zrepl would
-  crash when using the ``local`` transport type (:issue:`598`).
-* |bugfix| fix active child tasks panic that cold occur during replication plannig (:issue:`193abbe`)
-* |bugfix| ``zrepl status`` off-by-one error in display of completed step count (:commit:`ce6701f`)
-* |bugfix| Allow using day & week units for ``snapshotting.interval`` (:commit:`ffb1d89`)
-* |docs| ``docs/overview`` improvements (Thanks, `@jtagcat <https://github.com/jtagcat>`_).
-* |maint| Update to Go 1.19.
 
-0.5
+  * Ditch Hugo, move to Python Sphinx
----
+  * Improve & simplify tutorial (single SSH key per installation)
+  * Document pruning policies
+  * Document job types
+  * Document logging
+  * Start updating implementation overview
 
-* |feature| :ref:`Bandwidth limiting <job-send-recv-options--bandwidth-limit>` (Thanks, Prominic.NET, Inc.)
-* |feature| zrepl status: use a ``*`` to indicate which filesystem is currently replicating
-* |feature| include daemon environment variables in zrepl status (currently only in ``--raw``)
-* |bugfix| **fix encrypt-on-receive + placeholders use case** (:issue:`504`)
 
-  * Before this fix, **plain sends** to a receiver with an encrypted ``root_fs`` **could be received unencrypted** if zrepl needed to create placeholders on the receiver.
+0.0.1
-  * Existing zrepl users should :ref:`read the docs <job-recv-options--placeholder>` and check ``zfs get -r encryption,zrepl:placeholder PATH_TO_ROOTFS`` on the receiver.
-  * Thanks to `@mologie <https://github.com/mologie>`_ and `@razielgn <https://github.com/razielgn>`_ for reporting and testing!
-
-* |bugfix| Rename mis-spelled :ref:`send option <job-send-options>` ``embbeded_data`` to ``embedded_data``.
-* |bugfix| zrepl status: replication step numbers should start at 1
-* |bugfix| incorrect bandwidth averaging in ``zrepl status``.
-* |bugfix| FreeBSD with OpenZFS 2.0: zrepl would wait indefinitely for zfs send to exit on timeouts.
-* |bugfix| fix ``strconv.ParseInt: value out of range`` bug (and use the control RPCs).
-* |docs| improve description of multiple pruning rules.
-* |docs| document :ref:`platform tests <usage-platform-tests>`.
-* |docs| quickstart: make users aware that prune rules apply to all snapshots.
-* |maint| some platformtests were broken.
-* |maint| FreeBSD: release armv7 and arm64 binaries.
-* |maint| apt repo: update instructions due to ``apt-key`` deprecation.
-
-Note to all users: please read up on the following OpenZFS bugs, as you might be affected:
-
-* `ZFS send/recv with ashift 9->12 leads to data corruption <https://github.com/openzfs/zfs/issues/12762>`_.
-* Various bugs with encrypted send/recv (`Leadership meeting notes <https://openzfs.topicbox.com/groups/developer/T24bdaa2886c6cbf5-Mc039a11c3f1507ea0664817b/december-openzfs-leadership-meeting>`_)
-
-Finally, I'd like to point you to the `GitHub discussion <https://github.com/zrepl/zrepl/discussions/547>`_ about which bugfixes and features should be prioritized in zrepl 0.6 and beyond!
-
-0.4.0
 -----
 
-* |feature| support setting zfs send / recv flags in the config (send: ``-wLcepbS`` , recv: ``-ox`` ).
+* Initial release
-  Config docs :ref:`here <job-send-options>` and :ref:`here <job-recv-options>` .
-* |feature| parallel replication is now configurable (disabled by default, :ref:`config docs here <replication-option-concurrency>` ).
-* |feature| New ``zrepl status`` UI:
-
-  * Interactive job selection.
-  * Interactively ``zrepl signal`` jobs.
-  * Filter filesystems in the job view by name.
-  * An approximation of the old UI is still included as `--mode legacy` but will be removed in a future release of zrepl.
-
-* |bugfix| Actually use concurrency when listing zrepl abstractions & doing size estimation.
-  These operations were accidentally made sequential in zrepl 0.3.
-* |bugfix| Job hang-up during second replication attempt.
-* |bugfix| Data races conditions in the dataconn rpc stack.
-* |maint| Update to protobuf v1.25 and grpc 1.35.
-
-For users who skipped the 0.3.1 update: please make sure your pruning grid config is correct.
-The following bugfix in 0.3.1 :issue:`caused problems for some users <400>`:
-
-* |bugfix| pruning: ``grid``:  add all snapshots that do not match the regex to the rule's destroy list.
-
-0.3.1
------
-
-Mostly a bugfix release for :ref:`zrepl 0.3 <release-0.3>`.
-
-* |feature| pruning: add optional ``regex`` field to ``last_n`` rule
-* |docs| pruning: ``grid`` : improve documentation and add an example
-* |bugfix| pruning: ``grid``:  add all snapshots that do not match the regex to the rule's destroy list.
-  This brings the implementation in line with the docs.
-* |bugfix| ``easyrsa`` script in docs
-* |bugfix| platformtest: fix skipping encryption-only tests on systems that don't support encryption
-* |bugfix| replication: report AttemptDone if no filesystems are replicated
-* |feature| status + replication: warning if replication succeeeded without any filesystem being replicated
-* |docs| update multi-job & multi-host setup section
-* RPM Packaging
-* CI infrastructure rework
-* Continuous deployment of that new `stable` branch to zrepl.github.io.
-
-.. _release-0.3:
-
-0.3
----
-
-This is a big one! Headlining features:
-
-* **Resumable Send & Recv Support**
-  No knobs required, automatically used where supported.
-* **Encrypted Send & Recv Support** for OpenZFS native encryption,
-  :ref:`configurable <job-send-options>` at the job level, i.e., for all filesystems a job is responsible for.
-* **Replication Guarantees**
-  Automatic use of ZFS holds and bookmarks to protect a replicated filesystem from losing synchronization between sender and receiver.
-  By default, zrepl guarantees that incremental replication will always be possible and interrupted steps will always be resumable.
-
-.. TIP::
-
-   We highly recommend studying the updated :ref:`overview section of the configuration chapter <overview-how-replication-works>` to understand how replication works.
-
-.. TIP::
-
-   Go 1.15 changed the default TLS validation policy to **require Subject Alternative Names (SAN) in certificates**.
-   The openssl commands we provided in the quick-start guides up to and including the zrepl 0.3 docs seem not to work properly.
-   If you encounter certificate validation errors regarding SAN and wish to continue to use your old certificates, start the zrepl daemon with env var ``GODEBUG=x509ignoreCN=0``.
-   Alternatively, generate new certificates with SANs (see :ref:`both options int the TLS transport docs <transport-tcp+tlsclientauth-certgen>` ).
-
-Quick-start guides:
-
-* We have added :ref:`another quick-start guide for a typical workstation use case for zrepl <quickstart-backup-to-external-disk>`.
-  Check it out to learn how you can use zrepl to back up your workstation's OpenZFS natively-encrypted root filesystem to an external disk.
-
-Additional changelog:
-
-* |break| Go 1.15 TLS changes mentioned above.
-* |break| |break_config| **more restrictive job names than in prior zrepl versions**
-  Starting with this version, job names are going to be embedded into ZFS holds and bookmark names (see :ref:`this section for details <zrepl-zfs-abstractions>`).
-  Therefore you might need to adjust your job names.
-  **Note that jobs** cannot be renamed easily **once you start using zrepl 0.3.**
-* |break| |mig| replication cursor representation changed
-
-  * zrepl now manages the :ref:`replication cursor bookmark <zrepl-zfs-abstractions>` per job-filesystem tuple instead of a single replication cursor per filesystem.
-    In the future, this will permit multiple sending jobs to send from the same filesystems.
-  * ZFS does not allow bookmark renaming, thus we cannot migrate the old replication cursors.
-  * zrepl 0.3 will automatically create cursors in the new format for new replications, and warn if it still finds ones in the old format.
-  * Run ``zrepl migrate replication-cursor:v1-v2`` to safely destroy old-format cursors.
-    The migration will ensure that only those old-format cursors are destroyed that have been superseeded by new-format cursors.
-
-* |feature| New option ``listen_freebind`` (tcp, tls, prometheus listener)
-* |feature| :issue:`341` Prometheus metric for failing replications + corresponding Grafana panel
-* |feature| :issue:`265` transport/tcp: support for CIDR masks in client IP whitelist
-* |feature| documented subcommand to generate ``bash`` and ``zsh`` completions
-* |feature| :issue:`307` ``chrome://trace`` -compatible activity tracing of zrepl daemon activity
-* |feature| logging: trace IDs for better log entry correlation with concurrent replication jobs
-* |feature| experimental environment variable for parallel replication (see :issue:`306` )
-* |bugfix| missing logger context vars in control connection handlers
-* |bugfix| improved error messages on ``zfs send`` errors
-* |bugfix| |docs| snapshotting: clarify sync-up behavior and warn about filesystems
-* |bugfix| transport/ssh: do not leak zombie ssh process on connection failures
-  that will not be snapshotted until the sync-up phase is over
-* |docs| Installation: :ref:`FreeBSD jail with iocage <installation-freebsd-jail-with-iocage>`
-* |docs| Document new replication features in the :ref:`config overview <overview-how-replication-works>` and :repomasterlink:`replication/design.md`.
-* **[MAINTAINER NOTICE]** New platform tests in this version, please make sure you run them for your distro!
-* **[MAINTAINER NOTICE]** Please add the shell completions to the zrepl packages.
-
-0.2.1
------
-
-* |feature| Illumos (and Solaris) compatibility and binary builds (thanks, `MNX.io <https://mnx.io>`_ )
-* |feature| 32bit binaries for Linux and FreeBSD (untested, though)
-* |bugfix| better error messages in ``ssh+stdinserver`` transport
-* |bugfix| systemd + ``ssh+stdinserver``: automatically create ``/var/run/zrepl/stdinserver``
-* |bugfix| crash if Prometheus listening socket cannot be opened
-
-* [MAINTAINER NOTICE] ``Makefile`` refactoring, see commit :commit:`080f2c0`
-
-0.2
----
-
-* |feature| :ref:`Pre- and Post-Snapshot Hooks <job-snapshotting-hooks>`
-  with built-in support for MySQL and Postgres checkpointing
-  as well as custom scripts (thanks, `@overhacked <https://github.com/overhacked>`_!)
-* |feature| Use ``zfs destroy pool/fs@snap1,snap2,...`` CLI feature if available
-* |feature| Linux ARM64 Docker build support & binary builds
-* |feature| ``zrepl status`` now displays snapshotting reports
-* |feature| ``zrepl status --job <JOBNAME>`` filter flag
-* |bugfix| i386 build
-* |bugfix| early validation of host:port tuples in config
-* |bugfix| ``zrepl status`` now supports ``TERM=screen`` (tmux on FreeBSD / FreeNAS)
-* |bugfix| ignore *connection reset by peer* errors when shutting down connections
-* |bugfix| correct error messages when receive-side pool or ``root_fs`` dataset is not imported
-* |bugfix| fail fast for misconfigured local transport
-* |bugfix| race condition in replication report generation would crash the daemon when running ``zrepl status``
-* |bugfix| rpc goroutine leak in ``push`` mode if zfs recv fails on the ``sink`` side
-* [MAINTAINER NOTICE] Go modules for dependency management both inside and outside of GOPATH
-  (``lazy.sh`` and ``Makefile`` force ``GO111MODULE=on``)
-* [MAINTAINER NOTICE] ``make platformtest`` target to check zrepl's ZFS abstractions (screen scraping, etc.).
-  These tests only work on a system with ZFS installed, and must be run as root because they create a file-backed pool for each test case.
-  The pool name ``zreplplatformtest`` is reserved for this use case.
-  Only run ``make platformtest`` on test systems, e.g. a FreeBSD VM image.
-
-0.1.1
------
-
-* |bugfix| :issue:`162` commit :commit:`d6304f4` : fix I/O timeout errors on variable receive rate
-
-  * A significant reduction or sudden stall of the receive rate (e.g. recv pool has other I/O to do)
-    would cause a ``writev I/O timeout`` error after approximately ten seconds.
-
-0.1
----
-
-This release is a milestone for zrepl and required significant refactoring if not rewrites of substantial parts of the application.
-It breaks both configuration and transport format, and thus requires manual intervention and updates on both sides of a replication setup.
-
-.. DANGER::
-   The changes in the pruning system for this release require you to explicitly define **keep rules**:
-   for any snapshot that you want to keep, at least one rule must match.
-   This is different from previous releases where pruning only affected snapshots with the configured snapshotting prefix.
-   Make sure that snapshots to be kept or ignored by zrepl are covered, e.g. by using the ``regex`` keep rule.
-   :ref:`Learn more in the config docs... <prune>`
-
-
-Notes to Package Maintainers
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-* Notify users about config changes and migrations (see changes attributed with |break| and |mig| below)
-* If the daemon crashes, the stack trace produced by the Go runtime and possibly diagnostic output of zrepl will be written to stderr.
-  This behavior is independent from the ``stdout`` outlet type.
-  Please make sure the stderr output of the daemon is captured somewhere.
-  To conserve precious stack traces, make sure that multiple service restarts do not directly discard previous stderr output.
-* Make it obvious for users how to set the ``GOTRACEBACK`` environment variable to ``GOTRACEBACK=crash``.
-  This functionality will cause SIGABRT on panics and can be used to capture a coredump of the panicking process.
-  To that extend, make sure that your package build system, your OS's coredump collection and the Go delve debugger work together.
-  Use your build system to package the Go program in `this tutorial on Go coredumps and the delve debugger <https://rakyll.org/coredumps/>`_ , and make sure the symbol resolution etc. work on coredumps captured from the binary produced by your build system. (Special focus on symbol stripping, etc.)
-* Consider using the ``zrepl configcheck`` subcommand in startup scripts to abort a restart that would fail due to an invalid config.
-
-Changes
-~~~~~~~
-
-* |break| |mig| Placeholder property representation changed
-
-  * The :ref:`placeholder property <replication-placeholder-property>` now uses ``on|off`` as values
-    instead of hashes of the dataset path. This permits renames of the sink filesystem without
-    updating all placeholder properties.
-  * Relevant for 0.0.X-0.1-rc* to 0.1 migrations
-  * Make sure your config is valid with ``zrepl configcheck``
-  * Run ``zrepl migrate 0.0.X:0.1:placeholder``
-
-* |feature| :issue:`55` : Push replication (see :ref:`push job <job-push>` and :ref:`sink job <job-sink>`)
-* |feature| :ref:`TCP Transport <transport-tcp>`
-* |feature| :ref:`TCP + TLS client authentication transport <transport-tcp+tlsclientauth>`
-* |feature| :issue:`111`: RPC protocol rewrite
-
-  * |break| Protocol breakage; Update and restart of all zrepl daemons is required.
-  * Use `gRPC <https://grpc.io/>`_ for control RPCs and a custom protocol for bulk data transfer.
-  * Automatic retries for network-temporary errors
-
-    * Limited to errors during replication for this release.
-      Addresses the common problem of ISP-forced reconnection at night, but will become
-      way more useful with resumable send & recv support.
-      Pruning errors are handled per FS, i.e., a prune RPC is attempted at least once per FS.
-
-* |feature| Proper timeout handling for the :ref:`SSH transport <transport-ssh+stdinserver>`
-
-  * |break| Requires Go 1.11 or later.
-
-* |break| |break_config|: mappings are no longer supported
-
-  * Receiving sides (``pull`` and ``sink`` job) specify a single ``root_fs``.
-    Received filesystems are then stored *per client* in ``${root_fs}/${client_identity}``.
-    See :ref:`job-overview` for details.
-
-* |feature| |break| |break_config| Manual snapshotting + triggering of replication
-
-  * |feature| :issue:`69`: include manually created snapshots in replication
-  * |break_config| ``manual`` and ``periodic`` :ref:`snapshotting types <job-snapshotting-spec>`
-  * |feature| ``zrepl signal wakeup JOB`` subcommand to trigger replication + pruning
-  * |feature| ``zrepl signal reset JOB`` subcommand to abort current replication + pruning
-
-* |feature| |break| |break_config| New pruning system
-
-  * The active side of a replication (pull or push) decides what to prune for both sender and receiver.
-    The RPC protocol is used to execute the destroy operations on the remote side.
-  * New pruning policies (see :ref:`configuration documentation <prune>` )
-
-    * The decision what snapshots shall be pruned is now made based on *keep rules*
-    * |feature| :issue:`68`: keep rule ``not_replicated`` prevents divergence of sender and receiver
-
-  * |feature| |break| Bookmark pruning is no longer necessary
-
-    * Per filesystem, zrepl creates a single bookmark (``#zrepl_replication_cursor``) and moves it forward with the most recent successfully replicated snapshot on the receiving side.
-    * Old bookmarks created by prior versions of zrepl (named like their corresponding snapshot) must be deleted manually.
-    * |break_config| ``keep_bookmarks`` parameter of the ``grid`` keep rule has been removed
-
-* |feature| ``zrepl status`` for live-updating replication progress (it's really cool!)
-* |feature| :ref:`Snapshot- & pruning-only job type <job-snap>` (for local snapshot management)
-* |feature| :issue:`67`: Expose `Prometheus <https://prometheus.io>`_ metrics via HTTP (:ref:`config docs <monitoring-prometheus>`)
-
-  * Compatible Grafana dashboard shipping in ``dist/grafana``
-
-* |break_config| Logging outlet types must be specified using the ``type`` instead of ``outlet`` key
-* |break| :issue:`53`: CLI: ``zrepl control *`` subcommands have been made direct subcommands of ``zrepl *``
-* |bugfix| Goroutine leak on ssh transport connection timeouts
-* |bugfix| :issue:`81` :issue:`77` : handle failed accepts correctly (``source`` job)
-* |bugfix| :issue:`100`: fix incompatibility with ZoL 0.8
-* |feature| :issue:`115`: logging: configurable syslog facility
-* |feature| Systemd unit file in ``dist/systemd``
-
-.. |lastrelease| replace:: 0.0.3
-
-Previous Releases
------------------
-
-.. NOTE::
-    Due to limitations in our documentation system, we only show the changelog since the last release and the time this documentation is built.
-    For the changelog of previous releases, use the version selection in the hosted version of these docs at `zrepl.github.io <https://zrepl.github.io>`_.

docs/conf.py

@@ -1,187 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-#
-# zrepl documentation build configuration file, created by
-# sphinx-quickstart on Wed Nov  8 22:28:10 2017.
-#
-# This file is execfile()d with the current directory set to its
-# containing dir.
-#
-# Note that not all possible configuration values are present in this
-# autogenerated file.
-#
-# All configuration values have a default; values that are commented out
-# serve to show the default.
-
-# If extensions (or modules to document with autodoc) are in another directory,
-# add these directories to sys.path here. If the directory is relative to the
-# documentation root, use os.path.abspath to make it absolute, like shown here.
-#
-# import os
-# import sys
-# sys.path.insert(0, os.path.abspath('.'))
-
-
-# -- General configuration ------------------------------------------------
-
-# If your documentation needs a minimal Sphinx version, state it here.
-#
-# needs_sphinx = '1.0'
-
-# Add any Sphinx extension module names here, as strings. They can be
-# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
-# ones.
-extensions = [
-    'sphinx.ext.todo',
-    'sphinx.ext.githubpages',
-    'sphinx.ext.extlinks',
-]
-
-# suppress_warnings = ['image.nonlocal_uri']
-
-# Add any paths that contain templates here, relative to this directory.
-templates_path = ['./_templates']
-
-# The suffix(es) of source filenames.
-# You can specify multiple suffix as a list of string:
-#
-# source_suffix = ['.rst', '.md']
-source_suffix = '.rst'
-
-# The master toctree document.
-master_doc = 'index'
-
-# General information about the project.
-project = 'zrepl'
-copyright = '2017-2026, Christian Schwarz'
-author = 'Christian Schwarz'
-
-# The version info for the project you're documenting, acts as replacement for
-# |version| and |release|, also used in various other places throughout the
-# built documents.
-#
-# The short X.Y version.
-version = 'latest'
-# The full version, including alpha/beta/rc tags.
-release = 'latest'
-
-# The language for content autogenerated by Sphinx. Refer to documentation
-# for a list of supported languages.
-#
-# This is also used if you do content translation via gettext catalogs.
-# Usually you set "language" from the command line for these cases.
-language = 'en'
-
-# List of patterns, relative to source directory, that match files and
-# directories to ignore when looking for source files.
-# This patterns also effect to html_static_path and html_extra_path
-exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store', '.venv']
-
-# The name of the Pygments (syntax highlighting) style to use.
-pygments_style = 'sphinx'
-
-# If true, `todo` and `todoList` produce output, else they produce nothing.
-todo_include_todos = True
-
-
-# -- Options for HTML output ----------------------------------------------
-
-# The theme to use for HTML and HTML Help pages.  See the documentation for
-# a list of builtin themes.
-#
-html_theme = 'sphinx_rtd_theme'
-
-# Theme options are theme-specific and customize the look and feel of a theme
-# further.  For a list of options available for each theme, see the
-# documentation.
-#
-# html_theme_options = {}
-
-# Add any paths that contain custom static files (such as style sheets) here,
-# relative to this directory. They are copied after the builtin static files,
-# so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['_static']
-
-html_css_files = [
-    'banner.css',
-]
-
-html_logo = '_static/zrepl.svg'
-
-html_context = {
-  # https://github.com/rtfd/sphinx_rtd_theme/issues/205
-  # Add 'Edit on Github' link instead of 'View page source'
-  "display_github": True,
-  "github_user": "zrepl",
-  "github_repo": "zrepl",
-  "github_version": "master",
-  "conf_py_path": "/docs/",
-  "source_suffix": source_suffix,
-}
-
-# -- Options for HTMLHelp output ------------------------------------------
-
-# Output file base name for HTML help builder.
-htmlhelp_basename = 'zrepldoc'
-
-
-# -- Options for LaTeX output ---------------------------------------------
-
-latex_elements = {
-    # The paper size ('letterpaper' or 'a4paper').
-    #
-    # 'papersize': 'letterpaper',
-
-    # The font size ('10pt', '11pt' or '12pt').
-    #
-    # 'pointsize': '10pt',
-
-    # Additional stuff for the LaTeX preamble.
-    #
-    # 'preamble': '',
-
-    # Latex figure (float) alignment
-    #
-    # 'figure_align': 'htbp',
-}
-
-# Grouping the document tree into LaTeX files. List of tuples
-# (source start file, target name, title,
-#  author, documentclass [howto, manual, or own class]).
-latex_documents = [
-    (master_doc, 'zrepl.tex', 'zrepl Documentation',
-     'Christian Schwarz', 'manual'),
-]
-
-
-# -- Options for manual page output ---------------------------------------
-
-# One entry per manual page. List of tuples
-# (source start file, name, description, authors, manual section).
-man_pages = [
-    (master_doc, 'zrepl', 'zrepl Documentation',
-     [author], 1)
-]
-
-
-# -- Options for Texinfo output -------------------------------------------
-
-# Grouping the document tree into Texinfo files. List of tuples
-# (source start file, target name, title, author,
-#  dir menu entry, description, category)
-texinfo_documents = [
-    (master_doc, 'zrepl', 'zrepl Documentation',
-     author, 'zrepl', 'One line description of project.',
-     'Miscellaneous'),
-]
-
-
-# -- Options for the extlinks extension -----------------------------------
-# http://www.sphinx-doc.org/en/stable/ext/extlinks.html
-extlinks = {
-    'issue':('https://github.com/zrepl/zrepl/issues/%s', 'issue #%s'),
-    'repomasterlink':('https://github.com/zrepl/zrepl/blob/master/%s', '%s'),
-    'sampleconf':('https://github.com/zrepl/zrepl/blob/master/internal/config/samples%s', 'internal/config/samples%s'),
-    'commit':('https://github.com/zrepl/zrepl/commit/%s', '%s'),
-        }
-

docs/conf.py

@@ -0,0 +1 @@
+sphinxconf/conf.py

docs/configuration.rst

@@ -7,15 +7,10 @@ Configuration
 
 .. toctree::
 
-    configuration/overview
+    configuration/preface
     configuration/jobs
     configuration/transports
-    configuration/filter_syntax
+    configuration/map_filter_syntax
-    configuration/sendrecvoptions
-    configuration/replication
-    configuration/conflict_resolution
-    configuration/snapshotting
     configuration/prune
     configuration/logging
-    configuration/monitoring
     configuration/misc

docs/configuration/conflict_resolution.rst

@@ -1,37 +0,0 @@
-.. include:: ../global.rst.inc
-
-.. _conflict_resolution-options:
-
-Conflict Resolution Options
-===========================
-
-
-::
-
-   jobs:
-   - type: push
-     filesystems: ...
-     conflict_resolution:
-       initial_replication: most_recent | all | fail # default: most_recent
-
-     ...
-
-.. _conflict_resolution-initial_replication:
-
-
-``initial_replication`` option
-------------------------------
-
-The ``initial_replication`` option determines how many snapshots zrepl replicates if the filesystem has not been replicated before.
-If ``most_recent`` (the default), the initial replication will only transfer the most recent snapshot, while ignoring previous snapshots.
-If all snapshots should be replicated, specify ``all``.
-Use ``fail`` to make replication of the filesystem fail in case there is no corresponding fileystem on the receiver.
-
-For example, suppose there are snapshosts ``tank@1``, ``tank@2``, ``tank@3`` on a sender.
-Then ``most_recent`` will replicate just ``@3``, but ``all`` will replicate ``@1``, ``@2``, and ``@3``.
-
-If initial replication is interrupted, and there is at least one (maybe partial) snapshot on the receiver, zrepl will always resume in **incremental mode**.
-And that is regardless of where the initial replication was interrupted.
-
-For example, if ``initial_replication: all`` and the transfer of ``@1`` is interrupted, zrepl would retry/resume at ``@1``.
-And even if the user changes the config to ``initial_replication: most_recent`` before resuming, **incremental mode** will still resume at ``@1``.

docs/configuration/filter_syntax.rst

@@ -1,67 +0,0 @@
-.. include:: ../global.rst.inc
-
-.. _pattern-filter:
-
-Filter Syntax
-=============
-
-For :ref:`source<job-source>`, :ref:`push<job-push>` and :ref:`snap<job-snap>` jobs, a filesystem filter must be defined (field ``filesystems``).
-A filter takes a filesystem path (in the ZFS filesystem hierarchy) as parameter and returns ``true`` (pass) or ``false`` (block).
-
-A filter is specified as a **YAML dictionary** with patterns as keys and booleans as values.
-The following rules determine which result is chosen for a given filesystem path:
-
-* More specific path patterns win over less specific ones
-* Non-wildcard patterns (full path patterns) win over *subtree wildcards* (`<` at end of pattern)
-* If the path in question does not match any pattern, the result is ``false``.
-
-The **subtree wildcard** ``<`` means "the dataset left of ``<`` and all its children".
-   
-.. TIP::
-  You can try out patterns for a configured job using the ``zrepl test filesystems`` subcommand for push and source jobs.
-
-Examples
---------
-
-Full Access
-~~~~~~~~~~~
-
-The following configuration will allow access to all filesystems.
-
-::
-
-   jobs:
-   - type: source
-     filesystems: {
-       "<": true,
-     }
-     ...
-
-
-Fine-grained
-~~~~~~~~~~~~
-
-The following configuration demonstrates all rules presented above.
-
-::
-
-    jobs:
-    - type: source
-      filesystems: {
-        "tank<": true,          # rule 1
-        "tank/foo<": false,     # rule 2
-        "tank/foo/bar": true,  # rule 3
-      }
-      ...
-
-
-Which rule applies to given path, and what is the result?
-
-::
-
-    tank/foo/bar/loo => 2    false
-    tank/bar         => 1    true
-    tank/foo/bar     => 3    true
-    zroot            => NONE false
-    tank/var/log     => 1    true
-

docs/configuration/jobs.rst

@@ -1,101 +1,33 @@
 .. include:: ../global.rst.inc
 
+.. |patient| replace:: :ref:`patient <job-term-patient>`
+.. |serve-transport| replace:: :ref:`serve transport<transport-ssh+stdinserver-serve>`
+.. |connect-transport| replace:: :ref:`connect transport<transport-ssh+stdinserver-connect>`
+.. |mapping| replace:: :ref:`mapping <pattern-mapping>`
+.. |filter| replace:: :ref:`filter <pattern-filter>`
+.. |prune| replace:: :ref:`prune <prune>`
+
 .. _job:
 
-Job Types in Detail
+Job Types
-===================
+=========
 
-.. _job-push:
+A *job* is the unit of activity tracked by the zrepl daemon and configured in the |mainconfig|.
+Every job has a unique ``name``, a ``type`` and type-dependent fields which are documented on this page.
+Check out the :ref:`tutorial` and :sampleconf:`/` for examples on how job types are actually used.
 
-Job Type ``push``
+.. ATTENTION::
------------------
 
-.. list-table::
+    Currently, zrepl does not replicate filesystem properties.
-    :widths: 20 80
+    Whe receiving a filesystem, it is never mounted (`-u` flag)  and `mountpoint=none` is set.
-    :header-rows: 1
+    This is temporary and being worked on :issue:`24`.
-
-    * - Parameter
-      - Comment
-    * - ``type``
-      - = ``push``
-    * - ``name``
-      - unique name of the job :issue:`(must not change)<327>`
-    * - ``connect``
-      - |connect-transport|
-    * - ``filesystems``
-      - |filter-spec| for filesystems to be snapshotted and pushed to the sink
-    * - ``send``
-      - |send-options| 
-    * - ``snapshotting``
-      - |snapshotting-spec|
-    * - ``pruning``
-      - |pruning-spec|
-    * - ``replication``
-      - |replication-options|
-    * - ``conflict_resolution``
-      - |conflict-resolution-options|
-
-Example config: :sampleconf:`/push.yml`
-
-.. _job-sink:
-
-Job Type ``sink``
------------------
-
-.. list-table::
-    :widths: 20 80
-    :header-rows: 1
-
-    * - Parameter
-      - Comment
-    * - ``type``
-      - = ``sink``
-    * - ``name``
-      - unique name of the job :issue:`(must not change)<327>`
-    * - ``serve``
-      - |serve-transport|
-    * - ``root_fs``
-      - ZFS filesystems are received to
-        ``$root_fs/$client_identity/$source_path``
-
-Example config: :sampleconf:`/sink.yml`
-
-.. _job-pull:
-
-Job Type ``pull``
------------------
-
-.. list-table::
-    :widths: 20 80
-    :header-rows: 1
-
-    * - Parameter
-      - Comment
-    * - ``type``
-      - = ``pull``
-    * - ``name``
-      - unique name of the job :issue:`(must not change)<327>`
-    * - ``connect``
-      - |connect-transport|
-    * - ``root_fs``
-      - ZFS filesystems are received to
-        ``$root_fs/$source_path``
-    * - ``interval``
-      - | Interval at which to pull from the source job (e.g. ``10m``).
-        | ``manual`` disables periodic pulling, replication then only happens on :ref:`wakeup <cli-signal-wakeup>`.
-    * - ``pruning``
-      - |pruning-spec|
-    * - ``replication``
-      - |replication-options|
-    * - ``conflict_resolution``
-      - |conflict-resolution-options|
-
-Example config: :sampleconf:`/pull.yml`
 
 .. _job-source:
 
-Job Type ``source``
+Source Job
--------------------
+----------
+
+Example: :sampleconf:`pullbackup/productionhost.yml`.
 
 .. list-table::
     :widths: 20 80
@@ -106,35 +38,39 @@ Job Type ``source``
     * - ``type``
       - = ``source``
     * - ``name``
-      - unique name of the job :issue:`(must not change)<327>`
+      - unique name of the job
     * - ``serve``
       - |serve-transport|
     * - ``filesystems``
-      - |filter-spec| for filesystems to be snapshotted and exposed to connecting clients
+      - |filter| for filesystems to expose to client
-    * - ``send``
+    * - ``snapshot_prefix``
-      - |send-options| 
+      - prefix for ZFS snapshots taken by this job
-    * - ``snapshotting``
+    * - ``interval``
-      - |snapshotting-spec|
+      - snapshotting interval
-
+    * - ``prune``
-Example config: :sampleconf:`/source.yml`
+      - |prune| for versions of filesytems in ``filesystems``, versions prefixed with ``snapshot_prefix``
 
 
-.. _replication-local:
+- Snapshotting Task (every ``interval``, |patient|)
 
-Local replication
+  - A snapshot of filesystems matched by ``filesystems`` is taken every ``interval`` with prefix ``snapshot_prefix``.
------------------
+  - A bookmark of that snapshot is created with the same name.
+  - The ``prune`` policy is evaluated for versions of filesystems matched by ``filesystems``, versions prefixed with ``snapshot_prefix``.
 
-If you have the need for local replication (most likely between two local storage pools), you can use the :ref:`local transport type <transport-local>` to connect a local push job to a local sink job.
+- Serve Task
 
-Example config: :sampleconf:`/local.yml`.
+  - Wait for connections from pull job using ``serve``.
 
+A source job is the counterpart to a :ref:`job-pull`.
 
-.. _job-snap:
+Make sure you read the |prune| policy documentation.
 
-Job Type ``snap`` (snapshot & prune only)
+.. _job-pull:
------------------------------------------
 
-Job type that only takes snapshots and performs pruning on the local machine.
+Pull Job
+--------
+
+Example: :sampleconf:`pullbackup/backuphost.yml`
 
 .. list-table::
     :widths: 20 80
@@ -143,14 +79,109 @@ Job type that only takes snapshots and performs pruning on the local machine.
     * - Parameter
       - Comment
     * - ``type``
-      - = ``snap``
+      - = ``pull``
     * - ``name``
-      - unique name of the job :issue:`(must not change)<327>`
+      - unqiue name of the job
-    * - ``filesystems``
+    * - ``connect``
-      - |filter-spec| for filesystems to be snapshotted
+      - |connect-transport|
-    * - ``snapshotting``
+    * - ``interval``
-      - |snapshotting-spec|
+      - Interval between pull attempts
-    * - ``pruning``
+    * - ``mapping``
-      - |pruning-spec|
+      - |mapping| for remote to local filesystems
+    * - ``initial_repl_policy``
+      - default = ``most_recent``, initial replication policy
+    * - ``snapshot_prefix``
+      - prefix snapshots must match to be considered for replication & pruning
+    * - ``prune``
+      - |prune| policy for versions of filesystems of local filesystems reachable by ``mapping``, versions prefixed with ``snapshot_prefix``
 
-Example config: :sampleconf:`/snap.yml`
+* Main Task (every ``interval``, |patient|)
+ 
+  #. A connection to the remote source job is established using the strategy in ``connect``
+  #. ``mapping`` maps filesystems presented by the remote side to local *target filesystems*
+  #. Those remote filesystems with a local *target filesystem* are replicated
+ 
+     #. Only snapshots with prefix ``snapshot_prefix`` are replicated.
+     #. If possible, incremental replication takes place.
+     #. If the local target filesystem does not exist, ``initial_repl_policy`` is used.
+     #. On conflicts, an error is logged but replication of other filesystems with mapping continues.
+  
+  #. The ``prune`` policy is evaluated for all *target filesystems*
+
+A pull job is the counterpart to a :ref:`job-source`.
+
+Make sure you read the |prune| policy documentation.
+
+.. _job-local:
+
+Local Job
+---------
+
+Example: :sampleconf:`localbackup/host1.yml`
+
+.. list-table::
+    :widths: 20 80
+    :header-rows: 1
+
+    * - Parameter
+      - Comment
+    * - ``type``
+      - = ``local``
+    * - ``name``
+      - unqiue name of the job
+    * - ``mapping``
+      - |mapping| from source to target filesystem (both local)
+    * - ``snapshot_prefix``
+      - prefix for ZFS snapshots taken by this job
+    * - ``interval``
+      - snapshotting & replication interval
+    * - ``initial_repl_policy``
+      - default = ``most_recent``, initial replication policy
+    * - ``prune_lhs``
+      -  pruning policy on left-hand-side (source)
+    * - ``prune_rhs``
+      - pruning policy on right-hand-side (target)
+
+* Main Task (every ``interval``, |patient|)
+
+  #. Evaluate ``mapping`` for local filesystems, those with a *target filesystem* are called *mapped filesystems*.
+  #. Snapshot *mapped filesystems* with ``snapshot_prefix``.
+  #. Bookmark the snapshot created above.
+  #. Replicate *mapped filesystems* to their respective *target filesystems*:
+
+     #. Only snapshots with prefix ``snapshot_prefix`` are replicated.
+     #. If possible, incremental replication takes place.
+     #. If the *target filesystem* does not exist, ``initial_repl_policy`` is used.
+     #. On conflicts, an error is logged but replication of other *mapped filesystems* continues.
+
+  #. The ``prune_lhs`` policy is triggered for all *mapped filesystems*
+  #. The ``prune_rhs`` policy is triggered for all *target filesystems*
+
+A local job is combination of source & pull job executed on the same machine.
+
+Terminology
+-----------
+
+task
+
+    A job consists of one or more tasks and a task consists of one or more steps.
+    Some tasks may be periodic while others wait for an event to occur.
+
+
+patient task
+
+    .. _job-term-patient:
+
+    A patient task is supposed to execute some task every `interval`.
+    We call the start of the task an *invocation*.
+
+    * If the task completes in less than `interval`, the task is restarted at `last_invocation + interval`.
+    * Otherwise, a patient job
+        * logs a warning as soon as a task exceeds its configured `interval`
+        * waits for the last invocation to finish
+        * logs a warning with the effective task duration
+        * immediately starts a new invocation of the task
+
+filesystem version
+
+    A snapshot or a bookmark.

docs/configuration/logging.rst

@@ -7,18 +7,19 @@ Logging
 
 zrepl uses structured logging to provide users with easily processable log messages.
 
-Logging outlets are configured in the ``global`` section of the config file.
+Logging outlets are configured in the ``global`` section of the |mainconfig|.
+Check out :sampleconf:`random/logging.yml` for an example on how to configure multiple outlets:
 
 ::
 
     global:
       logging:
     
-        - type: OUTLET_TYPE
+        - outlet: OUTLET_TYPE
           level: MINIMUM_LEVEL
           format: FORMAT
     
-        - type: OUTLET_TYPE
+        - outlet: OUTLET_TYPE
           level: MINIMUM_LEVEL
           format: FORMAT
     
@@ -44,7 +45,7 @@ By default, the following logging configuration is used
     global:
       logging:
     
-        - type: "stdout"
+        - outlet: "stdout"
           level:  "warn"
           format: "human"
 
@@ -92,8 +93,8 @@ Formats
     * - Format
       - Description
     * - ``human``
-      - prints job and subsystem into brackets before the actual message,
+      - emphasizes context by putting job, task, step and other context variables into brackets
-        followed by remaining fields in logfmt style
+        before the actual message, followed by remaining fields in logfmt style|
     * - ``logfmt``
       - `logfmt <https://brandur.org/logfmt>`_ output. zrepl uses `this Go package <https://github.com/go-logfmt/logfmt>`_.
     * - ``json``
@@ -117,7 +118,7 @@ Outlets are the destination for log entries.
 
     * - Parameter
       - Comment
-    * - ``type``
+    * - ``outlet``
       - ``stdout``
     * - ``level``
       -  minimum  :ref:`log level <logging-levels>`
@@ -125,11 +126,9 @@ Outlets are the destination for log entries.
       - output :ref:`format <logging-formats>`
     * - ``time``
       - always include time in output (``true`` or ``false``)
-    * - ``color``
-      - colorize output according to log level (``true`` or ``false``)
 
 Writes all log entries with minimum level ``level`` formatted by ``format`` to stdout.
-If stdout is a tty, interactive usage is assumed and both ``time`` and ``color`` are set to ``true``.
+If stdout is a tty, interactive usage is assumed and the current time is included in the output.
 
 Can only be specified once.
 
@@ -141,14 +140,12 @@ Can only be specified once.
 
     * - Parameter
       - Comment
-    * - ``type``
+    * - ``outlet``
       - ``syslog``
     * - ``level``
       -  minimum  :ref:`log level <logging-levels>`
     * - ``format``
       - output :ref:`format <logging-formats>`
-    * - ``facility``
-      - Which syslog facility to use (default = ``local0``)
     * - ``retry_interval``
       - Interval between reconnection attempts to syslog (default = 0)
 
@@ -166,7 +163,7 @@ Can only be specified once.
 
     * - Parameter
       - Comment
-    * - ``type``
+    * - ``outlet``
       - ``tcp``
     * - ``level``
       -  minimum  :ref:`log level <logging-levels>`
@@ -182,9 +179,11 @@ Can only be specified once.
       - TLS config (see below)
 
 Establishes a TCP connection to ``address`` and sends log messages with minimum level ``level`` formatted by ``format``.
+
 If ``tls`` is not specified, an unencrypted connection is established.
+
 If ``tls`` is specified, the TCP connection is secured with TLS + Client Authentication.
-The latter is particularly useful in combination with log aggregation services.
+This is particularly useful in combination with log aggregation services that run on an other machine.
 
 .. list-table::
     :widths: 10 90
@@ -202,7 +201,7 @@ The latter is particularly useful in combination with log aggregation services.
 .. WARNING::
 
     zrepl drops log messages to the TCP outlet if the underlying connection is not fast enough.
-    Note that TCP buffering in the kernel must first run full before messages are dropped.
+    Note that TCP buffering in the kernel must first run full becfore messages are dropped.
 
     Make sure to always configure a ``stdout`` outlet as the special error outlet to be informed about problems
     with the TCP outlet (see :ref:`above <logging-error-outlet>` ).

docs/configuration/map_filter_syntax.rst

@@ -0,0 +1,108 @@
+.. include:: ../global.rst.inc
+
+Mapping & Filter Syntax
+=======================
+
+For various job types, a filesystem ``mapping`` or ``filter`` needs to be
+specified.
+
+Both have in common that they take a filesystem path (in the ZFS filesystem hierarchy)as parameters and return something.
+Mappings return a *target filesystem* and filters return a *filter result*.
+
+The pattern syntax is the same for mappings and filters and is documented in the following section.
+
+Common Pattern Syntax
+---------------------
+
+A mapping / filter is specified as a **YAML dictionary** with patterns as keys and
+results as values.
+The following rules determine which result is chosen for a given filesystem path:
+
+* More specific path patterns win over less specific ones
+* Non-wildcard patterns (full path patterns) win over *subtree wildcards* (`<` at end of pattern)
+
+The **subtree wildcard** ``<`` means "the dataset left of ``<`` and all its children".
+
+Example
+~~~~~~~
+
+::
+
+    # Rule number and its pattern
+    1: tank<            # tank and all its children
+    2: tank/foo/bar     # full path pattern (no wildcard)
+    3: tank/foo<        # tank/foo and all its children
+    
+    # Which rule applies to given path?
+    tank/foo/bar/loo => 3
+    tank/bar         => 1
+    tank/foo/bar     => 2
+    zroot            => NO MATCH
+    tank/var/log     => 1
+
+.. _pattern-mapping:
+
+Mappings
+--------
+
+Mappings map a *source filesystem path* to a *target filesystem path*.
+Per pattern, either a target filesystem path or ``"!"`` is specified as a result.
+
+* If no pattern matches, there exists no target filesystem (``NO MATCH``).
+* If the result is a ``"!"``, there exists no target filesystem (``NO MATCH``).
+* If the pattern is a non-wildcard pattern, the source path is mapped to the target path on the right.
+* If the pattern ends with a *subtree wildcard* (``<``), the source path is **prefix-trimmed** with the path specified left of ``<``.
+
+  * Note: this means that only for *wildcard-only* patterns (pattern= ``<`` ) is the source path simply appended to the target path.
+
+The example is from the :sampleconf:`localbackup/host1.yml` example config.
+
+::
+
+    jobs:
+    - name: mirror_local
+      type: local
+      mapping: {
+        "zroot/var/db<":    "storage/backups/local/zroot/var/db",
+        "zroot/usr/home<":  "storage/backups/local/zroot/usr/home",
+        "zroot/usr/home/paranoid":  "!", #don't backup paranoid user
+        "zroot/poudriere/ports<": "!", #don't backup the ports trees
+      }
+      ...
+
+
+::
+
+    zroot/var/db                        => storage/backups/local/zroot/var/db
+    zroot/var/db/a/child                => storage/backups/local/zroot/var/db/a/child
+    zroot/usr/home                      => storage/backups/local/zroot/usr/home
+    zroot/usr/home/paranoid             => NOT MAPPED
+    zroot/usr/home/bob                  => storage/backups/local/zroot/usr/home/bob
+    zroot/usr/src                       => NOT MAPPED
+    zroot/poudriere/ports/2017Q3        => NOT MAPPED
+    zroot/poudriere/ports/HEAD          => NOT MAPPED
+
+.. TIP::
+
+   You can try out patterns for a configured job using the ``zrepl test`` subcommand.
+
+.. _pattern-filter:
+
+Filters
+-------
+
+Valid filter results: ``ok`` or ``!``.
+
+The example below show the source job from the :ref:`tutorial <tutorial-configure-app-srv>`:
+The corresponding pull job is allowed access to ``zroot/var/db``, ``zroot/usr/home`` + children except ``zroot/usr/home/paranoid``::
+
+    jobs:
+    - name: pull_backup
+      type: source
+      ...
+      filesystems: {
+        "zroot/var/db": "ok",
+        "zroot/usr/home<": "ok",
+        "zroot/usr/home/paranoid": "!",
+      }
+      ...

docs/configuration/misc.rst

@@ -1,5 +1,3 @@
-.. _miscellaneous:
-
 Miscellaneous
 =============
 
@@ -8,16 +6,16 @@ Miscellaneous
 Runtime Directories & UNIX Sockets
 ----------------------------------
 
-The zrepl daemon needs to open various UNIX sockets in a runtime directory:
+zrepl daemon creates various UNIX sockets to allow communicating with it:
 
-* a ``control`` socket that the CLI commands use to interact with the daemon
+* the :ref:`transport-ssh+stdinserver` transport connects to a socket named after ``client_identity`` parameter
-* the :ref:`transport-ssh+stdinserver` listener opens one socket per configured client, named after ``client_identity`` parameter
+* the ``control`` CLI subcommand connects to a defined control socket
 
-There is no authentication on these sockets except the UNIX permissions.
+There is no further authentication on these sockets.
-The zrepl daemon will refuse to bind any of the above sockets in a directory that is world-accessible.
+Therefore we have to make sure they can only be created and accessed by ``zrepl daemon``.
+In fact, ``zrepl daemon`` will not bind a socket to a path in a directory that is world-accessible.
 
-The following sections of the ``global`` config shows the default paths.
+The directories can be configured in the main configuration file, the defaults are provided below:
-The shell script below shows how the default runtime directory can be created.
 
 ::
 
@@ -29,12 +27,6 @@ The shell script below shows how the default runtime directory can be created.
           sockdir: /var/run/zrepl/stdinserver
 
 
-::
-
-    mkdir -p /var/run/zrepl/stdinserver
-    chmod -R 0700 /var/run/zrepl
-
-
 Durations & Intervals
 ---------------------
 
@@ -44,3 +36,27 @@ Interval & duration fields in job definitions, pruning configurations, etc. must
 
     var durationStringRegex *regexp.Regexp = regexp.MustCompile(`^\s*(\d+)\s*(s|m|h|d|w)\s*$`)
     // s = second, m = minute, h = hour, d = day, w = week (7 days)
+
+Super-Verbose Job Debugging
+---------------------------
+
+You have probably landed here because you opened an issue on GitHub and some developer told you to do this...
+So just read the annotated comments ;)
+
+::
+
+    job:
+    - name: ...
+      ...
+     # JOB DEBUGGING OPTIONS
+      # should be equal for all job types, but each job implements the debugging itself
+      debug:
+        conn: # debug the io.ReadWriteCloser connection
+          read_dump: /tmp/connlog_read   # dump results of Read() invocations to this file
+          write_dump: /tmp/connlog_write # dump results of Write() invocations to this file
+        rpc: # debug the RPC protocol implementation
+          log: true # log output from rpc layer to the job log
+
+.. ATTENTION::
+
+    Connection dumps will almost certainly contain your or other's private data. Do not share it in a bug report.

docs/configuration/monitoring.rst

@@ -1,37 +0,0 @@
-.. include:: ../global.rst.inc
-
-.. _monitoring:
-
-Monitoring
-==========
-
-Monitoring endpoints are configured in the ``global.monitoring`` section of the config file.
-
-.. _monitoring-prometheus:
-
-Prometheus & Grafana
---------------------
-
-zrepl can expose `Prometheus metrics <https://prometheus.io/docs/instrumenting/exposition_formats/>`_ via HTTP.
-The ``listen`` attribute is a `net.Listen <https://golang.org/pkg/net/#Listen>`_  string for tcp, e.g. ``:9811`` or ``127.0.0.1:9811`` (port 9811 was reserved to zrepl `on the official list <https://github.com/prometheus/prometheus/wiki/Default-port-allocations/_compare/43e495dd251ee328ac0d08b58084665b5c0f7a7e...459195059b55b414193ebeb80c5ba463d2606951>`_).
-The ``listen_freebind`` attribute is :ref:`explained here <listen-freebind-explanation>`.
-The Prometheus monitoring job appears in the ``zrepl control`` job list and may be specified **at most once**.
-
-zrepl also ships with an importable `Grafana <https://grafana.com>`_ dashboard that consumes the Prometheus metrics:
-see :repomasterlink:`dist/grafana`.
-The dashboard also contains some advice on which metrics are important to monitor.
-
-.. NOTE::
-
-  At the time of writing, there is no stability guarantee on the exported metrics.
-
-::
-
-    global:
-      monitoring:
-        - type: prometheus
-          listen: ':9811'
-          listen_freebind: true # optional, default false
-
-
-

docs/configuration/overview.rst

@@ -1,364 +0,0 @@
-
-Overview & Terminology
-======================
-
-All work zrepl does is performed by the zrepl daemon which is configured in a single YAML configuration file loaded on startup.
-The following paths are searched, in this order:
-
-1. The path specified via the global ``--config`` flag
-2. ``/etc/zrepl/zrepl.yml``
-3. ``/usr/local/etc/zrepl/zrepl.yml``
-
-``zrepl configcheck`` can be used to validate the configuration.
-If the configuration is valid, it will output nothing and exit with code ``0``.
-The error messages vary in quality and usefulness: please report confusing config errors to the tracking :issue:`155`.
-
-Full example configs are available at :ref:`quick-start guides <quickstart-toc>` and :sampleconf:`/`.
-However, copy-pasting examples is no substitute for reading documentation!
-
-Config File Structure
----------------------
-
-.. code-block:: yaml
-
-   global: ...
-   jobs:
-   - name: backup
-     type: push
-   - ...
-
-A zrepl configuration file is divided in to two main sections: ``global`` and ``jobs``.
-``global`` has sensible defaults. It is covered in :ref:`logging <logging>`, :ref:`monitoring <monitoring>` \& :ref:`miscellaneous <miscellaneous>`.
-
-
-``conf.d``: including config files
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-It is possible to distribute zrepl configurations over multiple YAML configuration
-files. This is achieved by using the `include` key which can only exist in the main
-configuration file.
-
-The list of included paths must point to either individual YAML files or directories.
-If the path points to a directory then all YAML files with a `.yml` extention in the
-directory will be included.
-
-.. code-block:: yaml
-
-    # /etc/zrepl/zrep.yml
-    global: ...
-    include:
-      - ./jobs.d/
-      - /opt/some_job.yml
-
-    # /etc/zrepl/jobs.d/backup.yml
-    jobs:
-    - name: backup
-      type: push
-    - ...
-
-    # /opt/some_job.yml
-    jobs:
-    - name: another_job
-      ...
-    - ...
-
-
-
-The paths are treated as absolute when starting with `/`, otherwise
-as a relative path to the main config file's parent directory.
-
-Job names must be unique across all included configuration files.
-
-.. _job-overview:
-
-Jobs \& How They Work Together
-------------------------------
-
-A *job* is the unit of activity tracked by the zrepl daemon.
-The ``type`` of a job determines its role in a replication setup and in snapshot management.
-Jobs are identified by their ``name``, both in log files and the ``zrepl status`` command.
-
-.. NOTE::
-   The job name is persisted in several places on disk and thus :issue:`cannot be changed easily<327>`.
-
-Replication always happens between a pair of jobs: one **active side** and one **passive side**.
-The active side connects to the passive side using a :ref:`transport <transport>` and starts executing the replication logic.
-The passive side responds to requests from the active side after checking its permissions.
-
-The following table shows how different job types can be combined to achieve **both push and pull mode setups**.
-Note that snapshot-creation denoted by "(snap)" is orthogonal to whether a job is active or passive.
-
-+-----------------------+--------------+----------------------------------+------------------------------------------------------------------------------------+
-| Setup name            | active side  | passive side                     | use case                                                                           |
-+=======================+==============+==================================+====================================================================================+
-| Push mode             | ``push``     | ``sink``                         | * Laptop backup                                                                    |
-|                       | (snap)       |                                  | * NAS behind NAT to offsite                                                        |
-+-----------------------+--------------+----------------------------------+------------------------------------------------------------------------------------+
-| Pull mode             | ``pull``     | ``source``                       | * Central backup-server for many nodes                                             |
-|                       |              | (snap)                           | * Remote server to NAS behind NAT                                                  |
-+-----------------------+--------------+----------------------------------+------------------------------------------------------------------------------------+
-| Local replication     | | ``push`` + ``sink`` in one config             | * Backup to :ref:`locally attached disk <quickstart-backup-to-external-disk>`      |
-|                       | | with :ref:`local transport <transport-local>` | * Backup FreeBSD boot pool                                                         |
-+-----------------------+--------------+----------------------------------+------------------------------------------------------------------------------------+
-| Snap & prune-only     | ``snap``     | N/A                              | * | Snapshots & pruning but no replication                                         |
-|                       | (snap)       |                                  |   | required                                                                       |
-|                       |              |                                  | * Workaround for :ref:`source-side pruning <prune-workaround-source-side-pruning>` |
-+-----------------------+--------------+----------------------------------+------------------------------------------------------------------------------------+
-
-How the Active Side Works
--------------------------
-
-The active side (:ref:`push <job-push>` and :ref:`pull <job-pull>` job) executes the replication and pruning logic:
-
-1. Wakeup after snapshotting (``push`` job) or pull interval ticker (``pull`` job).
-2. Connect to the passive side and instantiate an RPC client.
-3. Replicate data from the sender to the receiver.
-4. Prune on sender & receiver.
-
-.. TIP::
-  The progress of the active side can be watched live using ``zrepl status``.
-
-.. _overview-passive-side--client-identity:
-
-How the Passive Side Works
---------------------------
-
-The passive side (:ref:`sink <job-sink>` and :ref:`source <job-source>`) waits for connections from the active side,
-on the :ref:`transport <transport>` specified with ``serve`` in the job configuration.
-The respective transport then perfoms authentication & authorization, resulting in a stable *client identity*.
-The passive side job uses this *client identity* as follows:
-
-   * In ``sink`` jobs, to map requests from different *client identities* to their respective sub-filesystem tree ``root_fs/${client_identity}``.
-   * *In the future, ``source`` might embed the client identity in :ref:`zrepl's ZFS abstraction names <zrepl-zfs-abstractions>`, to support multi-host replication.*
-
-.. TIP::
-   The use of the client identity in the ``sink`` job implies that it must be usable as a ZFS ZFS filesystem name component.
-
-.. _overview-how-replication-works:
-
-How Replication Works
----------------------
-
-One of the major design goals of the replication module is to avoid any duplication of the nontrivial logic.
-As such, the code works on abstract senders and receiver **endpoints**, where typically one will be implemented by a local program object and the other is an RPC client instance.
-Regardless of push- or pull-style setup, the logic executes on the active side, i.e. in the ``push`` or ``pull`` job.
-
-The following high-level steps take place during replication and can be monitored using ``zrepl status``:
-
-* Plan the replication:
-
-  * Compare sender and receiver filesystem snapshots
-  * Build the **replication plan**
-
-    * Per filesystem, compute a diff between sender and receiver snapshots
-    * Build a list of **replication steps**
-
-      * If possible, use incremental and resumable sends
-      * Otherwise, use full send of most recent snapshot on sender
-
-  * Retry on errors that are likely temporary (i.e. network failures).
-  * Give up on filesystems where a permanent error was received over RPC.
-
-* Execute the plan
-
-  * Perform replication steps in the following order:
-    Among all filesystems with pending replication steps, pick the filesystem whose next replication step's snapshot is the oldest.
-  * Create placeholder filesystems on the receiving side to mirror the dataset paths on the sender to ``root_fs/${client_identity}``.
-  * Acquire send-side *step-holds* on the step's `from` and `to` snapshots.
-  * Perform the replication step.
-  * Move the **replication cursor** bookmark on the sending side (see below).
-  * Move the **last-received-hold** on the receiving side (see below).
-  * Release the send-side step-holds.
-
-The idea behind the execution order of replication steps is that if the sender snapshots all filesystems simultaneously at fixed intervals, the receiver will have all filesystems snapshotted at time ``T1`` before the first snapshot at ``T2 = T1 + $interval`` is replicated.
-
-ZFS Background Knowledge
-^^^^^^^^^^^^^^^^^^^^^^^^
-This section gives some background knowledge about ZFS features that zrepl uses to provide guarantees for a replication filesystem.
-Specifically, zrepl guarantees by default that **incremental replication is always possible and that started replication steps can always be resumed if they are interrupted.**
-
-**ZFS Send Modes & Bookmarks**
-ZFS supports full sends (``zfs send fs@to``) and incremental sends (``zfs send -i @from fs@to``).
-Full sends are used to create a new filesystem on the receiver with the send-side state of ``fs@to``.
-Incremental sends only transfer the delta between ``@from`` and ``@to``.
-Incremental sends require that ``@from`` be present on the receiving side when receiving the incremental stream.
-Incremental sends can also use a ZFS bookmark as *from* on the sending side (``zfs send -i #bm_from fs@to``), where ``#bm_from`` was created using ``zfs bookmark fs@from fs#bm_from``.
-The receiving side must always have the actual snapshot ``@from``, regardless of whether the sending side uses ``@from`` or a bookmark of it.
-
-.. _zfs-background-knowledge-plain-vs-raw-sends:
-
-**Plain and raw sends**
-By default, ``zfs send`` sends the most generic, backwards-compatible data stream format (so-called 'plain send').
-If the sent uses newer features, e.g. compression or encryption, ``zfs send`` has to un-do these operations on the fly to produce the plain send stream.
-If the receiver uses newer features (e.g. compression or encryption inherited from the parent FS), it applies the necessary transformations again on the fly during ``zfs recv``.
-
-Flags such as ``-e``, ``-c`` and ``-L``  tell ZFS to produce a send stream that is closer to how the data is stored on disk.
-Sending with those flags removes computational overhead from sender and receiver.
-However, the receiver will not apply certain transformations, e.g., it will not compress with the receive-side ``compression`` algorithm.
-
-The ``-w`` (``--raw``) flag produces a send stream that is as *raw* as possible.
-For unencrypted datasets, its current effect is the same as ``-Lce``.
-
-Encrypted datasets can only be sent plain (unencrypted) or raw (encrypted) using the ``-w`` flag.
-
-**Resumable Send & Recv**
-The ``-s`` flag for ``zfs recv`` tells zfs to save the partially received send stream in case it is interrupted.
-To resume the replication, the receiving side filesystem's ``receive_resume_token`` must be passed to a new ``zfs send -t <value> | zfs recv`` command.
-A full send can only be resumed if ``@to`` still exists.
-An incremental send can only be resumed if ``@to`` still exists *and* either ``@from`` still exists *or* a bookmark ``#fbm`` of ``@from`` still exists.
-
-**ZFS Holds**
-ZFS holds prevent a snapshot from being deleted through ``zfs destroy``, letting the destroy fail with a ``datset is busy`` error.
-Holds are created and referred to by a *tag*. They can be thought of as a named, persistent lock on the snapshot.
-
-
-.. _zrepl-zfs-abstractions:
-
-ZFS Abstractions Managed By zrepl
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-With the background knowledge from the previous paragraph, we now summarize the different on-disk ZFS objects that zrepl manages to provide its functionality.
-
-.. _replication-placeholder-property:
-
-**Placeholder filesystems** on the receiving side are regular ZFS filesystems with the ZFS property ``zrepl:placeholder=on``.
-Placeholders allow the receiving side to mirror the sender's ZFS dataset hierarchy without replicating every filesystem at every intermediary dataset path component.
-Consider the following example: ``S/H/J`` shall be replicated to ``R/sink/job/S/H/J``, but neither ``S/H`` nor ``S`` shall be replicated.
-ZFS requires the existence of ``R/sink/job/S`` and ``R/sink/job/S/H`` in order to receive into ``R/sink/job/S/H/J``.
-Thus, zrepl creates the parent filesystems as placeholders on the receiving side.
-If at some point ``S/H`` and ``S`` shall be replicated, the receiving side invalidates the placeholder flag automatically.
-The ``zrepl test placeholder`` command can be used to check whether a filesystem is a placeholder.
-
-.. _replication-cursor-and-last-received-hold:
-
-The **replication cursor** bookmark and **last-received-hold** are managed by zrepl to ensure that future replications can always be done incrementally.
-The replication cursor is a send-side bookmark of the most recent successfully replicated snapshot,
-and the last-received-hold is a hold of that snapshot on the receiving side.
-Both are moved atomically after the receiving side has confirmed that a replication step is complete.
-
-The replication cursor has the format ``#zrepl_CUSOR_G_<GUID>_J_<JOBNAME>``.
-The last-received-hold tag has the format ``zrepl_last_received_J_<JOBNAME>``.
-Encoding the job name in the names ensures that multiple sending jobs can replicate the same filesystem to different receivers without interference.
-
-.. _tentative-replication-cursor-bookmarks:
-
-**Tentative replication cursor bookmarks** are short-lived bookmarks that protect the atomic moving-forward of the replication cursor and last-received-hold (see :issue:`this issue <340>`).
-They are only necessary if step holds are not used as per the :ref:`replication.protection <replication-option-protection>` setting.
-The tentative replication cursor has the format ``#zrepl_CUSORTENTATIVE_G_<GUID>_J_<JOBNAME>``.
-The ``zrepl zfs-abstraction list`` command provides a listing of all bookmarks and holds managed by zrepl.
-
-.. _step-holds:
-
-**Step holds** are zfs holds managed by zrepl to ensure that a replication step can always be resumed if it is interrupted, e.g., due to network outage.
-zrepl creates step holds before it attempts a replication step and releases them after the receiver confirms that the replication step is complete.
-For an initial replication ``full @initial_snap``, zrepl puts a zfs hold on ``@initial_snap``.
-For an incremental send ``@from -> @to``, zrepl puts a zfs hold on both ``@from`` and ``@to``.
-Note that ``@from`` is not strictly necessary for resumability -- a bookmark on the sending side would be sufficient --, but size-estimation in currently used OpenZFS versions only works if ``@from`` is a snapshot.
-The hold tag has the format ``zrepl_STEP_J_<JOBNAME>``.
-A job only ever has one active send per filesystem.
-Thus, there are never more than two step holds for a given pair of ``(job,filesystem)``.
-
-**Step bookmarks** are zrepl's equivalent for holds on bookmarks (ZFS does not support putting holds on bookmarks).
-They are intended for a situation where a replication step uses a bookmark ``#bm`` as incremental ``from`` where ``#bm`` is not managed by zrepl.
-To ensure resumability, zrepl copies ``#bm`` to step bookmark ``#zrepl_STEP_G_<GUID>_J_<JOBNAME>``.
-If the replication is interrupted and ``#bm`` is deleted by the user, the step bookmark remains as an incremental source for the resumable send.
-Note that zrepl does not yet support creating step bookmarks because the `corresponding ZFS feature for copying bookmarks <https://github.com/openzfs/zfs/pull/9571>`_ is not yet widely available .
-Subscribe to zrepl :issue:`326` for details.
-
-The ``zrepl zfs-abstraction list`` command provides a listing of all bookmarks and holds managed by zrepl.
-
-.. NOTE::
-
-    More details can be found in the design document :repomasterlink:`replication/design.md`.
-
-
-Caveats With Complex Setups (More Than 2 Jobs or Machines)
-----------------------------------------------------------
-
-Most users are served well with a single sender and a single receiver job.
-This section documents considerations for more complex setups.
-
-.. ATTENTION::
-
-   Before you continue, make sure you have a working understanding of :ref:`how zrepl works <overview-how-replication-works>`
-   and :ref:`what zrepl does to ensure <zrepl-zfs-abstractions>` that replication between sender and receiver is always
-   possible without conflicts.
-   This will help you understand why certain kinds of multi-machine setups do not (yet) work.
-
-.. NOTE::
-
-   If you can't find your desired configuration, have questions or would like to see improvements to multi-job setups, please `open an issue on GitHub <https://github.com/zrepl/zrepl/issues/new>`_.
-
-Multiple Jobs on One Machine
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-As a general rule, multiple jobs configured on one machine **must operate on disjoint sets of filesystems**.
-Otherwise, concurrently running jobs might interfere when operating on the same filesystem.
-
-On your setup, ensure that
-
-* all ``filesystems`` filter specifications are disjoint
-* no ``root_fs`` is a prefix or equal to another ``root_fs``
-
-  * For ``sink`` jobs, consider all possible ``root_fs/${client_identity}``.
-
-* no ``filesystems`` filter matches any ``root_fs``
-
-**Exceptions to the rule**:
-
-* A ``snap`` and ``push`` job on the same machine can match the same ``filesystems``.
-  To avoid interference, only one of the jobs should be pruning snapshots on the sender, the other one should keep all snapshots.
-  Since the jobs won't coordinate, errors in the log are to be expected, but :ref:`zrepl's ZFS abstractions <zrepl-zfs-abstractions>` ensure that ``push`` and ``sink`` can always replicate incrementally.
-  This scenario is detailed in one of the :ref:`quick-start guides <quickstart-backup-to-external-disk>`.
-
-
-Two Or More Machines
-^^^^^^^^^^^^^^^^^^^^
-
-This section might be relevant to users who wish to *fan-in* (N machines replicate to 1) or *fan-out* (replicate 1 machine to N machines).
-
-**Working setups**:
-
-* **Fan-in: N servers replicated to one receiver, disjoint dataset trees.**
-
-  * This is the common use case of a centralized backup server.
-
-  * Implementation:
-
-    * N ``push`` jobs (one per sender server), 1 ``sink`` (as long as the different push jobs have a different :ref:`client identity <overview-passive-side--client-identity>`)
-    * N ``source`` jobs (one per sender server), N ``pull`` on the receiver server (unique names, disjoint  ``root_fs``)
-
-  * The ``sink`` job automatically constrains each client to a disjoint sub-tree of the sink-side dataset hierarchy ``${root_fs}/${client_identity}``.
-    Therefore, the different clients cannot interfere.
-
-  * The ``pull`` job only pulls from one host, so it's up to the zrepl user to ensure that the different ``pull`` jobs don't interfere.
-
-.. _fan-out-replication:
-
-* **Fan-out: 1 server replicated to N receivers**
-
-  * Can be implemented either in a pull or push fashion.
-
-    * **pull setup**: 1 ``pull`` job on each receiver server, each with a corresponding **unique** ``source`` job on the sender server.
-    * **push setup**: 1 ``sink`` job on each receiver server, each with a corresponding **unique** ``push`` job on the sender server.
-
-  * It is critical that we have one sending-side job (``source``, ``push``) per receiver.
-    The reason is that :ref:`zrepl's ZFS abstractions <zrepl-zfs-abstractions>` (``zrepl zfs-abstraction list``) include the name of the ``source``/``push`` job, but not the receive-side job name or client identity (see :issue:`380`).
-    As a counter-example, suppose we used multiple ``pull`` jobs with only one ``source`` job.
-    All ``pull`` jobs would share the same :ref:`replication cursor bookmark <replication-cursor-and-last-received-hold>` and trip over each other, breaking incremental replication guarantees quickly.
-    The anlogous problem exists for 1 ``push`` to N ``sink`` jobs.
-
-  * The ``filesystems`` matched by the sending side jobs (``source``, ``push``) need not necessarily be disjoint.
-    For this to work, we need to avoid interference between snapshotting and pruning of the different sending jobs.
-    The solution is to centralize sender-side snapshot management in a separate ``snap`` job.
-    Snapshotting in the ``source``/``push`` job should then be disabled (``type: manual``).
-    And sender-side pruning (``keep_sender``) needs to be disabled in the active side (``pull`` / ``push``), since that'll be done by the ``snap job``.
-
-  * **Restore limitations**: when restoring from one of the ``pull`` targets (e.g., using ``zfs send -R``), the replication cursor bookmarks don't exist on the restored system.
-    This can break incremental replication to all other receive-sides after restore.
-
-  * See :ref:`the fan-out replication quick-start guide <quickstart-fan-out-replication>` for an example of this setup.
-
-
-**Setups that do not work**:
-
-* N ``pull`` identities, 1 ``source`` job. Tracking :issue:`380`.
-

docs/configuration/preface.rst

@@ -0,0 +1,39 @@
+
+.. _configuration_preface:
+
+=======
+Preface
+=======
+
+-----------------------
+Configuration File Path
+-----------------------
+
+zrepl searches for its main configuration file in the following locations (in that order):
+
+* If set, the location specified via the global ``--config`` flag
+* ``/etc/zrepl/zrepl.yml``
+* ``/usr/local/etc/zrepl/zrepl.yml``
+
+The examples in the :ref:`tutorial` or the ``cmd/sampleconf`` directory should provide a good starting point.
+
+-------------------
+Runtime Directories
+-------------------
+
+zrepl requires runtime directories for various UNIX sockets --- they are documented in the :ref:`config file<conf-runtime-directories>`.
+Your package maintainer / init script should take care of creating them.
+Alternatively, for default settings, the following should to the trick.
+
+::
+
+    mkdir -p /var/run/zrepl/stdinserver
+    chmod -R 0700 /var/run/zrepl
+
+
+----------
+Validating
+----------
+
+The config can be validated using the ``zrepl test`` subcommand.
+

docs/configuration/prune.rst

@@ -3,282 +3,82 @@
 Pruning Policies
 ================
 
-In zrepl, *pruning* means *destroying snapshots*.
+In zrepl, *pruning* means *destroying filesystem versions by some policy* where filesystem versions are bookmarks and snapshots.
-Pruning must happen on both sides of a replication or the systems would inevitably run out of disk space at some point.
 
-Typically, the requirements to temporal resolution and maximum retention time differ per side.
+A *pruning policy* takes a list of filesystem versions and decides for each whether it should be kept or destroyed.
-For example, when using zrepl to back up a busy database server, you will want high temporal resolution (snapshots every 10 min) for the last 24h in case of administrative disasters, but cannot afford to store them for much longer because you might have high turnover volume in the database.
-On the receiving side, you may have more disk space available, or need to comply with other backup retention policies.
 
-zrepl uses a set of  **keep rules** per sending and receiving side to determine which snapshots shall be kept per filesystem.
+The job context defines which snapshots are even considered for pruning, for example through the ``snapshot_prefix`` variable.
-**A snapshot that is not kept by any rule is destroyed.**
+Check the respective :ref:`job definition <job>` for details.
-The keep rules are **evaluated on the active side** (:ref:`push <job-push>` or :ref:`pull job <job-pull>`) of the replication setup, for both active and passive side, after replication completed or was determined to have failed permanently.
 
+Currently, the :ref:`prune-retention-grid` is the only supported pruning policy.
 
+.. TIP::
 
-Example Configuration:
+    You can perform a dry-run of a job's pruning policy using the ``zrepl test`` subcommand.
+
+.. _prune-retention-grid:
+
+Retention Grid
+--------------
 
 ::
 
     jobs:
-     - type: push
+    - name: pull_app-srv
-       name: ...
+      type: pull
-       connect: ...
-       filesystems: {
-         "<": true,
-         "tmp": false
-       }
-       snapshotting:
-         type: periodic
-         prefix: zrepl_
-         interval: 10m
-       pruning:
-         keep_sender:
-           - type: not_replicated
-           # make sure manually created snapshots by the administrator are kept
-           - type: regex
-             regex: "^manual_.*"
-           - type: grid
-             grid: 1x1h(keep=all) | 24x1h | 14x1d
-             regex: "^zrepl_.*"
-         keep_receiver:
-           - type: grid
-             grid: 1x1h(keep=all) | 24x1h | 35x1d | 6x30d
-             regex: "^zrepl_.*"
-           # manually created snapshots will be kept forever on receiver
-           - type: regex
-             regex: "^manual_.*"
-
-.. DANGER::
-
-    You might have **existing snapshots** of filesystems affected by pruning which you want to keep, i.e. not be destroyed by zrepl.
-    Make sure to actually add the necessary ``regex`` keep rules on both sides, like with ``manual`` in the example above.
-
-.. _prune-keep-not-replicated:
-
-Policy ``not_replicated``
--------------------------
-::
-
-   jobs:
-   - type: push
-     pruning:
-       keep_sender:
-       - type: not_replicated
       ...
+      prune:
+        policy: grid
+        grid: 1x1h(keep=all) | 24x1h | 35x1d | 6x30d
+                │               │
+                └─ one hour interval
+                                │
+                                └─ 24 adjacent one-hour intervals
 
-``not_replicated`` keeps all snapshots that have not been replicated to the receiving side.
+    - name: pull_backup
-It only makes sense to specify this rule for the ``keep_sender``.
+      type: source
-The reason is that, by definition, all snapshots on the receiver have already been replicated to there from the sender.
+      interval: 10m
-To determine whether a sender-side snapshot has already been replicated, zrepl uses the :ref:`replication cursor bookmark <replication-cursor-and-last-received-hold>` which corresponds to the most recent successfully replicated snapshot.
+      prune:
+        policy: grid
+        grid: 1x1d(keep=all)
+        keep_bookmarks: 144
+
+
+The retention grid can be thought of as a time-based sieve:
+The ``grid`` field specifies a list of adjacent time intervals:
+the left edge of the leftmost (first) interval is the ``creation`` date of the youngest snapshot.
+All intervals to its right describe time intervals further in the past.
+
+Each interval carries a maximum number of snapshots to keep.
+It is secified via ``(keep=N)``, where ``N`` is either ``all`` (all snapshots are kept) or a positive integer.
+The default value is **1**.
+
+Bookmarks are not affected by the above.
+Instead, the ``keep_bookmarks`` field specifies the number of bookmarks to be kept per filesystem.
+You only need to specify ``keep_bookmarks`` at the source-side of a replication setup since the destination side does not receive bookmarks.
+You can specify ``all`` as a value to keep all bookmarks, but be warned that you should install some other way to prune unneeded ones then (see below).
+
+The following procedure happens during pruning:
+
+#. The list of snapshots eligible for pruning is sorted by ``creation``
+#. The left edge of the first interval is aligned to the ``creation`` date of the youngest snapshot
+#. A list of buckets is created, one for each interval
+#. The list of snapshots is split up into the buckets.
+#. For each bucket
+
+   #. the contained snapshot list is sorted by creation.
+   #. snapshots from the list, oldest first, are destroyed until the specified ``keep`` count is reached.
+   #. all remaining snapshots on the list are kept.
+#. The list of bookmarks eligible for pruning is sorted by ``createtxg`` and the most recent ``keep_bookmarks`` bookmarks are kept.
+
+.. _replication-downtime:
 
 .. ATTENTION::
 
-    If you use ``not_replicated`` in your pruning rules, make sure to :ref:`monitor <monitoring>` replication.
+    Be aware that ``keep_bookmarks x interval`` (interval of the job level) controls the **maximum allowable replication downtime** between source and destination.
-    If your replication gets stuck then ``not_replicated`` causes snapshots to pile uip on the sender.
+    If replication does not work for whatever reason, source and destination will eventually run out of sync because the source will continue pruning snapshots.
-    ZFS and especially the ``zfs`` management command are known to degrade in performance with a lot of snapshots.
+    The only recovery in that case is full replication, which may not always be viable due to disk space or traffic constraints.
-    Such degradation impacts zrepl, any other scripts, and human ability to manage your zpool.
 
+    Further note that while bookmarks consume a constant amount of disk space, listing them requires temporary dynamic **kernel memory** proportional to the number of bookmarks.
+    Thus, do not use ``all`` or an inappropriately high value without good reason.
 
-.. _prune-keep-retention-grid:
-
-Policy ``grid``
----------------
-
-::
-
-    jobs:
-    - type: pull
-      pruning:
-        keep_receiver:
-        - type: grid
-          regex: "^zrepl_.*"
-          grid: 1x1h(keep=all) | 24x1h | 35x1d | 6x30d
-                │                │               │
-                └─ 1 repetition of a one-hour interval with keep=all
-                                 │               │
-                                 └─ 24 repetitions of a one-hour interval with keep=1
-                                                 │
-                                                 └─ 6 repetitions of a 30-day interval with keep=1
-      ...
-
-The retention grid can be thought of as a time-based sieve that thins out snapshots as they get older.
-
-The ``grid`` field specifies a list of adjacent time intervals.
-Each interval is a bucket with a maximum capacity of ``keep`` snapshots.
-The following procedure happens during pruning:
-
-#. The list of snapshots is filtered by the regular expression in ``regex``.
-   Only snapshots names that match the regex are considered for this rule, all others will be pruned unless another rule keeps them.
-#. The snapshots that match ``regex`` are placed onto a time axis according to their ``creation`` date.
-   The youngest snapshot is on the left, the oldest on the right.
-#. The first buckets are placed "under" that axis so that the ``grid`` spec's first bucket's left edge aligns with youngest snapshot.
-#. All subsequent buckets are placed adjacent to their predecessor bucket.
-#. Now each snapshot on the axis either falls into one bucket or it is older than our rightmost bucket.
-   Buckets are left-inclusive and right-exclusive which means that a snapshot on the edge of bucket will always 'fall into the right one'.
-#. Snapshots older than the rightmost bucket are **not kept** by the grid specification.
-#. For each bucket, we only keep the ``keep`` oldest snapshots.
-
-The syntax to describe the bucket list is as follows:
-
-::
-
-     Repeat x Duration (keep=all)
-
-* The **duration** specifies the length of the interval.
-* The **keep** count specifies the number of snapshots that fit into the bucket.
-  It can be either a positive integer or ``all`` (all snapshots are kept).
-* The **repeat** count repeats the bucket definition for the specified number of times.
-
-**Example**:
-
-::
-
-   Assume the following grid specification:
-
-      grid: 1x1h(keep=all) | 2x2h | 1x3h
-
-   This grid specification produces the following constellation of buckets:
-
-   0h        1h        2h        3h        4h        5h        6h        7h        8h        9h
-   |         |         |         |         |         |         |         |         |         |
-   |-Bucket1-|-----Bucket2-------|------Bucket3------|-----------Bucket4-----------|
-   | keep=all|      keep=1       |       keep=1      |            keep=1           |
-
-
-
-   Now assume that we have a set of snapshots @a, @b, ..., @D.
-   Snapshot @a is the most recent snapshot.
-   Snapshot @D is the oldest snapshot, it is almost 9 hours older than snapshot @a.
-   We place the snapshots on the same timeline as the buckets:
-
-
-   0h        1h        2h        3h        4h        5h        6h        7h        8h        9h
-   |         |         |         |         |         |         |         |         |         |
-   |-Bucket1-|-----Bucket2-------|------Bucket3------|-----------Bucket4-----------|
-   | keep=all|      keep=1       |       keep=1      |            keep=1           |
-   |         |                   |                   |                             |
-   | a  b  c | d  e  f  g  h  i  j  k  l  m  n  o  p |q  r  s  t  u  v  w  x  y  z |A  B  C  D
-
-   We obtain the following mapping of snapshots to buckets:
-
-   Bucket1:   a,b,c
-   Bucket2:   d,e,f,g,h,i
-   Bucket3:   j,k,l,m,n,o,p
-   Bucket4:   q,r,s,t,u,v,w,x,y,z
-   No bucket: A,B,C,D
-
-   For each bucket, we now prune snapshots until it only contains `keep` snapshots.
-   Newer snapshots are destroyed first.
-   Snapshots that do not fall into a bucket are always destroyed.
-
-   Result after pruning:
-
-   0h        1h        2h        3h        4h        5h        6h        7h        8h        9h
-   |         |         |         |         |         |         |         |         |         |
-   |-Bucket1-|-----Bucket2-------|------Bucket3------|-----------Bucket4-----------|
-   |         |                   |                   |                             |
-   | a  b  c |                i  |                 p |                           z |
-
-.. _prune-keep-last-n:
-
-Policy ``last_n``
------------------
-
-::
-
-   jobs:
-     - type: push
-       pruning:
-         keep_receiver:
-         - type: last_n
-           count: 10
-           regex: ^zrepl_.*$ # optional
-     ...
-
-``last_n`` filters the snapshot list by ``regex``, then keeps the last ``count`` snapshots in that list (last = youngest = most recent creation date)
-All snapshots that don't match ``regex`` or exceed ``count`` in the filtered list are destroyed unless matched by other rules.
-
-.. _prune-keep-regex:
-
-Policy ``regex``
-----------------
-
-::
-
-   jobs:
-     - type: push
-       pruning:
-         keep_receiver:
-         # keep all snapshots with prefix zrepl_ or manual_
-         - type: regex
-           regex: "^(zrepl|manual)_.*"
-
-     - type: push
-       snapshotting:
-         prefix: zrepl_
-       pruning:
-         keep_sender:
-         # keep all snapshots that were not created by zrepl
-         - type: regex
-           negate: true
-           regex: "^zrepl_.*"
-
-``regex`` keeps all snapshots whose names are matched by the regular expression in ``regex``.
-Like all other regular expression fields in prune policies, zrepl uses Go's `regexp.Regexp <https://golang.org/pkg/regexp/#Compile>`_ Perl-compatible regular expressions (`Syntax <https://golang.org/pkg/regexp/syntax>`_).
-The optional `negate` boolean field inverts the semantics: Use it if you want to keep all snapshots that *do not* match the given regex.
-
-.. _prune-workaround-source-side-pruning:
-
-Source-side snapshot pruning
-----------------------------
-
-A :ref:`source jobs<job-source>` takes snapshots on the system it runs on.
-The corresponding :ref:`pull job <job-pull>` on the replication target connects to the source job and replicates the snapshots.
-Afterwards, the pull job coordinates pruning on both sender (the source job side) and receiver (the pull job side).
-
-There is no built-in way to define and execute pruning on the source side independently of the pull side.
-The source job will continue taking snapshots which will not be pruned until the pull side connects.
-This means that **extended replication downtime will fill up the source's zpool with snapshots**.
-
-If the above is a conceivable situation for you, consider using :ref:`push mode <job-push>`, where pruning happens on the same side where snapshots are taken.
-
-Workaround using ``snap`` job
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-As a workaround (see GitHub :issue:`102` for development progress), a pruning-only :ref:`snap job <job-snap>` can be defined on the source side:
-The snap job is in charge of snapshot creation & destruction, whereas the source job's role is reduced to just serving snapshots.
-However, since, jobs are run independently, it is possible that the snap job will prune snapshots that are queued for replication / destruction by the remote pull job that connects to the source job.
-Symptoms of such race conditions are spurious replication and destroy errors.
-
-Example configuration:
-
-::
-
-  # source side
-  jobs:
-  - type: snap
-    snapshotting:
-      type: periodic
-    pruning:
-      keep:
-        # source side pruning rules go here
-    ...
-
-  - type: source
-    snapshotting:
-      type: manual
-    root_fs: ...
-
-  # pull side
-  jobs:
-  - type: pull
-    pruning:
-      keep_sender:
-        # let the source-side snap job do the pruning
-        - type: regex
-          regex: ".*"
-        ...
-      keep_receiver:
-        # feel free to prune on the pull side as desired
-        ...

docs/configuration/replication.rst

@@ -1,75 +0,0 @@
-.. include:: ../global.rst.inc
-
-.. _replication-options:
-
-Replication Options
-===================
-
-
-::
-   
-   jobs:
-   - type: push
-     filesystems: ...
-     replication:
-       protection:
-         initial:     guarantee_resumability # guarantee_{resumability,incremental,nothing}
-         incremental: guarantee_resumability # guarantee_{resumability,incremental,nothing}
-       concurrency:
-         size_estimates: 4
-         steps: 1
-
-     ...
-
-.. _replication-option-protection:
-
-``protection`` option
---------------------------
-
-The ``protection`` variable controls the degree to which a replicated filesystem is protected from getting out of sync through a zrepl pruner or external tools that destroy snapshots.
-zrepl can guarantee :ref:`resumability <step-holds>` or just :ref:`incremental replication <replication-cursor-and-last-received-hold>`.
-
-``guarantee_resumability`` is the **default** value and guarantees that a replication step is always resumable and that incremental replication will always be possible.
-The implementation uses replication cursors, last-received-hold and step holds.
-
-``guarantee_incremental`` only guarantees that incremental replication will always be possible.
-If a step ``from -> to`` is interrupted and its `to` snapshot is destroyed, zrepl will remove the half-received ``to``'s resume state and start a new step ``from -> to2``.
-The implementation uses replication cursors, tentative replication cursors and last-received-hold.
-
-``guarantee_nothing`` does not make any guarantees with regards to keeping sending and receiving side in sync.
-No bookmarks or holds are created to protect sender and receiver from diverging.
-
-**Tradeoffs**
-
-Using ``guarantee_incremental`` instead of ``guarantee_resumability`` obviously removes the resumability guarantee.
-This means that replication progress is no longer monotonic which might lead to a replication setup that never makes progress if mid-step interruptions are too frequent (e.g. frequent network outages).
-However, the advantage and :issue:`reason for existence <288>` of the ``incremental`` mode is that it allows the pruner to delete snapshots of interrupted replication steps
-which is useful if replication happens so rarely (or fails so frequently) that the amount of disk space exclusively referenced by the step's snapshots becomes intolerable.
-
-.. NOTE::
-
-   When changing this flag, obsoleted zrepl-managed bookmarks and holds will be destroyed on the next replication step that is attempted for each filesystem.
-
-
-.. _replication-option-concurrency:
-
-``concurrency`` option
-----------------------
-
-The ``concurrency`` options control the maximum amount of concurrency during replication.
-The default values allow some concurrency during size estimation but no parallelism for the actual replication.
-
-* ``concurrency.steps`` (default = 1) controls the maximum number of concurrently executed :ref:`replication steps <overview-how-replication-works>`.
-  The planning step for each file system is counted as a single step.
-* ``concurrency.size_estimates`` (default = 4) controls the maximum number of concurrent step size estimations done by the job.
-
-Note that initial replication cannot start replicating child filesystems before the parent filesystem's initial replication step has completed.
-
-Some notes on tuning these values:
-
-* Disk: Size estimation is less I/O intensive than step execution because it does not need to access the data blocks.
-* CPU: Size estimation is usually a dense CPU burst whereas step execution CPU utilization is stretched out over time because of disk IO.
-  Faster disks, sending a compressed dataset in :ref:`plain mode <zfs-background-knowledge-plain-vs-raw-sends>` and the zrepl transport mode all contribute to higher CPU requirements.
-* Network bandwidth: Size estimation does not consume meaningful amounts of bandwidth, step execution does.
-* :ref:`zrepl ZFS abstractions <zrepl-zfs-abstractions>`: for each replication step zrepl needs to update its ZFS abstractions through the ``zfs`` command which often waits multiple seconds for the zpool to sync.
-  Thus, if the actual send & recv time of a step is small compared to the time spent on zrepl ZFS abstractions then increasing step execution concurrency will result in a lower overall turnaround time.

docs/configuration/sendrecvoptions.rst

@@ -1,287 +0,0 @@
-.. include:: ../global.rst.inc
-
-
-Send & Recv Options
-===================
-
-.. _job-send-options:
-
-Send Options
-~~~~~~~~~~~~
-
-:ref:`Source<job-source>` and :ref:`push<job-push>` jobs have an optional ``send`` configuration section.
-
-::
-
-   jobs:
-   - type: push
-     filesystems: ...
-     send:
-       # flags from the table below go here
-     ...
-
-The following table specifies the list of (boolean) options.
-Flags with an entry in the ``zfs send`` column map directly to the zfs send CLI flags.
-zrepl does not perform feature checks for these flags.
-If you enable a flag that is not supported by the installed version of ZFS, the zfs error will show up at runtime in the logs and zrepl status.
-See the `upstream man page <https://openzfs.github.io/openzfs-docs/man/8/zfs-send.8.html>`_ (``man zfs-send``) for their semantics.
-
-.. list-table::
-    :widths: 20 10 70
-    :header-rows: 1
-
-    * - ``send.``
-      - ``zfs send``
-      - Comment
-    * - ``encrypted``
-      -
-      - Specific to zrepl, :ref:`see below <job-send-options-encrypted>`.
-    * - ``bandwidth_limit``
-      -
-      - Specific to zrepl, :ref:`see below <job-send-recv-options--bandwidth-limit>`.
-    * - ``raw``
-      - ``-w``
-      - Use ``encrypted`` to only allow encrypted sends. Mixed sends are not supported.
-    * - ``send_properties``
-      - ``-p``
-      - **Be careful**, read the :ref:`note on property replication below <job-note-property-replication>`.
-    * - ``backup_properties``
-      - ``-b``
-      - **Be careful**, read the :ref:`note on property replication below <job-note-property-replication>`.
-    * - ``large_blocks``
-      - ``-L``
-      - **Potential data loss on OpenZFS < 2.0**, see :ref:`warning below <job-send-options-large-blocks>`.
-    * - ``compressed``
-      - ``-c``
-      -
-    * - ``embedded_data``
-      - ``-e``
-      -
-    * - ``saved``
-      - ``-S``
-      -
-
-.. _job-send-options-encrypted:
-
-``encrypted``
--------------
-
-The ``encrypted`` option controls whether the matched filesystems are sent as `OpenZFS native encryption <http://open-zfs.org/wiki/ZFS-Native_Encryption>`_ raw sends.
-More specifically, if ``encrypted=true``, zrepl
-
-* checks for any of the filesystems matched by ``filesystems`` whether the ZFS ``encryption`` property indicates that the filesystem is actually encrypted with ZFS native encryption and
-* invokes the ``zfs send`` subcommand with the ``-w`` option (raw sends) and
-* expects the receiving side to support OpenZFS native encryption (recv will fail otherwise)
-
-Filesystems matched by ``filesystems`` that are not encrypted are not sent and will cause error log messages.
-
-If ``encrypted=false``, zrepl expects that filesystems matching ``filesystems`` are not encrypted or have loaded encryption keys.
-
-.. NOTE::
-
-   Use ``encrypted`` instead of ``raw`` to make your intent clear that zrepl must only replicate filesystems that are actually encrypted by OpenZFS native encryption.
-   It is meant as a safeguard to prevent unintended sends of unencrypted filesystems in raw mode.
-
-.. _job-send-options-properties:
-
-``send_properties``
--------------------
-Sends the dataset properties along with snapshots.
-Please be careful with this option and read the :ref:`note on property replication below <job-note-property-replication>`.
-
-.. _job-send-options-backup-properties:
-
-``backup_properties``
----------------------
-
-When properties are modified on a filesystem that was received from a send stream with ``send.properties=true``, ZFS archives the original received value internally.
-This also applies to :ref:`inheriting or overriding properties during zfs receive <job-recv-options--inherit-and-override>`.
-
-When sending those received filesystems another hop, the ``backup_properties`` flag instructs ZFS to send the original property values rather than the current locally set values.
-
-This is useful for replicating properties across multiple levels of backup machines.
-**Example:**
-Suppose we want to flow snapshots from Machine A to B, then from B to C.
-A will enable the :ref:`properties send option <job-send-options-properties>`.
-B will want to override :ref:`critical properties such as mountpoint or canmount <job-note-property-replication>`.
-But the job that replicates from B to C should be sending the original property values received from A.
-Thus, B sets the ``backup_properties`` option.
-
-Please be careful with this option and read the :ref:`note on property replication below <job-note-property-replication>`.
-
-.. _job-send-options-large-blocks:
-
-``large_blocks``
-----------------
-
-This flag should not be changed after initial replication.
-Prior to `OpenZFS commit 7bcb7f08 <https://github.com/openzfs/zfs/pull/10383/files#diff-4c1e47568f46fb63546e984943b09a3e6b051e2242649523f7835bbdfe2a9110R337-R342>`_
-it was possible to change this setting which resulted in **data loss on the receiver**.
-The commit in question is included in OpenZFS 2.0 and works around the problem by prohibiting receives of incremental streams with a flipped setting.
-
-.. WARNING::
-
-   This bug has **not been fixed in the OpenZFS 0.8 releases** which means that changing this flag after initial replication might cause **data loss** on the receiver.
-
-.. _job-recv-options:
-
-Recv Options
-~~~~~~~~~~~~
-
-:ref:`Sink<job-sink>` and :ref:`pull<job-pull>` jobs have an optional ``recv`` configuration section:
-
-::
-
-   jobs:
-   - type: pull
-     recv:
-       properties:
-         inherit:
-           - "mountpoint"
-         override: {
-           "org.openzfs.systemd:ignore": "on"
-         }
-       bandwidth_limit: ...
-       placeholder:
-         encryption: unspecified | off | inherit
-     ...
-
-Jump to
-:ref:`properties <job-recv-options--inherit-and-override>` ,
-:ref:`bandwidth_limit <job-send-recv-options--bandwidth-limit>` , and
-:ref:`placeholder <job-recv-options--placeholder>`.
-
-.. _job-recv-options--inherit-and-override:
-
-``properties``
---------------
-
-``override`` maps directly to the `zfs recv -o flag <https://openzfs.github.io/openzfs-docs/man/8/zfs-recv.8.html>`_.
-Property name-value pairs specified in this map will apply to all received filesystems, regardless of whether the send stream contains properties or not.
-
-``inherit`` maps directly to the `zfs recv -x flag <https://openzfs.github.io/openzfs-docs/man/8/zfs-recv.8.html>`_.
-Property names specified in this list will be inherited from the receiving side's parent filesystem (e.g. ``root_fs``).
-
-With both options, the sending side's property value is still stored on the receiver, but the local override or inherit is the one that takes effect.
-You can send the original properties from the first receiver to another receiver using :ref:`send.backup_properties<job-send-options-backup-properties>`.
-
-
-.. _job-note-property-replication:
-
-A Note on Property Replication
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-If a send stream contains properties, as per ``send.send_properties`` or ``send.backup_properties``,
-the default ZFS behavior is to use those properties on the receiving side, verbatim.
-
-In many use cases for zrepl, this can have devastating consequences.
-For example, when backing up a filesystem that has ``mountpoint=/`` to a storage server,
-that storage server's root filesystem will be shadowed by the received file system on some platforms.
-Also, many scripts and tools use ZFS user properties for configuration and do not check the property source (``local`` vs. ``received``).
-If they are installed on the receiving side as well as the sending side, property replication could have unintended effects.
-
-**zrepl currently does not provide any automatic safe-guards for property replication:**
-
-* Make sure to read the entire man page on zfs recv (`man zfs recv <https://openzfs.github.io/openzfs-docs/man/8/zfs-recv.8.html>`_) before enabling this feature.
-* Use ``recv.properties.override`` whenever possible, e.g. for ``mountpoint=none`` or ``canmount=off``.
-* Use ``recv.properties.inherit`` if that makes more sense to you.
-
-Below is an **non-exhaustive list of problematic properties**.
-Please open a pull request if you find a property that is missing from this list.
-(Both with regards to core ZFS tools and other software in the broader ecosystem.)
-
-Mount behaviour
----------------
-
-* ``mountpoint``
-* ``canmount``
-* ``overlay``
-
-Note: Before `OpenZFS 2.0.5 <https://github.com/openzfs/zfs/issues/11416>`_, inheriting or overriding the ``mountpoint`` property on ZVOLs fails in ``zfs recv``.
-If you are on such an older version, consider creating separate zrepl jobs for your ZVOL and filesystem datasets.
-
-Systemd
--------
-
-With systemd, you should also consider the properties processed by the `zfs-mount-generator <https://manpages.debian.org/buster-backports/zfsutils-linux/zfs-mount-generator.8.en.html>`_ .
-
-Most notably:
-
-* ``org.openzfs.systemd:ignore``
-* ``org.openzfs.systemd:wanted-by``
-* ``org.openzfs.systemd:required-by``
-
-Encryption
-----------
-
-If the sender filesystems are encrypted but the sender does :ref:`plain sends <zfs-background-knowledge-plain-vs-raw-sends>`
-and property replication is enabled, the receiver must :ref:`inherit the following properties<job-recv-options--inherit-and-override>`:
-
-* ``keylocation``
-* ``keyformat``
-* ``encryption``
-
-Sharing
--------
-
-You may not want the replicated filesystem shared in the same way as the source is.
-
-* ``sharenfs``
-* ``sharesmb``
-
-.. _job-recv-options--placeholder:
-
-Placeholders
-~~~~~~~~~~~~
-
-::
-
-   placeholder:
-     encryption: unspecified | off | inherit
-
-During replication, zrepl :ref:`creates placeholder datasets <replication-placeholder-property>` on the receiving side if the sending side's ``filesystems`` filter creates gaps in the dataset hierarchy.
-This is generally fully transparent to the user.
-However, with OpenZFS Native Encryption, placeholders require zrepl user attention.
-Specifically, the problem is that, when zrepl attempts to create the placeholder dataset on the receiver, and that placeholder's parent dataset is encrypted, ZFS wants to inherit encryption to the placeholder.
-This is relevant to two use cases that zrepl supports:
-
-1. **encrypted-send-to-untrusted-receiver** In this use case, the sender sends an :ref:`encrypted send stream <job-send-options-encrypted>` and the receiver doesn't have the key loaded.
-2. **send-plain-encrypt-on-receive** The receive-side ``root_fs`` dataset is encrypted, and the senders are unencrypted.
-   The key of ``root_fs`` is loaded, and the goal is that the plain sends (e.g., from production) are encrypted on-the-fly during receive, with ``root_fs``'s key.
-
-For **encrypted-send-to-untrusted-receiver**, the placeholder datasets need to be created with ``-o encryption=off``.
-Without it, creation would fail with an error, indicating that the placeholder's parent dataset's key needs to be loaded.
-But we don't trust the receiver, so we can't expect that to ever happen.
-
-However, for **send-plain-encrypt-on-receive**, we cannot set ``-o encryption=off``.
-The reason is that if we did, any of the (non-placeholder) child datasets below the placeholder would inherit ``encryption=off``, thereby silently breaking our encrypt-on-receive use case.
-So, to cover this use case, we need to create placeholders without specifying ``-o encryption``.
-This will make ``zfs create`` inherit the encryption mode from the parent dataset, and thereby transitively from ``root_fs``.
-
-The zrepl config provides the `recv.placeholder.encryption` knob to control this behavior.
-In ``undefined`` mode (default), placeholder creation bails out and asks the user to configure a behavior.
-In ``off`` mode, the placeholder is created with ``encryption=off``, i.e., **encrypted-send-to-untrusted-rceiver** use case.
-In ``inherit`` mode, the placeholder is created without specifying ``-o encryption`` at all, i.e., the **send-plain-encrypt-on-receive** use case.
-
-
-Common Options
-~~~~~~~~~~~~~~
-
-.. _job-send-recv-options--bandwidth-limit:
-
-Bandwidth Limit (send & recv)
------------------------------
-
-::
-
-   bandwidth_limit:
-     max: 23.5 MiB # -1 is the default and disabled rate limiting
-     bucket_capacity: # token bucket capacity in bytes; defaults to 128KiB
-
-Both ``send`` and ``recv`` can be limited to a maximum bandwidth through ``bandwidth_limit``.
-For most users, it should be sufficient to just set ``bandwidth_limit.max``.
-The ``bandwidth_limit.bucket_capacity`` refers to the `token bucket size <https://github.com/juju/ratelimit>`_.
-
-The bandwidth limit only applies to the payload data, i.e., the ZFS send stream.
-It does not account for transport protocol overheads.
-The scope is the job level, i.e., all :ref:`concurrent <replication-option-concurrency>` sends or incoming receives of a job share the bandwidth limit.

docs/configuration/snapshotting.rst

@@ -1,301 +0,0 @@
-.. include:: ../global.rst.inc
-
-.. _job-snapshotting-spec:
-
-Taking Snaphots
-===============
-
-You can configure zrepl to take snapshots of the filesystems in the ``filesystems`` field specified in ``push``, ``source`` and ``snap`` jobs.
-
-The following snapshotting types are supported:
-
-
-.. list-table::
-    :widths: 20 70
-    :header-rows: 1
-
-    * - ``snapshotting.type``
-      - Comment
-    * - ``periodic``
-      - Ensure that snapshots are taken at a particular interval.
-    * - ``cron``
-      - Use cron spec to take snapshots at particular points in time.
-    * - ``manual``
-      - zrepl does not take any snapshots by itself.
-
-The ``periodic`` and ``cron`` snapshotting types share some common options and behavior:
-
-* **Naming:** The snapshot names are composed of a user-defined ``prefix`` followed by a UTC date formatted like ``20060102_150405_000`` by default.
-  We either use UTC or timestamp with timezone information because it will avoid name conflicts when switching time zones or between summer and winter time.
-  Note that if timestamp with time zone information is used, the "+" of the timezone (i.e. +02:00) is replaced by "_" (i.e. _02:00) to conform to allowed characters in ZFS snapshots.
-* **Hooks:** You can configure hooks to run before or after zrepl takes the snapshots. See :ref:`below <job-snapshotting-hooks>` for details.
-* **Push replication:** After creating all snapshots, the snapshotter will wake up the replication part of the job, if it's a ``push`` job.
-  Note that snapshotting is decoupled from replication, i.e., if it is down or takes too long, snapshots will still be taken.
-  Note further that other jobs are not woken up by snapshotting.
-
-.. NOTE::
-
-   There is **no concept of ownership** of the snapshots that are created by ``periodic`` or ``cron``.
-   Thus, there is no distinction between zrepl-created snapshots and user-created snapshots during replication or pruning.
-
-   In particular, pruning will take all snapshots into consideration by default.
-   To constrain pruning to just zrepl-created snapshots:
-
-     1. Assign a unique `prefix` to the snapshotter and
-     2. Use the ``regex`` functionality of the various pruning ``keep`` rules to just consider snapshots with that prefix.
-
-  There is currently no way to constrain replication to just zrepl-created snapshots.
-  Follow and comment at :issue:`403` if you need this functionality.
-
-.. NOTE::
-
-  The ``zrepl signal wakeup JOB`` subcommand does not trigger snapshotting.
-
-``periodic`` Snapshotting
--------------------------
-
-::
-
-   jobs:
-   - ...
-     filesystems: { ... }
-     snapshotting:
-       type: periodic
-       prefix: zrepl_
-       interval: 10m
-       # Timestamp format that is used as snapshot suffix.
-       # Can be any of "dense" (default), "human", "iso-8601", "unix-seconds" or a custom Go time format (see https://go.dev/src/time/format.go)
-       timestamp_format: dense
-       # Specifies in which time zone the snapshot suffix is generated, optional, defaults to UTC, used only if time zone information is part of timestamp_format.
-       # Can be "UTC" (default), "Local" (time zone information from the OS) or any of the IANA Time Zone names (see https://nodatime.org/TimeZones)
-       timestamp_location: UTC
-       hooks: ...
-    pruning: ...
-
-The ``periodic`` snapshotter ensures that snapshots are taken in the specified ``interval``.
-If you use zrepl for backup, this translates into your recovery point objective (RPO).
-To meet your RPO, you still need to monitor that replication, which happens asynchronously to snapshotting, actually works.
-
-It is desirable to get all ``filesystems`` snapshotted simultaneously because it results in a more consistent backup.
-To accomplish this while still maintaining the ``interval``, the ``periodic`` snapshotter attempts to get the snapshotting rhythms in sync.
-To find that sync point, the most recent snapshot, created by the snapshotter, in any of the matched ``filesystems`` is used.
-A filesystem that does not have snapshots by the snapshotter has lower priority than filesystem that do, and thus might not be snapshotted (and replicated) until it is snapshotted at the next sync point.
-The snapshotter uses the ``prefix`` to identify which snapshots it created.
-
-.. _job-snapshotting--cron:
-
-``cron`` Snapshotting
----------------------
-
-::
-
-   jobs:
-   - type: snap
-     filesystems: { ... }
-     snapshotting:
-       type: cron
-       prefix: zrepl_
-       # (second, optional) minute hour day-of-month month day-of-week
-       # This example takes snapshots daily at 3:00.
-       cron: "0 3 * * *"
-       # Timestamp format that is used as snapshot suffix.
-       # Can be any of "dense" (default), "human", "iso-8601", "unix-seconds" or a custom Go time format (see https://go.dev/src/time/format.go)
-       timestamp_format: dense
-       # Specifies in which time zone the snapshot suffix is generated, optional, defaults to UTC, used only if time zone information is part of timestamp_format.
-       # Can be "UTC" (default), "Local" (time zone information from the OS) or any of the IANA Time Zone names (see https://nodatime.org/TimeZones)
-       timestamp_location: UTC
-     pruning: ...
-
-In ``cron`` mode, the snapshotter takes snaphots at fixed points in time.
-See https://en.wikipedia.org/wiki/Cron for details on the syntax.
-zrepl uses the ``the github.com/robfig/cron/v3`` Go package for parsing.
-An optional field for "seconds" is supported to take snapshots at sub-minute frequencies.
-
-
-``manual`` Snapshotting
------------------------
-
-::
-
-   jobs:
-   - type: push
-     snapshotting:
-       type: manual
-     ...
-
-In ``manual`` mode, zrepl does not take snapshots by itself.
-Manual snapshotting is most useful if you have existing infrastructure for snapshot management.
-Or, if you want to decouple snapshot management from replication using a zrepl ``snap`` job.
-See :ref:`this quickstart guide <quickstart-backup-to-external-disk>` for an example.
-
-To trigger replication after taking snapshots, use the ``zrepl signal wakeup JOB`` command.
-
-.. _job-snapshotting-timestamp_format:
-
-Timestamp Format
-----------------
-
-The ``cron`` and ``periodic`` snapshotter support configuring a custom timestamp format that is used as suffix for the snapshot name.
-
-By default, the current time is converted to UTC and then formatted using the ``dense`` format.
-
-Both time zone and format can be customized by setting ``timestamp_format`` and/or ``timestamp_location``.
-
-``timestamp_location`` is fed verbatim into `time.LoadLocation <https://pkg.go.dev/time#LoadLocation>`_.
-
- ``timestamp_format`` supports the following values (case-insensitive).
-
-* ``dense`` => Go format string ``20060102_150405_000``
-* ``human`` => Go format string ``2006-01-02_15:04:05``
-* ``iso-8601`` => Go format string ``2006-01-02T15:04:05.000Z07:00``
-* ``unix-seconds`` => Unix seconds since the epoch, formatted as a decimal string (`time.Time.Unix <https://pkg.go.dev/time#Time.Unix>`_)
-* Any custom Go time format accepted by `time.Time#Format <https://go.dev/src/time/format.go>`_.
-
-The ``dense``, ``human``, and ``unix-seconds`` formats require the ``timestamp_location`` to be ``UTC`` because they do not contain timezone information.
-
-The ``+`` character is `not allowed in ZFS snapshot names <https://github.com/openzfs/zfs/blob/1d3ba0bf01020f5459b1c28db3979129088924c0/module/zcommon/zfs_namecheck.c#L334-L351>`_.
-Format strings are disallowed to contain the ``+`` character (it's not a character that is interpreted by ``time.Time.Format``, so there's no reason to use it).
-If a format _produces_ a ``+``, that ``+`` is replaced by an `_`` character.
-
-.. _job-snapshotting-hooks:
-
-Pre- and Post-Snapshot Hooks
-----------------------------
-
-Jobs with `periodic snapshots <job-snapshotting-spec_>`_ can run hooks before and/or after taking the snapshot specified in ``snapshotting.hooks``:
-Hooks are called per filesystem before and after the snapshot is taken (pre- and post-edge).
-Pre-edge invocations are in configuration order, post-edge invocations in reverse order, i.e. like a stack.
-If a pre-snapshot invocation fails, ``err_is_fatal=true`` cuts off subsequent hooks, does not take a snapshot, and only invokes post-edges corresponding to previous successful pre-edges.
-``err_is_fatal=false`` logs the failed pre-edge invocation but does not affect subsequent hooks nor snapshotting itself.
-Post-edges are only invoked for hooks whose pre-edges ran without error.
-Note that hook failures for one filesystem never affect other filesystems.
-
-The optional ``timeout`` parameter specifies a period after which zrepl will kill the hook process and report an error.
-The default is 30 seconds and may be specified in any units understood by `time.ParseDuration <https://golang.org/pkg/time/#ParseDuration>`_.
-
-The optional ``filesystems`` filter which limits the filesystems the hook runs for. This uses the same |filter-spec| as jobs.
-
-Most hook types take additional parameters, please refer to the respective subsections below.
-
-.. list-table::
-    :widths: 20 10 70
-    :header-rows: 1
-
-    * - Hook ``type``
-      - Details
-      - Description
-    * - ``command``
-      - :ref:`Details <job-hook-type-command>`
-      - Arbitrary pre- and post snapshot scripts.
-    * - ``postgres-checkpoint``
-      - :ref:`Details <job-hook-type-postgres-checkpoint>`
-      - Execute Postgres ``CHECKPOINT`` SQL command before snapshot.
-    * - ``mysql-lock-tables``
-      - :ref:`Details <job-hook-type-mysql-lock-tables>`
-      - Flush and read-Lock MySQL tables while taking the snapshot.
-
-.. _job-hook-type-command:
-
-``command`` Hooks
-~~~~~~~~~~~~~~~~~
-
-::
-
-
-    jobs:
-    - type: push
-      filesystems: {
-        "<": true,
-        "tmp": false
-      }
-      snapshotting:
-        type: periodic
-        prefix: zrepl_
-        interval: 10m
-        hooks:
-        - type: command
-          path: /etc/zrepl/hooks/zrepl-notify.sh
-          timeout: 30s
-          err_is_fatal: false
-        - type: command
-          path: /etc/zrepl/hooks/special-snapshot.sh
-          filesystems: {
-            "tank/special": true
-          }
-      ...
-
-
-``command`` hooks take a ``path`` to an executable script or binary to be executed before and after the snapshot.
-``path`` must be absolute (e.g. ``/etc/zrepl/hooks/zrepl-notify.sh``).
-No arguments may be specified; create a wrapper script if zrepl must call an executable that requires arguments.
-The process standard output is logged at level INFO. Standard error is logged at level WARN.
-The following environment variables are set:
-
-* ``ZREPL_HOOKTYPE``: either "pre_snapshot" or "post_snapshot"
-* ``ZREPL_FS``: the ZFS filesystem name being snapshotted
-* ``ZREPL_SNAPNAME``: the zrepl-generated snapshot name (e.g. ``zrepl_20380119_031407_000``)
-* ``ZREPL_DRYRUN``: set to ``"true"`` if a dry run is in progress so scripts can print, but not run, their commands
-
-An empty template hook can be found in :sampleconf:`/hooks/template.sh`.
-
-.. _job-hook-type-postgres-checkpoint:
-
-``postgres-checkpoint`` Hook
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Connects to a Postgres server and executes the ``CHECKPOINT`` statement pre-snapshot.
-Checkpointing applies the WAL contents to all data files and syncs the data files to disk.
-This is not required for a consistent database backup: it merely forward-pays the "cost" of WAL replay to the time of snapshotting instead of at restore.
-However, the Postgres manual recommends against checkpointing during normal operation.
-Further, the operation requires Postgres superuser privileges.
-zrepl users must decide on their own whether this hook is useful for them (it likely isn't).
-
-.. ATTENTION::
-    Note that WALs and Postgres data directory (with all database data files) must be on the same filesystem to guarantee a correct point-in-time backup with the ZFS snapshot.
-
-DSN syntax documented here: `<https://godoc.org/github.com/lib/pq>`_
-
-.. code-block:: sql
-
-   CREATE USER zrepl_checkpoint PASSWORD yourpasswordhere;
-   ALTER ROLE zrepl_checkpoint SUPERUSER;
-
-.. code-block:: yaml
-
-  - type: postgres-checkpoint
-    dsn: "host=localhost port=5432 user=postgres password=yourpasswordhere sslmode=disable"
-    filesystems: {
-        "p1/postgres/data11": true
-    }
-
-.. _job-hook-type-mysql-lock-tables:
-
-``mysql-lock-tables`` Hook
-~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-Connects to MySQL and executes
-
-* pre-snapshot ``FLUSH TABLES WITH READ LOCK`` to lock all tables in all databases in the MySQL server we connect to (`docs <https://dev.mysql.com/doc/refman/8.0/en/flush.html#flush-tables-with-read-lock>`_)
-* post-snapshot ``UNLOCK TABLES``  reverse above operation.
-
-Above procedure is documented in the `MySQL manual <https://dev.mysql.com/doc/mysql-backup-excerpt/5.7/en/backup-methods.html>`_
-as a means to produce a consistent backup of a MySQL DBMS installation (i.e., all databases).
-
-`DSN syntax <https://github.com/go-sql-driver/mysql#dsn-data-source-name>`_: ``[username[:password]@][protocol[(address)]]/dbname[?param1=value1&...&paramN=valueN]``
-
-.. ATTENTION::
-    All MySQL databases must be on the same ZFS filesystem to guarantee a consistent point-in-time backup with the ZFS snapshot.
-
-.. code-block:: sql
-
-   CREATE USER zrepl_lock_tables IDENTIFIED BY 'yourpasswordhere';
-   GRANT RELOAD ON *.* TO zrepl_lock_tables;
-   FLUSH PRIVILEGES;
-
-.. code-block:: yaml
-
-  - type: mysql-lock-tables
-    dsn: "zrepl_lock_tables:yourpasswordhere@tcp(localhost)/"
-    filesystems: {
-      "tank/mysql": true
-    }

docs/configuration/transports.rst

@@ -5,280 +5,42 @@
 Transports
 ==========
 
-The zrepl RPC layer uses **transports** to establish a single, bidirectional data stream between an active and passive job.
+A transport provides an authenticated `io.ReadWriteCloser <https://golang.org/pkg/io/#ReadWriteCloser>`_ to the RPC layer.
-On the passive (serving) side, the transport also provides the **client identity** to the upper layers:
+(An ``io.ReadWriteCloser`` is essentially a bidirectional reliable communication channel.)
-this string is used for access control and separation of filesystem sub-trees in :ref:`sink jobs <job-sink>`.
-Transports are specified in the ``connect`` or ``serve`` section of a job definition.
-
-.. contents::
-
-.. ATTENTION::
-
-    The **client identities must be valid ZFS dataset path components**
-    because the :ref:`sink job <job-sink>` uses ``${root_fs}/${client_identity}`` to determine the client's subtree.
-
-.. _transport-tcp:
-
-``tcp`` Transport
------------------
-
-The ``tcp`` transport uses plain TCP, which means that the data is **not encrypted** on the wire.
-Clients are identified by their IPv4 or IPv6 addresses, and the client identity is established through a mapping on the server.
-
-This transport may also be used in conjunction with network-layer encryption and/or VPN tunnels to provide encryption on the wire.
-To make the IP-based client authentication effective, such solutions should provide authenticated IP addresses.
-Some options to consider:
-
-.. _transport-tcp-tunneling:
-
-* `WireGuard <https://www.wireguard.com/>`_: Linux-focussed, in-kernel TLS
-* `OpenVPN <https://openvpn.net/>`_: Cross-platform VPN, uses tun on \*nix
-* `IPSec <https://en.wikipedia.org/wiki/IPsec>`_: Properly standardized, in-kernel network-layer VPN
-* `spiped <http://www.tarsnap.com/spiped.html>`_: think of it as an encrypted pipe between two servers
-* SSH
-
-  * `sshuttle <https://sshuttle.readthedocs.io/en/stable/overview.html>`_: VPN-like solution, but using SSH
-  * `SSH port forwarding <https://help.ubuntu.com/community/SSH/OpenSSH/PortForwarding>`_: Systemd user unit & make it start before the zrepl service.
-
-Serve
-~~~~~
-
-::
-
-    jobs:
-    - type: sink
-      serve:
-        type: tcp
-        listen: ":8888"
-        listen_freebind: true # optional, default false
-        clients: {
-          "192.168.122.123" :               "mysql01",
-          "192.168.122.42" :                "mx01",
-          "2001:0db8:85a3::8a2e:0370:7334": "gateway",
-
-          # CIDR masks require a '*' in the client identity string
-          # that is expanded to the client's IP address
-
-          "10.23.42.0/24":       "cluster-*"
-          "fde4:8dba:82e1::/64": "san-*"
-        }
-      ...
-
-.. _listen-freebind-explanation:
-
-``listen_freebind`` controls whether the socket is allowed to bind to non-local or unconfigured IP addresses (Linux ``IP_FREEBIND`` , FreeBSD ``IP_BINDANY``).
-Enable this option if you want to ``listen`` on a specific IP address that might not yet be configured when the zrepl daemon starts.
-
-Connect
-~~~~~~~
-
-::
-
-    jobs:
-     - type: push
-       connect:
-         type: tcp
-         address: "10.23.42.23:8888"
-         dial_timeout: # optional, default 10s
-       ...
-
-.. _transport-tcp+tlsclientauth:
-
-``tls`` Transport
------------------
-
-The ``tls`` transport uses TCP + TLS with client authentication using client certificates.
-The client identity is the common name (CN) presented in the client certificate.
-
-It is recommended to set up a dedicated CA infrastructure for this transport, e.g. using OpenVPN's `EasyRSA <https://github.com/OpenVPN/easy-rsa>`_.
-For a simple 2-machine setup, mutual TLS might also be sufficient.
-We provide :ref:`copy-pastable instructions to generate the certificates below <transport-tcp+tlsclientauth-certgen>`.
-
-The implementation uses `Go's TLS library <https://golang.org/pkg/crypto/tls/>`_.
-Since Go binaries are statically linked, you or your distribution need to recompile zrepl when vulnerabilities in that library are disclosed.
-
-All file paths are resolved relative to the zrepl daemon's working directory.
-Specify absolute paths if you are unsure what directory that is (or find out from your init system).
-
-If intermediate CAs are used, the **full chain** must be present in either in the ``ca`` file or the individual ``cert`` files.
-Regardless, the client's certificate must be first in the ``cert`` file, with each following certificate directly certifying the one preceding it (see `TLS's specification <https://tools.ietf.org/html/rfc5246#section-7.4.2>`_).
-This is the common default when using a CA management tool.
-
-.. NOTE::
-
-   As of Go 1.15 (zrepl 0.3.0 and newer), the Go TLS / x509 library **requrires Subject Alternative Names**
-   be present in certificates. You might need to re-generate your certificates using one of the :ref:`two alternatives
-   provided below<transport-tcp+tlsclientauth-certgen>`.
-
-   Note further that zrepl continues to use the CommonName field to assign client identities.
-   Hence, we recommend to keep the Subject Alternative Name and the CommonName in sync.
-
-
-Serve
-~~~~~
-
-::
-
-    jobs:
-      - type: sink
-        root_fs: "pool2/backup_laptops"
-        serve:
-          type: tls
-          listen: ":8888"
-          listen_freebind: true # optional, default false
-          ca:   /etc/zrepl/ca.crt
-          cert: /etc/zrepl/prod.fullchain
-          key:  /etc/zrepl/prod.key
-          client_cns:
-            - "laptop1"
-            - "homeserver"
-
-The ``ca`` field specified the certificate authority used to validate client certificates.
-The ``client_cns`` list specifies a list of accepted client common names (which are also the client identities for this transport).
-The ``listen_freebind`` field is :ref:`explained here <listen-freebind-explanation>`.
-
-Connect
-~~~~~~~
-
-::
-
-    jobs:
-    - type: pull
-      connect:
-        type: tls
-        address: "server1.foo.bar:8888"
-        ca:   /etc/zrepl/ca.crt
-        cert: /etc/zrepl/backupserver.fullchain
-        key:  /etc/zrepl/backupserver.key
-        server_cn: "server1"
-        dial_timeout: # optional, default 10s
-
-The ``ca`` field specifies the CA which signed the server's certificate (``serve.cert``).
-The ``server_cn`` specifies the expected common name (CN) of the server's certificate.
-It overrides the hostname specified in ``address``.
-The connection fails if either do not match.
-
-.. _transport-tcp+tlsclientauth-certgen:
-
-.. _transport-tcp+tlsclientauth-2machineopenssl:
-
-Mutual-TLS between Two Machines
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-For a two-machine setup, self-signed certificates distributed using an out-of-band mechanism will also work just fine:
-
-Suppose you have a push-mode setup, with `backups.example.com` running the :ref:`sink job <job-sink>`, and `prod.example.com` running the :ref:`push job <job-push>`.
-Run the following OpenSSL commands on each host, substituting HOSTNAME in both filenames and the interactive input prompt by OpenSSL:
-
-.. code-block:: bash
-
-   (name=HOSTNAME; openssl req -x509 -sha256 -nodes \
-    -newkey rsa:4096 \
-    -days 365 \
-    -keyout $name.key \
-    -out $name.crt -addext "subjectAltName = DNS:$name" -subj "/CN=$name")
-
-Now copy each machine's ``HOSTNAME.crt`` to the other machine's ``/etc/zrepl/HOSTNAME.crt``, for example using `scp`.
-The serve & connect configuration will thus look like the following:
-
-::
-
-   # on backups.example.com
-   - type: sink
-     serve:
-       type: tls
-       listen: ":8888"
-       ca: "/etc/zrepl/prod.example.com.crt"
-       cert: "/etc/zrepl/backups.example.com.crt"
-       key: "/etc/zrepl/backups.example.com.key"
-       client_cns:
-         - "prod.example.com"
-     ...
-
-   # on prod.example.com
-   - type: push
-     connect:
-       type: tls
-       address:"backups.example.com:8888"
-       ca: /etc/zrepl/backups.example.com.crt
-       cert: /etc/zrepl/prod.example.com.crt
-       key:  /etc/zrepl/prod.example.com.key
-       server_cn: "backups.example.com"
-     ...
-
-
-Certificate Authority using EasyRSA
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-For more than two machines, it might make sense to set up a CA infrastructure.
-Tools like `EasyRSA <https://github.com/OpenVPN/easy-rsa>`_ make this very easy:
-
-::
-
-     #!/usr/bin/env bash
-     set -euo pipefail
-
-     HOSTS=(backupserver prod1 prod2 prod3 10.23.42.1)
-
-     curl -L https://github.com/OpenVPN/easy-rsa/releases/download/v3.2.5/EasyRSA-3.2.5.tgz > EasyRSA-3.2.5.tgz
-     echo "662ee3b453155aeb1dff7096ec052cd83176c460cfa82ac130ef8568ec4df490  EasyRSA-3.2.5.tgz" | sha256sum -c
-     rm -rf EasyRSA-3.2.5
-     tar -xf EasyRSA-3.2.5.tgz
-     cd EasyRSA-3.2.5
-     ./easyrsa --batch
-     ./easyrsa --batch init-pki
-     ./easyrsa --batch build-ca nopass
-
-     for host in "${HOSTS[@]}"; do
-         ./easyrsa --batch --auto-san build-serverClient-full $host nopass
-         echo cert for host $host available at pki/issued/$host.crt
-         echo key for host $host available at pki/private/$host.key
-     done
-     echo ca cert available at pki/ca.crt
 
+Currently, only the ``ssh+stdinserver`` transport is supported.
 
 .. _transport-ssh+stdinserver:
 
 ``ssh+stdinserver`` Transport
 -----------------------------
 
-``ssh+stdinserver`` uses the ``ssh`` command and some features of the server-side SSH ``authorized_keys`` file.
+The way the ``ssh+stdinserver`` transport works is inspired by `git shell <https://git-scm.com/docs/git-shell>`_ and `Borg Backup <https://borgbackup.readthedocs.io/en/stable/deployment.html>`_.
-It is less efficient than other transports because the data passes through two more pipes.
+It is provided by the Go package ``github.com/problame/go-netssh``.
-However, it is fairly convenient to set up and allows the zrepl daemon to not be directly exposed to the internet, because all traffic passes through the system's SSH server.
+The config excerpts are taken from the :ref:`tutorial` which you should complete before reading further.
-
-The concept is inspired by `git shell <https://git-scm.com/docs/git-shell>`_ and `Borg Backup <https://borgbackup.readthedocs.io/en/stable/deployment.html>`_.
-The implementation is provided by the Go package ``github.com/problame/go-netssh``.
-
-.. NOTE::
-
-   ``ssh+stdinserver`` generally provides inferior error detection and handling compared to the ``tcp`` and ``tls`` transports.
-   When encountering such problems, consider using  ``tcp`` or ``tls`` transports, or help improve package go-netssh.
 
 .. _transport-ssh+stdinserver-serve:
 
-Serve
+Serve Mode
-~~~~~
+~~~~~~~~~~
 
 ::
 
     jobs:
-    - type: source
+    - name: pull_backup
+      type: source
       serve:
         type: stdinserver
-        client_identities:
+        client_identity: backup-srv.example.com
-        - "client1"
-        - "client2"
       ...
 
-First of all, note that ``type=stdinserver`` in this case:
+The serving job opens a UNIX socket named after ``client_identity`` in the runtime directory, e.g. ``/var/run/zrepl/stdinserver/backup-srv.example.com``.
-Currently, only ``connect.type=ssh+stdinserver`` can connect to a ``serve.type=stdinserver``, but we want to keep that option open for future extensions.
 
-The serving job opens a UNIX socket named after ``client_identity`` in the runtime directory.
+On the same machine, the ``zrepl stdinserver $client_identity`` command connects to that socket.
-In our example above, that is ``/var/run/zrepl/stdinserver/client1`` and ``/var/run/zrepl/stdinserver/client2``.
+For example, ``zrepl stdinserver backup-srv.example.com`` connects to the UNIX socket ``/var/run/zrepl/stdinserver/backup-srv.example.com``.
 
-On the same machine, the ``zrepl stdinserver $client_identity`` command connects to ``/var/run/zrepl/stdinserver/$client_identity``.
 It then passes its stdin and stdout file descriptors to the zrepl daemon via *cmsg(3)*.
-zrepl daemon in turn combines them into an object implementing ``net.Conn``:
+zrepl daemon in turn combines them into an ``io.ReadWriteCloser``:
 a ``Write()`` turns into a write to stdout, a ``Read()`` turns into a read from stdin.
 
 Interactive use of the ``stdinserver`` subcommand does not make much sense.
@@ -292,8 +54,8 @@ This can be achieved with an entry in the ``authorized_keys`` file of the servin
     # for older OpenSSH versions
     command="zrepl stdinserver CLIENT_IDENTITY",no-port-forwarding,no-X11-forwarding,no-pty,no-agent-forwarding,no-user-rc CLIENT_SSH_KEY
 
-* CLIENT_IDENTITY is substituted with an entry from ``client_identities`` in our example
+* CLIENT_IDENTITY is substituted with ``backup-srv.example.com`` in our example
-* CLIENT_SSH_KEY is substituted with the public part of the SSH keypair specified in the ``connect.identity_file`` directive on the connecting host.
+* CLIENT_SSH_KEY is substituted with the public part of the SSH keypair specified in the ``connect`` directive on the connecting host.
 
 .. NOTE::
 
@@ -302,30 +64,29 @@ This can be achieved with an entry in the ``authorized_keys`` file of the servin
 
 To recap, this is of how client authentication works with the ``ssh+stdinserver`` transport:
 
-* Connections to the ``/var/run/zrepl/stdinserver/${client_identity}`` UNIX socket are blindly trusted by zrepl daemon.
+* Connections to the ``client_identity`` UNIX socket are blindly trusted by zrepl daemon.
-  The connection client identity is the name of the socket, i.e. ``${client_identity}``.
+* Thus, the runtime directory must be private to the zrepl user (checked by zrepl daemon)
-* Thus, the runtime directory must be private to the zrepl user (this is checked by zrepl daemon)
 * The admin of the host with the serving zrepl daemon controls the ``authorized_keys`` file.
 * Thus, the administrator controls the mapping ``PUBKEY -> CLIENT_IDENTITY``.
 
 .. _transport-ssh+stdinserver-connect:
 
-Connect
+Connect Mode
-~~~~~~~
+~~~~~~~~~~~~
 
 ::
 
     jobs:
-    - type: pull
+    - name: pull_app-srv
+      type: pull
       connect:
         type: ssh+stdinserver
-        host: prod.example.com
+        host: app-srv.example.com
         user: root
         port: 22
         identity_file: /etc/zrepl/ssh/identity
-        # options: # optional, default [], `-o` arguments passed to ssh
+        options: # optional
-        # - "Compression=yes"
+        - "Compression=on"
-        # dial_timeout: 10s # optional, default 10s, max time.Duration until initial handshake is completed
 
 The connecting zrepl daemon
 
@@ -340,47 +101,12 @@ The connecting zrepl daemon
       #. The remote user, host and port correspond to those configured.
       #. Further options can be specified using the ``options`` field, which appends each entry in the list to the command line using ``-o $entry``.
 
-#. Wraps the pipe ends in a ``net.Conn`` and returns it to the RPC layer.
+1. Wraps the pipe ends in an ``io.ReadWriteCloser`` and uses it for RPC.
 
 As discussed in the section above, the connecting zrepl daemon expects that ``zrepl stdinserver $client_identity`` is  executed automatically via an ``authorized_keys`` file entry.
 
-The ``known_hosts`` file used by the ssh command must contain an entry for ``connect.host`` prior to starting zrepl.
-Thus, run the following on the pulling host's command line (substituting ``connect.host``):
-
-::
-
-    ssh -i /etc/zrepl/ssh/identity root@prod.example.com
-
 .. NOTE::
 
     The environment variables of the underlying SSH process are cleared. ``$SSH_AUTH_SOCK`` will not be available.
     It is suggested to create a separate, unencrypted SSH key solely for that purpose.
 
-
-.. _transport-local:
-
-``local`` Transport
--------------------
-
-The local transport can be used to implement :ref:`local replication <replication-local>`, i.e., push replication between a push and sink job defined in the same configuration file.
-
-The ``listener_name`` is analogous to a hostname and must match between ``serve`` and ``connect``.
-The ``client_identity`` is used by the sink as documented above.
-
-::
-
-    jobs:
-    - type: sink
-      serve:
-        type: local
-        listener_name: localsink
-      ...
-
-    - type: push
-      connect:
-        type: local
-        listener_name: localsink
-        client_identity: local_backup
-        dial_timeout: 2s # optional, 0 for no timeout
-      ...
-

docs/future.rst

@@ -1,33 +0,0 @@
-.. include:: global.rst.inc
-
-.. _future:
-
-Future
-======
-
-This page contains some notes about future plans for zrepl.
-Then again, Christian only has very limited time available for zrepl maintenance these days.
-So, don't count on any of this happening in the near future.
-
-One big development that has happened in recent years is this fork of zrepl: https://github.com/dsh2dsh/zrepl
-We should figure out whether we can pull in features from there.
-
-The next major step in terms of new feature development for zrepl would be to revise snapshot management:
-
-- Make it easy to decouple snapshot management (snapshotting, pruning) from replication.
-- Ability to include/exclude snapshots from replication.
-  This is useful for aforementioned decoupling, e.g., separate snapshot prefixes for local & remote replication.
-  Also, it makes explicit that by default, zrepl replicates all snapshots, and that
-  replication has no concept of "zrepl-created snapshots", which is a common misconception.
-- Use of ``zfs snapshot`` comma syntax or channel programs to take snapshots of multiple
-  datasets atomically.
-- Provide an alternative to the ``grid`` pruning policy.
-  Most likely something based on hourly/daily/weekly/monthly "trains" plus a count.
-- Ability to prune at the granularity of the **group** of snapshots created at a given
-  time, as opposed to the individual snapshots within a dataset.
-  Maybe this will be addressed by the alternative to the ``grid`` pruning policy,
-  as it will likely be more predictable.
-
-Those changes will likely come with some breakage in the config.
-However, I want to avoid breaking **use cases** that are satisfied by the current design.
-There will be beta/RC releases to give users a chance to evaluate.

docs/global.rst.inc

@@ -1,37 +1,3 @@
 .. global list of substitutions
 
-.. 
+.. |mainconfig| replace:: :ref:`main configuration file <mainconfigfile>`
-  .. |macroname| replace:: macro expansion
-
-.. |GitHub license| image:: https://img.shields.io/github/license/zrepl/zrepl.svg
-   :target: https://github.com/zrepl/zrepl/blob/master/LICENSE
-.. |Language: Go| image:: https://img.shields.io/badge/lang-Go-6ad7e5.svg
-   :target: https://golang.org/
-.. |User Docs| image:: https://img.shields.io/badge/docs-web-blue.svg
-   :target: https://zrepl.github.io
-.. |Donate via PayPal| image:: https://img.shields.io/badge/donate-paypal-yellow.svg
-   :target: https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=R5QSXJVYHGX96
-.. |Donate via Liberapay| image:: https://img.shields.io/liberapay/patrons/zrepl.svg?logo=liberapay
-   :target: https://liberapay.com/zrepl/donate
-.. |Donate via Patreon| image:: https://img.shields.io/badge/dynamic/json?color=yellow&label=Patreon&query=data.attributes.patron_count&url=https%3A%2F%2Fwww.patreon.com%2Fapi%2Fcampaigns%2F3095079
-   :target: https://www.patreon.com/zrepl
-.. |Donate via GitHub Sponsors| image:: https://img.shields.io/static/v1?label=Sponsor&message=%E2%9D%A4&logo=GitHub&style=flat&color=yellow
-   :target: https://github.com/sponsors/problame
-.. |Twitter| image:: https://img.shields.io/twitter/url/https/github.com/zrepl/zrepl.svg?style=social
-   :target: https://twitter.com/intent/tweet?text=Wow:&url=https%3A%2F%2Fgithub.com%2Fzrepl%2Fzrepl
-.. |Matrix| image:: https://img.shields.io/badge/chat-matrix-blue.svg
-   :target: https://matrix.to/#/#zrepl:matrix.org
-
-.. |serve-transport| replace:: :ref:`serve specification<transport>`
-.. |connect-transport| replace:: :ref:`connect specification<transport>`
-.. |send-options| replace:: :ref:`send options<job-send-options>`, e.g. for encrypted sends
-.. |recv-options| replace:: :ref:`recv options<job-recv-options>`
-.. |replication-options| replace:: :ref:`replication options<replication-options>`
-.. |conflict-resolution-options| replace:: :ref:`conflict resolution options<conflict_resolution-options>`
-.. |snapshotting-spec| replace:: :ref:`snapshotting specification <job-snapshotting-spec>`
-.. |pruning-spec| replace:: :ref:`pruning specification <prune>`
-.. |filter-spec| replace:: :ref:`filter specification<pattern-filter>`
-
-.. |br| raw:: html
-
-   <br/>

docs/implementation.rst

@@ -0,0 +1,62 @@
+.. _implementation_toc:
+
+Implementation Overview
+=======================
+
+.. WARNING::
+
+    Incomplete / under construction
+
+The following design aspects may convince you that ``zrepl`` is superior to a hacked-together shell script solution.
+
+Testability & Performance
+-------------------------
+
+zrepl is written in Go, a real programming language with type safety,
+reasonable performance, testing infrastructure and an (opinionated) idea of
+software engineering.
+
+* key parts & algorithms of zrepl are covered by unit tests (work in progress)
+* zrepl is noticably faster than comparable shell scripts
+
+
+RPC protocol
+------------
+
+While it is tempting to just issue a few ``ssh remote 'zfs send ...' | zfs recv``, this has a number of drawbacks:
+
+* The snapshot streams need to be compatible.
+* Communication is still unidirectional. Thus, you will most likely
+
+  * either not take advantage of features such as *compressed send & recv*
+  * or issue additional ``ssh`` commands in advance to figure out what features are supported on the other side.
+
+* Advanced logic in shell scripts is ugly to read, poorly testable and a pain to maintain.
+
+zrepl takes a different approach:
+
+* Define an RPC protocol.
+* Establish an encrypted, authenticated, bidirectional communication channel...
+* ... with zrepl running at both ends of it.
+
+This has several obvious benefits:
+
+* No blank root shell access is given to the other side.
+* Instead, an *authenticated* peer can *request* filesystem lists, snapshot streams, etc.
+* Requests are then checked against job-specific ACLs, limiting a client to the filesystems it is actually allowed to replicate.
+* The :ref:`transport mechanism <transport>` is decoupled from the remaining logic, keeping it extensible.
+
+Protocol Implementation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+zrepl implements its own RPC protocol.
+This is mostly due to the fact that existing solutions do not provide efficient means to transport large amounts of data.
+
+Package `github.com/zrepl/zrepl/rpc <https://github.com/zrepl/zrepl/tree/master/rpc>`_ builds a special-case handling around returning an ``io.Reader`` as part of a unary RPC call.
+
+Measurements show only a single memory-to-memory copy of a snapshot stream is made using ``github.com/zrepl/zrepl/rpc``, and there is still potential for further optimizations.
+
+Logging & Transparency
+----------------------
+
+zrepl comes with :ref:`rich, structured and configurable logging <logging>`, allowing administators to understand what the software is actually doing.

docs/index.rst

@@ -5,114 +5,65 @@
 
 .. include:: global.rst.inc
 
-|GitHub license| |Language: Go| |Twitter| |Donate via Patreon| |Donate via GitHub Sponsors| |Donate via Liberapay| |Donate via PayPal| |Matrix|
-
-
 zrepl - ZFS replication
 -----------------------
 
-**zrepl** is a one-stop, integrated solution for ZFS replication.
+.. ATTENTION::
-
+    zrepl as well as this documentation is still under active development.
-.. raw:: html
+    It is neither feature complete nor is there a stability guarantee on the configuration format.
-
+    Use & test at your own risk ;)
-   <div style="margin-bottom: 1em; background: #2e3436; color: white; font-size: 0.8em; min-height: 6em; max-width: 100%; overflow: auto;">
-     <pre>
-     Job: prod_to_backups
-     Type: push
-     Replication:
-         Attempt #1
-         Status: fan-out-filesystems
-         Progress: [=========================\----] 246.7 MiB / 264.7 MiB @ 11.5 MiB/s
-           zroot              STEPPING (step 1/2, 624 B/1.2 KiB) next: @a => @b
-           zroot/ROOT         DONE (step 2/2, 1.2 KiB/1.2 KiB)
-         * zroot/ROOT/default STEPPING (step 1/2, 123.4 MiB/129.3 MiB) next: @a => @b
-           zroot/tmp          STEPPING (step 1/2, 29.9 KiB/44.2 KiB) next: @a => @b
-           zroot/usr          STEPPING (step 1/2, 624 B/1.2 KiB) next: @a => @b
-         * zroot/usr/home     STEPPING (step 1/2, 123.3 MiB/135.3 MiB) next: @a => @b
-           zroot/var          STEPPING (step 1/2, 624 B/1.2 KiB) next: @a => @b
-           zroot/var/audit    DONE (step 2/2, 1.2 KiB/1.2 KiB)
-           zroot/var/crash    DONE (step 2/2, 1.2 KiB/1.2 KiB)
-           zroot/var/log      STEPPING (step 1/2, 22.0 KiB/29.2 KiB) next: @a => @b
-           zroot/var/mail     STEPPING (step 1/2, 624 B/1.2 KiB) next: @a => @b
-     Pruning Sender:
-         ...
-     Pruning Receiver:
-     </pre>
-   </div>
-
 
 Getting started
 ~~~~~~~~~~~~~~~
 
-The :ref:`10 minute quick-start guides <quickstart-toc>` give you a first impression.
+The :ref:`5 minute tutorial setup <tutorial>` gives you a first impression.
 
 Main Features
 ~~~~~~~~~~~~~
 
-* **Filesystem replication**
+* Filesystem Replication
 
-  * [x] Pull & Push mode
+  * [x] Local & Remote
-  * [x] Multiple :ref:`transport modes <transport>`: TCP, TCP + TLS client auth, SSH
+  * [x] Pull mode
+  * [ ] Push mode
+  * [x] Access control checks when pulling datasets
+  * [x] :ref:`Flexible mapping <pattern-mapping>` rules
+  * [x] Bookmarks support
+  * [ ] Feature-negotiation for
 
-  * Advanced replication features
+    * Resumable `send & receive`
+    * Compressed `send & receive`
+    * Raw encrypted `send & receive` (as soon as it is available)
 
-    * [x] Automatic retries for temporary network errors
+* Automatic snapshot creation
-    * [x] Automatic resumable send & receive
-    * [x] Automatic ZFS holds during send & receive
-    * [x] Automatic bookmark \& hold management for guaranteed incremental send & recv 
-    * [x] Encrypted raw send & receive to untrusted receivers (OpenZFS native encryption)
-    * [x] Properties send & receive
-    * [x] Compressed send & receive
-    * [x] Large blocks send & receive
-    * [x] Embedded data send & receive
-    * [x] Resume state send & receive
-    * [x] Bandwidth limiting
 
-* **Automatic snapshot management**
+  * [x] Ensure fixed time interval between snapshots
 
-  * [x] Periodic :ref:`filesystem snapshots <job-snapshotting-spec>`
+* Automatic snapshot :ref:`pruning <prune>`
-  * [x] Support for :ref:`pre- and post-snapshot hooks <job-snapshotting-hooks>` with builtins for MySQL & Postgres
-  * [x] Flexible :ref:`pruning rule system <prune>`
 
   * [x] Age-based fading (grandfathering scheme)
-    * [x] Bookmarks to avoid divergence between sender and receiver
 
-* **Sophisticated Monitoring & Logging**
+* Flexible, detailed & structured :ref:`logging <logging>`
 
-  * [x] Live progress reporting via `zrepl status` :ref:`subcommand <usage>`
+  * [x] ``human``, ``logfmt`` and ``json`` formatting
-  * [x] Comprehensive, structured :ref:`logging <logging>`
+  * [x] stdout, syslog and TCP (+TLS client auth) outlets
 
-    * ``human``, ``logfmt`` and ``json`` formatting
+* Maintainable implementation in Go
-    * stdout, syslog and TCP (+TLS client auth) outlets
-
-  * [x] Prometheus :ref:`monitoring <monitoring>` endpoint
-
-* **Maintainable implementation in Go**
 
   * [x] Cross platform
-  * [x] Dynamic feature checking
   * [x] Type safe & testable code
 
-
-.. ATTENTION::
-    zrepl as well as this documentation is still under active development.
-    There is no stability guarantee on the RPC protocol or configuration format,
-    but we do our best to document breaking changes in the :ref:`changelog`.
-
-
 Contributing
 ~~~~~~~~~~~~
 
 We are happy about any help we can get!
 
-* :ref:`Financial Support <supporters>`
 * Explore the codebase
 
   * These docs live in the ``docs/`` subdirectory
 
 * Document any non-obvious / confusing / plain broken behavior you encounter when setting up zrepl for the first time
-* Check the *Issues* and *Projects* sections for things to do.
+* Check the *Issues* and *Projects* sections for things to do
-  The `good first issues <https://github.com/zrepl/zrepl/labels/good%20first%20issue>`_ and `docs <https://github.com/zrepl/zrepl/labels/docs>`_ are suitable starting points.
 
 .. admonition:: Development Workflow
     :class: note
@@ -121,8 +72,6 @@ We are happy about any help we can get!
     Make sure to read the `Developer Documentation section <https://github.com/zrepl/zrepl>`_ and open new issues or pull requests there.
 
 
-
-
 Table of Contents
 ~~~~~~~~~~~~~~~~~
 
@@ -130,14 +79,11 @@ Table of Contents
    :maxdepth: 2
    :caption: Contents:
 
-   quickstart
+   tutorial
    installation
    configuration
    usage
-   pr
+   implementation
    changelog
-   future
    GitHub Repository & Issue Tracker <https://github.com/zrepl/zrepl>
-   Chat: Matrix <https://matrix.to/#/#zrepl:matrix.org>
+   pr
-   supporters
-

docs/installation.rst

@@ -1,19 +1,88 @@
+.. _binary releases: https://github.com/zrepl/zrepl/releases
 
-.. _installation_toc:
+.. _installation:
 
-************
 Installation
-************
+============
 
 .. TIP::
 
-    Note: check out the :ref:`quick-start guides <quickstart-toc>` if you want a first impression of zrepl.
+    Note: check out the :ref:`tutorial` if you want a first impression of zrepl.
 
-.. toctree::
+User Privileges
+---------------
+
+It is possible to run zrepl as an unprivileged user in combination with
+`ZFS delegation <https://www.freebsd.org/doc/handbook/zfs-zfs-allow.html>`_.
+Also, there is the possibility to run it in a jail on FreeBSD by delegating a dataset to the jail.
+However, until we get around documenting those setups, you will have to run zrepl as root or experiment yourself :)
+
+Packages
+--------
+
+zrepl source releases are signed & tagged by the author in the git repository.
+Your OS vendor may provide binary packages of zrepl through the package manager.
+Additionally, `binary releases`_ are provided on GitHub.
+The following list may be incomplete, feel free to submit a PR with an update:
+
+.. list-table::
+    :header-rows: 1
+
+    * - OS / Distro
+      - Install Command
+      - Link
+    * - FreeBSD
+      - ``pkg install zrepl``
+      - `<https://www.freshports.org/sysutils/zrepl/>`_
+    * - Others
+      -
+      - Use `binary releases`_ or build from source.
+
+Compile From Source
+~~~~~~~~~~~~~~~~~~~
+
+Producing a release requires **Go 1.9** or newer and **Python 3** + **pip3** + ``docs/requirements.txt`` for the Sphinx documentation.
+A tutorial to install Go is available over at `golang.org <https://golang.org/doc/install>`_.
+Python and pip3 should probably be installed via your distro's package manager.
+
+Alternatively, you can use the Docker build process:
+it is used to produce the official zrepl `binary releases`_
+and serves as a reference for build dependencies and procedure:
+
+::
+
+    git clone https://github.com/zrepl/zrepl.git
+    cd zrepl
+    sudo docker build -t zrepl_build -f build.Dockerfile .
+    sudo docker run -it --rm \
+        -v "${PWD}:/go/src/github.com/zrepl/zrepl" \
+        --user "$(id -u):$(id -g)" \
+        zrepl_build make vendordeps release
+
+Alternatively, you can install build dependencies on your local system and then build in your ``$GOPATH``:
+
+::
+
+    mkdir -p "${GOPATH}/src/github.com/zrepl/zrepl"
+    git clone https://github.com/zrepl/zrepl.git "${GOPATH}/src/github.com/zrepl/zrepl"
+    cd "${GOPATH}/src/github.com/zrepl/zrepl"
+    ./lazy.sh devsetup
+    make vendordeps release
+
+Build results are located in the ``artifacts/`` directory.
+
+.. NOTE::
+
+    It is your job to install the apropriate binary in the zrepl users's ``$PATH``, e.g. ``/usr/local/bin/zrepl``.
+    Otherwise, the examples in the :ref:`tutorial` may need to be adjusted.
+
+.. _mainconfigfile:
+
+What next?
+----------
+
+Read the :ref:`configuration chapter<configuration_toc>` and optionally create the :ref:`runtime directories <configuration_preface>`.
+Afterwards, continue with the :ref:`usage chapter<usage>`.
+
+**Reminder**: If you want a quick introduction, please read the :ref:`tutorial`.
 
-    installation/packages
-    installation/apt-repos
-    installation/rpm-repos
-    installation/user-privileges
-    installation/freebsd-jail-with-iocage
-    installation/what-next

docs/installation/apt-repos.rst

@@ -1,40 +0,0 @@
-
-.. _installation-apt-repos:
-
-Debian / Ubuntu APT repositories
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-We maintain APT repositories for Debian, Ubuntu and derivatives.
-The fingerprint of the signing key is ``E101 418F D3D6 FBCB 9D65  A62D 7086 99FC 5F2E BF16``.
-It is available at `<https://zrepl.cschwarz.com/apt/apt-key.asc>`_ .
-Please open an issue in on GitHub if you encounter any issues with the repository.
-
-::
-
-    (
-    set -ex
-    zrepl_apt_key_url=https://zrepl.cschwarz.com/apt/apt-key.asc
-    zrepl_apt_key_dst=/usr/share/keyrings/zrepl.gpg
-    zrepl_apt_repo_file=/etc/apt/sources.list.d/zrepl.list
-
-    # Install dependencies for subsequent commands
-    sudo apt update && sudo apt install curl gnupg lsb-release
-
-    # Deploy the zrepl apt key.
-    curl -fsSL "$zrepl_apt_key_url" | tee | gpg --dearmor | sudo tee "$zrepl_apt_key_dst" > /dev/null
-
-    # Add the zrepl apt repo.
-    ARCH="$(dpkg --print-architecture)"
-    CODENAME="$(lsb_release -i -s | tr '[:upper:]' '[:lower:]') $(lsb_release -c -s | tr '[:upper:]' '[:lower:]')"
-    echo "Using Distro and Codename: $CODENAME"
-    echo "deb [arch=$ARCH signed-by=$zrepl_apt_key_dst] https://zrepl.cschwarz.com/apt/$CODENAME main" | sudo tee "$zrepl_apt_repo_file" > /dev/null
-
-    # Update apt repos.
-    sudo apt update
-    )
-
-.. NOTE::
-
-   Until zrepl reaches 1.0, the repositories will be updated to the latest zrepl release immediately.
-   This includes breaking changes between zrepl versions.
-   Use ``apt-mark hold zrepl`` to prevent upgrades of zrepl.

docs/installation/freebsd-jail-with-iocage.rst

@@ -1,152 +0,0 @@
-.. include:: ../global.rst.inc
-
-.. _installation-freebsd-jail-with-iocage:
-
-FreeBSD Jail With iocage
-========================
-
-
-This tutorial shows how zrepl can be installed on FreeBSD, or FreeNAS in a jail using iocage.
-While this tutorial focuses on using iocage, much of the setup would be similar
-using a different jail manager.
-
-.. NOTE::
-
-   From a security perspective, just keep in mind that ``zfs send``/``recv`` was never designed with
-   jails in mind, an attacker could probably crash the receive-side kernel or worse induce stateful
-   damage to the receive-side pool if they were able to get access to the jail.
-
-   The jail doesn't provide security benefits, but only management ones.
-
-Requirements
-------------
-
-A dataset that will be delegated to the jail needs to be created if one does not already exist.
-For the tutorial ``tank/zrepl`` will be used.
-
-.. code-block:: bash
-
-   zfs create -o mountpoint=none tank/zrepl
-
-The only software requirements on the host system are ``iocage``, which can be installed
-from ports or packages.
-
-.. code-block:: bash
-
-   pkg install py37-iocage
-
-.. NOTE::
-
-   By default ``iocage`` will "activate" on first use which will set up some defaults such as
-   which pool will be used. To activate ``iocage`` manually the ``iocage activate`` command can be used.
-
-Jail Creation
--------------
-
-There are two options for jail creation using FreeBSD.
-
-1. Manually set up the jail from scratch
-2. Create the jail using the ``zrepl`` plugin. On FreeNAS this is possible from the user interface using the community index.
-
-Manual Jail
-###########
-
-Create a jail, using the same release as the host, called ``zrepl`` that will be automatically started at boot.
-The jail will have ``tank/zrepl`` delegated into it.
-
-.. code-block:: bash
-
-   iocage create --release "$(freebsd-version -k | cut -d '-' -f '1,2')" --name zrepl \
-          boot=on nat=1 \
-          jail_zfs=on \
-          jail_zfs_dataset=zrepl \
-          jail_zfs_mountpoint='none'
-
-Enter the jail:
-
-.. code-block:: bash
-
-   iocage console zrepl
-
-Install ``zrepl``
-
-.. code-block:: bash
-
-   pkg update && pkg upgrade
-   pkg install zrepl
-
-Create the log file ``/var/log/zrepl.log``
-
-.. code-block:: bash
-
-   touch /var/log/zrepl.log && service newsyslog restart
-
-Tell syslogd to redirect facility local0 to the ``zrepl.log`` file:
-
-.. code-block:: bash
-
-   service syslogd reload
-
-Enable the zrepl daemon to start automatically at boot:
-
-.. code-block:: bash
-
-   sysrc zrepl_enable="YES"
-
-Now jump to :ref:`the summary <installation-freebsd-jail-summary>` below.
-
-Plugin
-######
-
-When using the plugin, ``zrepl`` will be installed for you in a jail using the following ``iocage`` properties.
-
-* ``nat=1``
-* ``jail_zfs=on``
-* ``jail_zfs_mountpoint=none``
-
-Additionally the delegated dataset should be specified upon creation, and optionally start on boot can be set.
-This can also be done from the FreeNAS webui.
-
-.. code-block:: bash
-
-   fetch https://raw.githubusercontent.com/ix-plugin-hub/iocage-plugin-index/master/zrepl.json -o /tmp/zrepl.json
-   iocage fetch -P /tmp/zrepl.json --name zrepl jail_zfs_dataset=zrepl boot=on
-
-Configuration
--------------
-
-Now ``zrepl`` can be configured.
-
-Enter the jail.
-
-.. code-block:: bash
-
-   iocage console zrepl
-
-Modify the ``/usr/local/etc/zrepl/zrepl.yml`` configuration file.
-
-.. TIP::
-
-    Note: check out the :ref:`quick-start guides <quickstart-toc>` for examples of a ``sink`` job.
-
-Now ``zrepl`` can be started.
-
-.. code-block:: bash
-
-   service zrepl start
-
-Now jump to :ref:`the summary <installation-freebsd-jail-summary>` below.
-
-.. _installation-freebsd-jail-summary:
-
-Summary
--------
-
-Congratulations, you have a working jail!
-
-.. NOTE::
-
-   With FreeBSD 13's transition to OpenZFS 2.0, please ensure that your jail's FreeBSD version matches the one in the kernel module.
-   If you are getting cryptic errors such as
-   ``cannot receive new filesystem stream: invalid backup stream``
-   the instructions posted `here <https://github.com/zrepl/zrepl/issues/500#issuecomment-966215205>`_ might help.

docs/installation/packages.rst

@@ -1,50 +0,0 @@
-.. _installation-packages:
-
-.. _binary releases: https://github.com/zrepl/zrepl/releases
-
-Packages
---------
-
-zrepl source releases are signed & tagged by the author in the git repository.
-Your OS vendor may provide binary packages of zrepl through the package manager.
-Additionally, `binary releases`_ are provided on GitHub.
-The following list may be incomplete, feel free to submit a PR with an update:
-
-.. list-table::
-    :header-rows: 1
-
-    * - OS / Distro
-      - Install Command
-      - Link
-    * - any
-      - Statically linked binaries.
-      - `Official GitHub releases <binary releases_>`_
-    * - FreeBSD
-      - ``pkg install zrepl``
-      - `<https://www.freshports.org/sysutils/zrepl/>`_
-
-        :ref:`installation-freebsd-jail-with-iocage`
-    * - FreeNAS
-      -
-      - :ref:`installation-freebsd-jail-with-iocage`
-    * - MacOS
-      - ``brew install zrepl``
-      - Available on `homebrew <https://brew.sh>`_
-    * - Arch Linux
-      - ``yay install zrepl``
-      - Available on `AUR <https://aur.archlinux.org/packages/zrepl>`_
-    * - Fedora / RHEL / OpenSUSE
-      - ``dnf install zrepl``
-      - :ref:`RPM repository config <installation-rpm-repos>`
-    * - Debian + Ubuntu
-      - ``apt install zrepl``
-      - :ref:`APT repository config <installation-apt-repos>`
-    * - OmniOS
-      - ``pkg install zrepl``
-      - Available since `r151030 <https://pkg.omniosce.org/r151030/extra/en/search.shtml?token=zrepl&action=Search>`_
-    * - Void Linux
-      - ``xbps-install zrepl``
-      - Available since `a88a2a4 <https://github.com/void-linux/void-packages/commit/a88a2a4d7bf56072dadf61ab56b8424e39155890>`_
-    * - any
-      - Build from source
-      - :repomasterlink:`README.md`.

docs/installation/rpm-repos.rst

@@ -1,55 +0,0 @@
-.. _installation-rpm-repos:
-
-RPM repositories
-~~~~~~~~~~~~~~~~
-
-We provide a single RPM repository for all RPM-based Linux distros.
-The zrepl binary in the repo is the same as the one published to GitHub.
-Since Go binaries are statically linked, the RPM should work about everywhere.
-Please open an issue on GitHub if you encounter any issues with the repository.
-
-The fingerprint of the repo & package signing key is:
-``F6F6 E8EA 6F2F 1462 2878 B5DE 50E3 4417 826E 2CE6``.
-It is available at `<https://zrepl.cschwarz.com/rpm/rpm-key.asc>`_ .
-
-.. NOTE::
-
-   Until zrepl reaches 1.0, the repository will be updated to the latest zrepl release immediately.
-   This includes breaking changes between zrepl versions.
-   If that bothers you, use the `dnf versionlock plugin <https://dnf-plugins-core.readthedocs.io/en/latest/versionlock.html>`_ or ``zypper addlock zrepl`` to pin the version of zrepl on your system.
-
-
-
-Fedora / AlmaLinux / Rocky Linux / RHEL
-^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-
-For ``dnf``/``yum``-based distributions, copy-paste the following snippet:
-
-.. code-block:: bash
-
-    rpmkeys --import 'https://zrepl.cschwarz.com/rpm/rpm-key.asc'
-    cat > /etc/yum.repos.d/zrepl.repo <<EOF
-    [zrepl]
-    name = zrepl
-    baseurl = https://zrepl.cschwarz.com/rpm/repo
-    gpgkey = https://zrepl.cschwarz.com/rpm/rpm-key.asc
-    repo_gpgcheck = 1
-    EOF
-    dnf install zrepl
-    # Or if you're on an older system:
-    # yum install zrepl
-
-You will be asked by dnf to verify repository metadata (``repo_gpgcheck``).
-There is no way to automate that prompt.
-
-openSUSE
-^^^^^^^^
-
-For SUSE-based distributions that use ``zypper``, copy-paste the following snippet:
-
-.. code-block:: bash
-
-    rpmkeys --import 'https://zrepl.cschwarz.com/rpm/rpm-key.asc'
-    zypper ar --check --gpgcheck-strict --refresh https://zrepl.cschwarz.com/rpm/repo zrepl
-    zypper install zrepl
-

docs/installation/user-privileges.rst

@@ -1,12 +0,0 @@
-.. _installation-user-privileges:
-
-User Privileges
----------------
-
-It is possible to run zrepl as an unprivileged user in combination with
-`ZFS delegation <https://www.freebsd.org/doc/handbook/zfs-zfs-allow.html>`_.
-Also, there is the possibility to run it in a jail on FreeBSD by delegating a dataset to the jail.
-
-.. TIP::
-
-    Note: check out the :ref:`installation-freebsd-jail-with-iocage` for FreeBSD jail setup instructions.