docs: changelog for 0.6.1

should replace loop with `copy.outs[level] = append(copy.outs[level], os.outs[level]...)` (gosimple)

.circleci/config.yml

@@ -1,145 +1,296 @@
-version: 2.0
+version: 2.1
+orbs:
+  # NB: 1.7.2 is not the Go version, but the Orb version
+  # https://circleci.com/developer/orbs/orb/circleci/go#usage-go-modules-cache
+  go: circleci/go@1.7.2
+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
+
+  invoke-lazy-sh:
+    parameters:
+      subcommand:
+        type: string
+    steps:
+      - run:
+          environment:
+            TERM: xterm
+          command: ./lazy.sh <<parameters.subcommand>>
+
+  apt-update-and-install-common-deps:
+    steps:
+      - run: sudo apt-get update
+      - run: sudo apt-get install -y gawk make
+      # CircleCI doesn't update its cimg/go images.
+      # So, need to update manually to get up-to-date trust chains.
+      # The need for this was required for cimg/go:1.12, but let's future proof this here and now.
+      - run: sudo apt-get install -y git ca-certificates
+
+  install-godep:
+    steps:
+      - apt-update-and-install-common-deps
+      - invoke-lazy-sh:
+          subcommand: godep
+
+  install-docdep:
+    steps:
+      - apt-update-and-install-common-deps
+      - run: sudo apt install python3 python3-pip libgirepository1.0-dev
+      - invoke-lazy-sh:
+          subcommand: docdep
+
+  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"
+
+      # https://circleci.com/docs/2.0/add-ssh-key/#adding-multiple-keys-with-blank-hostnames
+      - run: ssh-add -D
+      # the default circleci ssh config only additional ssh keys for Host !github.com
+      - run:
+          command: |
+            cat > ~/.ssh/config \<<EOF
+            Host *
+              IdentityFile /home/circleci/.ssh/id_rsa_458e62c517f6c480e40452126ce47421
+            EOF
+      - add_ssh_keys:
+          fingerprints:
+            # deploy key for zrepl.github.io
+            - "45:8e:62:c5:17:f6:c4:80:e4:04:52:12:6c:e4:74:21"
+
+      # 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
+
+  release_docker_baseimage_tag:
+    type: string
+    default: "1.21"
+
 workflows:
   version: 2
-  build:
-    jobs:
-      - build-1.11
-      - build-1.12
-      - build-1.13
-      - build-1.14
-      - build-latest
-      - test-build-in-docker
-jobs:
 
-  # build-latest serves as the template
+  ci:
-  # we use YAML anchors & aliases to exchange the docker image (and hence Go version used for the build)
+    when: << pipeline.parameters.do_ci >>
-  build-latest: &build-latest
+    jobs:
-    description: Builds zrepl
+      - quickcheck-docs
+      - quickcheck-go: &quickcheck-go-smoketest
+          name: quickcheck-go-amd64-linux-1.21
+          goversion: &latest-go-release "1.21"
+          goos: linux
+          goarch: amd64
+      - test-go-on-latest-go-release:
+          goversion: *latest-go-release
+      - quickcheck-go:
+          requires:
+            - quickcheck-go-amd64-linux-1.21 #quickcheck-go-smoketest.name
+          matrix: &quickcheck-go-matrix
+            alias: quickcheck-go-matrix
+            parameters:
+              goversion: [*latest-go-release, "1.20"]
+              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"]
+          requires:
+            - 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:
+      - publish-zrepl-github-io:
+          filters:
+            branches:
+              only:
+                - stable
+
+jobs:
+  quickcheck-docs:
+    docker:
+      - image: cimg/base:2023.09
+    steps:
+      - checkout
+      - install-docdep
+      # do the current docs build
+      - run: make docs
+      # does the publish.sh script still work?
+      - docs-publish-sh:
+          push: false
+
+  quickcheck-go:
     parameters:
-      image:
+      goversion:
-        description: "the docker image that the job should use"
+        type: string
+      goos:
+        type: string
+      goarch:
         type: string
     docker:
-      - image: circleci/golang:latest
+      - image: cimg/go:<<parameters.goversion>>
     environment:
-      # required by lazy.sh
+      GOOS: <<parameters.goos>>
-      TERM: xterm
+      GOARCH: <<parameters.goarch>>
-    working_directory: /go/src/github.com/zrepl/zrepl
+
     steps:
-      - run:
-          name: Setup environment variables
-          command: |
-            # used by pip (for docs)
-            echo 'export PATH="$HOME/.local/bin:$PATH"' >> $BASH_ENV
-            # we use modules
-            echo 'export GO111MODULE=on' >> $BASH_ENV
-
-      - restore_cache:
-          keys:
-            - source
-            - protobuf
-
       - checkout
 
-      - save_cache:
+      - go/load-cache:
-          key: source
+          key: quickcheck-<<parameters.goversion>>
-          paths:
+      - install-godep
-            - ".git"
+      - run: go mod download
-      
+      - run: cd build && go mod download
-      # install deps
+      - go/save-cache:
-      - run: wget https://github.com/protocolbuffers/protobuf/releases/download/v3.6.1/protoc-3.6.1-linux-x86_64.zip
+          key: quickcheck-<<parameters.goversion>>
-      - run: echo "6003de742ea3fcf703cfec1cd4a3380fd143081a2eb0e559065563496af27807  protoc-3.6.1-linux-x86_64.zip" | sha256sum -c
-      - run: sudo unzip -d /usr protoc-3.6.1-linux-x86_64.zip
-      - save_cache:
-          key: protobuf
-          paths:
-            - "/usr/include/google/protobuf"
-            
-      - run: sudo apt update && sudo apt install python3 python3-pip libgirepository1.0-dev gawk
-      - run: ./lazy.sh devsetup
 
-      - run: make zrepl-bin
+      - run: make formatcheck
+      - run: make generate-platform-test-list
+      - run: make zrepl-bin test-platform-bin
       - run: make vet
-      - run: make test
       - run: make lint
-      - run: make release
-
 
+      - run: rm -f artifacts/generate-platform-test-list
       - store_artifacts:
-          path: ./artifacts/release
+          path: artifacts
-          when: always
+      - persist_to_workspace:
+          root: .
+          paths: [.]
 
-      - run:
+  platformtest:
-          shell: /bin/bash -eo pipefail
+    parameters:
-          when: always
+      goversion:
-          command: |
+        type: string
-            if [ -n "$CIRCLE_PR_NUMBER" ]; then # CIRCLE_PR_NUMBER is guaranteed to be only present in forked PRs (external)
+      goos:
-              echo "Forked PR detected. Sry, can't trust you with credentials to external artifact store, use CircleCI's instead."
+        type: string
-              exit 0
+      goarch:
-            fi
+        type: string
-            set -u # from now on
+    machine:
-            
+      image: ubuntu-2204:current
-            # Download and install minio
+    resource_class: medium
-            curl -sSL https://dl.minio.io/client/mc/release/linux-amd64/mc -o ${GOPATH}/bin/mc
-            chmod +x ${GOPATH}/bin/mc
-            mc config host add --api s3v4 zrepl-minio https://minio.cschwarz.com ${MINIO_ACCESS_KEY} ${MINIO_SECRET_KEY}
-            
-            # Upload artifacts
-            echo "$CIRCLE_BUILD_URL" > ./artifacts/release/cirlceci_build_url
-            mc cp -r artifacts/release "zrepl-minio/zrepl-ci-artifacts/${CIRCLE_SHA1}/${CIRCLE_JOB}/"
-
-            # Push Artifact Link to GitHub
-            REPO="zrepl/zrepl"
-            COMMIT="${CIRCLE_SHA1}"
-            JOB_NAME="${CIRCLE_JOB}"
-            curl "https://api.github.com/repos/$REPO/statuses/$COMMIT?access_token=$GITHUB_COMMIT_STATUS_TOKEN" \
-              -H "Content-Type: application/json" \
-              -X POST \
-              -d '{"context":"zrepl/publish-ci-artifacts", "state": "success", "description":"CI Build Artifacts for '"$JOB_NAME"'", "target_url":"https://minio.cschwarz.com/minio/zrepl-ci-artifacts/'"$COMMIT"'/"}'
-
-  build-1.11:
-    <<: *build-latest
-    docker:
-    - image: circleci/golang:1.11
-
-  build-1.12:
-    <<: *build-latest
-    docker:
-    - image: circleci/golang:1.12
-
-  build-1.13:
-    <<: *build-latest
-    docker:
-    - image: circleci/golang:1.13
-
-  build-1.14:
-    <<: *build-latest
-    docker:
-    - image: circleci/golang:1.14
-
-  # this job tries to mimic the build-in-docker instructions
-  # given in docs/installation.rst
-  #
-  # However, CircleCI doesn't support volume mounts, so we have to copy
-  # the source into the build-container by modifying the Dockerfile
-  test-build-in-docker:
-    description: Check that build-in-docker works
-    docker:
-      - image: circleci/golang:latest
     environment:
-    working_directory: /go/src/github.com/zrepl/zrepl
+      GOOS: <<parameters.goos>>
+      GOARCH: <<parameters.goarch>>
     steps:
-    - checkout
+      - attach_workspace:
-    - setup_remote_docker
+          at: .
-    - run:
+      - run: sudo apt-get update
-        name: (hacky) circleci doesn't allow volume mounts, so copy src to container
+      - run: sudo apt-get install -y zfsutils-linux
-        command: echo "ADD . /src" >> build.Dockerfile
+      - run: sudo zfs version
-    - run:
+      - run: sudo make test-platform GOOS="$GOOS" GOARCH="$GOARCH"
-        name: (hacky) commit modified Dockerfile to avoid failing git clean check in Makefile
+
-        command: git -c user.name='circleci' -c user.email='circleci@localhost' commit -m 'CIRCLECI modified Dockerfile with zrepl src' --author 'autoauthor <circleci@localhost>' -- build.Dockerfile
+  test-go-on-latest-go-release:
-    - run:
+    parameters:
-        name: build the build image (build deps)
+      goversion:
-        command: docker build -t zrepl_build -f build.Dockerfile .
+        type: string
-    - run:
+    docker:
-        name: try compiling
+      - image: cimg/go:<<parameters.goversion>>
-        command: docker run -it zrepl_build make release
+    steps:
+      - checkout
+      - go/load-cache:
+          key: make-test-go
+      - run: make test-go
+      - go/save-cache:
+          key: make-test-go
+
+  release-build:
+    machine:
+      image: ubuntu-2004:202201-02
+    steps:
+      - checkout
+      - run: make release-docker RELEASE_DOCKER_BASEIMAGE_TAG=<<pipeline.parameters.release_docker_baseimage_tag>>
+      - persist_to_workspace:
+          root: .
+          paths: [.]
+  release-deb:
+    machine:
+      image: ubuntu-2004:202201-02
+    steps:
+      - attach_workspace:
+          at: .
+      - run: make debs-docker
+      - persist_to_workspace:
+          root: .
+          paths:
+            - "artifacts/*.deb"
+
+  release-rpm:
+    machine:
+      image: ubuntu-2004:202201-02
+    steps:
+      - attach_workspace:
+          at: .
+      - run: make rpms-docker
+      - persist_to_workspace:
+          root: .
+          paths:
+            - "artifacts/*.rpm"
+
+  release-upload:
+    docker:
+      - image: cimg/base:2020.08
+    steps:
+      - attach_workspace:
+          at: .
+      - store_artifacts:
+          path: artifacts
+
+  publish-zrepl-github-io:
+    docker:
+      - image: cimg/base:2023.09
+    steps:
+      - checkout
+      - install-docdep
+      - docs-publish-sh:
+          push: true

.circleci/trigger_debian_binary_packaging_workflow.bash

@@ -0,0 +1,10 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+COMMIT="$1"
+GO_VERSION="$2"
+
+curl -v -X POST https://api.github.com/repos/zrepl/debian-binary-packaging/dispatches \
+    -H 'Accept: application/vnd.github.v3+json' \
+    -H "Authorization: token $GITHUB_ACCESS_TOKEN" \
+    --data '{"event_type": "push", "client_payload": { "zrepl_main_repo_commit": "'"$COMMIT"'", "go_version": "'"$GO_VERSION"'" }}'

.github/FUNDING.yml

@@ -1,3 +1,4 @@
+github: problame
 patreon: zrepl
 liberapay: zrepl
 custom: https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=R5QSXJVYHGX96

.golangci.yml

@@ -7,4 +7,10 @@ issues:
     - path: _test\.go
       linters:
         - errcheck
+    # Disable staticcheck 'Empty body in an if or else branch' as it's useful
+    # to put a comment into an empty else-clause that explains why whatever
+    # is done in the if-caluse is not necessary if the condition is false.
+    - linters:
+        - staticcheck
+      text: "SA9003:"
 

.travis.yml

@@ -1,70 +0,0 @@
-dist: xenial
-services:
-- docker
-
-env: # for allow_failures: https://docs.travis-ci.com/user/customizing-the-build/
-matrix:
-  include:
-
-  - language: go
-    name: "Build in Docker (docs/installation.rst)"
-    script:
-    - 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
-
-  - &zrepl_build_template
-    language: go
-    go_import_path: github.com/zrepl/zrepl
-    before_install:
-        - wget https://github.com/protocolbuffers/protobuf/releases/download/v3.6.1/protoc-3.6.1-linux-x86_64.zip
-        - echo "6003de742ea3fcf703cfec1cd4a3380fd143081a2eb0e559065563496af27807  protoc-3.6.1-linux-x86_64.zip" | sha256sum -c
-        - sudo unzip -d /usr protoc-3.6.1-linux-x86_64.zip
-        - ./lazy.sh godep
-        - make vendordeps
-    script:
-      - make
-      - make vet
-      - make test
-      - make lint
-      - make artifacts/zrepl-freebsd-amd64
-      - make artifacts/zrepl-linux-amd64
-      - make artifacts/zrepl-darwin-amd64
-    go:
-    - "1.11"
-
-  - <<: *zrepl_build_template
-    go:
-    - "1.12"
-
-  - <<: *zrepl_build_template
-    go:
-    - "master"
-
-  - &zrepl_docs_template 
-    language: python
-    python:
-    - "3.4"
-    install:
-    - sudo apt-get install libgirepository1.0-dev
-    - pip install -r docs/requirements.txt
-    script:
-    - make docs
-  - <<: *zrepl_docs_template
-    python:
-    - "3.5"
-  - <<: *zrepl_docs_template
-    python:
-    - "3.6"
-  - <<: *zrepl_docs_template
-    python:
-    - "3.7"
-  
-
-  allow_failures:
-  - <<: *zrepl_build_template
-    go:
-    - "master"

Makefile

@@ -27,6 +27,10 @@ GO_EXTRA_BUILDFLAGS :=
 GO_BUILDFLAGS := $(GO_MOD_READONLY) $(GO_EXTRA_BUILDFLAGS)
 GO_BUILD := $(GO_ENV_VARS) $(GO) build $(GO_BUILDFLAGS) -ldflags $(GO_LDFLAGS)
 GOLANGCI_LINT := golangci-lint
+GOCOVMERGE := gocovmerge
+RELEASE_DOCKER_BASEIMAGE_TAG ?= 1.21
+RELEASE_DOCKER_BASEIMAGE ?= golang:$(RELEASE_DOCKER_BASEIMAGE_TAG)
+
 ifneq ($(GOARM),)
 	ZREPL_TARGET_TUPLE := $(GOOS)-$(GOARCH)v$(GOARM)
 else
@@ -45,16 +49,101 @@ printvars:
 
 release: clean
 	# no cross-platform support for target test
-	$(MAKE) test
+	$(MAKE) test-go
 	$(MAKE) bins-all
 	$(MAKE) noarch
 	$(MAKE) wrapup-and-checksum
 	$(MAKE) check-git-clean
 ifeq (SIGN, 1)
-	$(make) sign
+	$(MAKE) sign
 endif
 	@echo "ZREPL RELEASE ARTIFACTS AVAILABLE IN artifacts/release"
 
+release-docker: $(ARTIFACTDIR)
+	sed 's/FROM.*!SUBSTITUTED_BY_MAKEFILE/FROM $(RELEASE_DOCKER_BASEIMAGE)/' build.Dockerfile > artifacts/release-docker.Dockerfile
+	docker build -t zrepl_release --pull -f artifacts/release-docker.Dockerfile .
+	docker run --rm -i -v $(CURDIR):/src -u $$(id -u):$$(id -g) \
+		zrepl_release \
+		make release GOOS=$(GOOS) GOARCH=$(GOARCH) GOARM=$(GOARM)
+
+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)"
+	mkdir -p "$(_ZREPL_RPM_TOPDIR_ABS)"/{SPECS,RPMS,BUILD,BUILDROOT}
+	sed "s/^Version:.*/Version:          $(_ZREPL_RPM_VERSION)/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)
+
+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))"; \
+		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)
+
 # expects `release` target to have run before
 NOARCH_TARBALL := $(ARTIFACTDIR)/zrepl-noarch.tar
 wrapup-and-checksum:
@@ -90,6 +179,10 @@ check-git-clean:
 		fi; \
 	fi;
 
+tag-release:
+	test -n "$(ZREPL_TAG_VERSION)" || exit 1
+	git tag -u E27CA5FC -m "$(ZREPL_TAG_VERSION)" "$(ZREPL_TAG_VERSION)"
+
 sign:
 	gpg -u "89BC 5D89 C845 568B F578  B306 CDBD 8EC8 E27C A5FC" \
 		--armor \
@@ -99,13 +192,15 @@ clean: docs-clean
 	rm -rf "$(ARTIFACTDIR)"
 
 ##################### BINARIES #####################
-.PHONY: bins-all lint test vet zrepl-bin platformtest-bin
+.PHONY: bins-all lint test-go test-platform cover-merge cover-html vet zrepl-bin test-platform-bin generate-platform-test-list
 
-BINS_ALL_TARGETS := zrepl-bin platformtest-bin vet lint
+BINS_ALL_TARGETS := zrepl-bin test-platform-bin vet lint
 GO_SUPPORTS_ILLUMOS := $(shell $(GO) version | gawk -F '.' '/^go version /{split($$0, comps, " "); split(comps[3], v, "."); if (v[1] == "go1" && v[2] >= 13) { print "illumos"; } else { print "noillumos"; }}')
 bins-all:
 	$(MAKE) $(BINS_ALL_TARGETS) GOOS=freebsd   GOARCH=amd64
 	$(MAKE) $(BINS_ALL_TARGETS) GOOS=freebsd   GOARCH=386
+	$(MAKE) $(BINS_ALL_TARGETS) GOOS=freebsd   GOARCH=arm	  GOARM=7
+	$(MAKE) $(BINS_ALL_TARGETS) GOOS=freebsd   GOARCH=arm64
 	$(MAKE) $(BINS_ALL_TARGETS) GOOS=linux     GOARCH=amd64
 	$(MAKE) $(BINS_ALL_TARGETS) GOOS=linux     GOARCH=arm64
 	$(MAKE) $(BINS_ALL_TARGETS) GOOS=linux     GOARCH=arm     GOARM=7
@@ -122,44 +217,102 @@ endif
 
 lint:
 	$(GO_ENV_VARS) $(GOLANGCI_LINT) run ./...
-test:
+
-	$(GO_ENV_VARS) $(GO) test $(GO_BUILDFLAGS) ./...
 vet:
 	$(GO_ENV_VARS) $(GO) vet $(GO_BUILDFLAGS) ./...
 
+test-go: $(ARTIFACTDIR)
+	rm -f "$(ARTIFACTDIR)/gotest.cover"
+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)"
 
-platformtest-bin:
+generate-platform-test-list:
-	$(GO_BUILD) -o "$(ARTIFACTDIR)/platformtest-$(ZREPL_TARGET_TUPLE)" ./platformtest/harness
+	$(GO_BUILD) -o $(ARTIFACTDIR)/generate-platform-test-list ./platformtest/tests/gen
 
-##################### DEV TARGETS #####################
+COVER_PLATFORM_BIN_PATH := $(ARTIFACTDIR)/platformtest-cover-$(ZREPL_TARGET_TUPLE)
-# not part of the build, must do that manually
+cover-platform-bin:
-.PHONY: generate format platformtest
+	$(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
 
-generate:
+TEST_PLATFORM_BIN_PATH := $(ARTIFACTDIR)/platformtest-$(ZREPL_TARGET_TUPLE)
-	protoc -I=replication/logic/pdu --go_out=plugins=grpc:replication/logic/pdu replication/logic/pdu/pdu.proto
+test-platform-bin:
-	$(GO_ENV_VARS) $(GO) generate $(GO_BUILDFLAGS) -x ./...
+	$(GO_BUILD) -o "$(TEST_PLATFORM_BIN_PATH)" ./platformtest/harness
-
+test-platform:
-format:
+	export _TEST_PLATFORM_CMD="\"$(TEST_PLATFORM_BIN_PATH)\""; \
-	goimports -srcdir . -local 'github.com/zrepl/zrepl' -w $(shell find . -type f -name '*.go' -not -path "./vendor/*" -not -name '*.pb.go' -not -name '*_enumer.go')
+		$(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 := 
+ZREPL_PLATFORMTEST_ARGS :=
-platformtest: # do not track dependency on platformtest-bin to allow build of platformtest outside of test VM
+_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 "$(ARTIFACTDIR)/platformtest.cover"
 	platformtest/logmockzfs/logzfsenv "$(ZREPL_PLATFORMTEST_ZFS_LOG)" `which zfs` \
-	"$(ARTIFACTDIR)/platformtest-$(ZREPL_TARGET_TUPLE)" \
+		$(_TEST_PLATFORM_CMD) \
 		-poolname "$(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 "$$(id -u)" = "0" || echo "MUST RUN AS ROOT" 1>&2
+	$(MAKE) test-go COVER=1
+	$(MAKE) cover-platform-bin
+	$(MAKE) cover-platform
+	$(MAKE) cover-html
+
+##################### DEV TARGETS #####################
+# not part of the build, must do that manually
+.PHONY: generate formatcheck format
+
+generate: generate-platform-test-list
+	protoc -I=replication/logic/pdu --go_out=replication/logic/pdu --go-grpc_out=replication/logic/pdu replication/logic/pdu/pdu.proto
+	protoc -I=rpc/grpcclientidentity/example --go_out=rpc/grpcclientidentity/example/pdu --go-grpc_out=rpc/grpcclientidentity/example/pdu rpc/grpcclientidentity/example/grpcauth.proto
+	$(GO_ENV_VARS) $(GO) generate $(GO_BUILDFLAGS) -x ./...
+
+GOIMPORTS := goimports -srcdir . -local 'github.com/zrepl/zrepl'
+FINDSRCFILES := find . -type f -name '*.go' -not -path "./vendor/*" -not -name '*.pb.go' -not -name '*_enumer.go'
+
+formatcheck:
+	@# goimports doesn't have a knob to exit with non-zero status code if formatting is needed
+	@# see https://go-review.googlesource.com/c/tools/+/237378
+	@ affectedfiles=$$($(GOIMPORTS) -l $(shell $(FINDSRCFILES)) | tee /dev/stderr | wc -l); test "$$affectedfiles" = 0
+
+format:
+	@ $(GOIMPORTS) -w -d $(shell  $(FINDSRCFILES))
+
 ##################### NOARCH #####################
 .PHONY: noarch $(ARTIFACTDIR)/bash_completion $(ARTIFACTDIR)/_zrepl.zsh_completion $(ARTIFACTDIR)/go_env.txt docs docs-clean
 
@@ -182,13 +335,16 @@ $(ARTIFACTDIR)/_zrepl.zsh_completion:
 
 $(ARTIFACTDIR)/go_env.txt:
 	$(GO_ENV_VARS) $(GO) env > $@
+	$(GO) version >> $@
 
 docs: $(ARTIFACTDIR)/docs
-	make -C docs \
+	# https://www.sphinx-doc.org/en/master/man/sphinx-build.html
+	$(MAKE) -C docs \
 		html \
 		BUILDDIR=../artifacts/docs \
+		SPHINXOPTS="-W --keep-going -n"
 
 docs-clean:
-	make -C docs \
+	$(MAKE) -C docs \
 		clean \
 		BUILDDIR=../artifacts/docs

README.md

@@ -1,10 +1,12 @@
 [![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/language-Go-6ad7e5.svg)](https://golang.org/)
+[![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)
-[![Donate via Patreon](https://img.shields.io/endpoint.svg?url=https%3A%2F%2Fshieldsio-patreon.herokuapp.com%2Fzrepl%2Fpledges&style=flat&color=yellow)](https://www.patreon.com/zrepl)
+[![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 Liberapay](https://img.shields.io/liberapay/receives/zrepl.svg?logo=liberapay)](https://liberapay.com/zrepl/donate)
+[![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.
@@ -20,7 +22,7 @@ zrepl is a one-stop ZFS backup & replication solution.
 
 ## Feature Requests
 
-1. Does you feature request require default values / some kind of configuration?
+1. Does your 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.
@@ -29,82 +31,83 @@ zrepl is a one-stop ZFS backup & replication solution.
 The above does not apply if you already implemented everything.
 Check out the *Coding Workflow* section below for details.
 
-## Package Maintainer Information
+## Building, Releasing, Downstream-Packaging
 
-* Follow the steps in `docs/installation.rst -> Compiling from Source` and read the Makefile / shell scripts used in this process.
+This section provides an overview of the zrepl build & release process.
-* Make sure your distro is compatible with the paths in `docs/installation.rst`.
+Check out `docs/installation/compile-from-source.rst` for build-from-source instructions.
-* 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.
-* Ship other material provided in `./dist`, e.g. in `/usr/share/zrepl/`.
-* Use `make release ZREPL_VERSION='mydistro-1.2.3_1'`
-    * Your distro's name and any versioning supplemental to zrepl's (e.g. package revision) should be in this string
-* Use `make platformtest` **on a test system** to validate that zrepl's abstractions on top of ZFS work with the system ZFS.
-* Make sure you are informed about new zrepl versions, e.g. by subscribing to GitHub's release RSS feed.
 
-## Developer Documentation
+### Overview
 
 zrepl is written in [Go](https://golang.org) and uses [Go modules](https://github.com/golang/go/wiki/Modules) to manage dependencies.
 The documentation is written in [ReStructured Text](http://docutils.sourceforge.net/rst.html) using the [Sphinx](https://www.sphinx-doc.org) framework.
 
-To get started, run `./lazy.sh devsetup` to easily install build dependencies and read `docs/installation.rst -> Compiling from Source`.
+Install **build dependencies** using  `./lazy.sh devsetup`.
 `lazy.sh` uses `python3-pip` to fetch the build dependencies for the docs - you might want to use a [venv](https://docs.python.org/3/library/venv.html).
 If you just want to install the Go dependencies, run `./lazy.sh godep`.
 
-### Project Structure
+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`.
-├── artifacts               # build artifcats generate by make
+An HTML report can be generated using `make cover-html`.
-├── cli                     # wrapper around CLI package cobra
-├── client                  # all subcommands that are not `daemon`
-├── config                  # config data types (=> package yaml-config)
-│   └── samples
-├── daemon                  # the implementation of `zrepl daemon` subcommand
-│   ├── filters
-│   ├── hooks               # snapshot hooks
-│   ├── job                 # job implementations
-│   ├── logging             # logging outlets + formatters
-│   ├── nethelpers
-│   ├── prometheus
-│   ├── pruner              # pruner implementation
-│   ├── snapper             # snapshotter implementation
-├── dist                    # supplemental material for users & package maintainers
-├── 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
-├── endpoint                # implementation of replication endpoints (=> package replication)
-├── logger                  # our own logger package
-├── platformtest            # test suite for our zfs abstractions (error classification, etc)
-├── pruning                 # pruning rules (the logic, not the actual execution)
-│   └── retentiongrid
-├── replication
-│   ├── driver              # the driver of the replication logic (status reporting, error handling)
-│   ├── logic               # planning & executing replication steps via rpc
-|   |   └── pdu             # the generated gRPC & protobuf code used in replication (and endpoints)
-│   └── report              # the JSON-serializable report datastructures exposed to the client
-├── rpc                     # the hybrid gRPC + ./dataconn RPC client: connects to a remote replication.Endpoint
-│   ├── dataconn            # Bulk data-transfer RPC protocol
-│   ├── grpcclientidentity  # adaptor to inject package transport's 'client identity' concept into gRPC contexts
-│   ├── netadaptor          # adaptor to convert a package transport's Connecter and Listener into net.* primitives
-│   ├── transportmux        # TCP connecter and listener used to split control & data traffic
-│   └── versionhandshake    # replication protocol version handshake perfomed on newly established connections
-├── tlsconf                 # abstraction for Go TLS server + client config
-├── transport               # transport implementations
-│   ├── fromconfig
-│   ├── local
-│   ├── ssh
-│   ├── tcp
-│   └── tls
-├── util
-├── version                 # abstraction for versions (filled during build by Makefile)
-└── zfs                     # zfs(8) wrappers
-```
 
-### Coding Workflow
+**Code generation** is triggered by `make generate`. Generated code is committed to the source tree.
+
+### Build & Release Process
+
+**The `Makefile` is catering to the needs of developers & CI, not distro packagers**.
+It provides phony targets for
+* local development (building, running tests, etc)
+* building a release in Docker (used by the CI & release management)
+* building .deb and .rpm packages out of the release artifacts.
+
+**Build tooling & dependencies** are documented as code in `lazy.sh`.
+Go dependencies are then fetched by the go command and pip dependencies are pinned through a `requirements.txt`.
+
+**We use CircleCI for continuous integration**.
+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.
+  Artifacts are published to minio.cschwarz.com (see GitHub Commit Status).
+
+* `release` runs
+  * on manual triggers through the CircleCI API (in order to produce a release)
+  * periodically on `master`
+  Artifacts are published to minio.cschwarz.com (see GitHub Commit Status).
+
+**Releases** are issued via Git tags + GitHub Releases feature.
+The procedure to issue a release is as follows:
+* Issue the source release:
+  * Git tag the release on the `master` branch.
+  * Push the tag.
+  * Run `./docs/publish.sh` to re-build & push zrepl.github.io.
+* Issue the official binary release:
+  * Run the `release` pipeline (triggered via CircleCI API)
+  * Download the artifacts to the release manager's machine.
+  * Create a GitHub release, edit the changelog, upload all the release artifacts, including .rpm and .deb files.
+  * Issue the GitHub release.
+  * Add the .rpm and .deb files to the official zrepl repos, publish those.
+
+**Official binary releases are not re-built when Go receives an update. If the Go update is critical to zrepl (e.g. a Go security update that affects zrepl), we'd issue a new source release**.
+The rationale for this is that whereas distros provide a mechanism for this (`$zrepl_source_release-$distro_package_revision`), GitHub Releases doesn't which means we'd need to update the existing GitHub release's assets, which nobody would notice (no RSS feed updates, etc.).
+Downstream packagers can read the changelog to determine whether they want to push that minor release into their distro or simply skip it.
+
+### Additional Notes to Distro Package Maintainers
+
+* 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.
+* Ship other material provided in `./dist`, e.g. in `/usr/share/zrepl/`.
+* 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.
+
+
+## Contributing Code
 
 * Open an issue when starting to hack on a new feature
 * Commits should reference the issue they are related to
@@ -114,9 +117,6 @@ If you just want to install the Go dependencies, run `./lazy.sh godep`.
 
 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 />
@@ -133,16 +133,3 @@ variables and types are often named *dataset* when they in fact refer to a *file
 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.
-
-### RPC debugging
-
-Optionally, there are various RPC-related environment variables, that if set to something != `""` will produce additional debug output on stderr:
-
-https://github.com/zrepl/zrepl/blob/master/rpc/rpc_debug.go#L11
-
-https://github.com/zrepl/zrepl/blob/master/rpc/dataconn/dataconn_debug.go#L11
-
-https://github.com/zrepl/zrepl/blob/master/rpc/dataconn/stream/stream_debug.go#L11
-
-https://github.com/zrepl/zrepl/blob/master/rpc/dataconn/heartbeatconn/heartbeatconn_debug.go#L11
-

build.Dockerfile

@@ -1,13 +1,19 @@
-FROM golang:latest
+FROM !SUBSTITUTED_BY_MAKEFILE
 
 RUN apt-get update && apt-get install -y \
     python3-pip \
+    python3-venv \
     unzip \
     gawk
 
 ADD build.installprotoc.bash ./
 RUN bash build.installprotoc.bash
 
+# setup venv
+ENV VIRTUAL_ENV=/opt/venv
+RUN python3 -m venv $VIRTUAL_ENV
+ENV PATH="$VIRTUAL_ENV/bin:$PATH"
+
 ADD lazy.sh /tmp/lazy.sh
 ADD docs/requirements.txt /tmp/requirements.txt
 ENV ZREPL_LAZY_DOCS_REQPATH=/tmp/requirements.txt

build/go.mod

@@ -3,49 +3,38 @@ module github.com/zrepl/zrepl/build
 go 1.12
 
 require (
-	github.com/OpenPeeDeeP/depguard v0.0.0-20181229194401-1f388ab2d810 // indirect
 	github.com/alvaroloes/enumer v1.1.1
-	github.com/fatih/color v1.7.0 // indirect
+	github.com/breml/bidichk v0.2.6 // indirect
-	github.com/go-critic/go-critic v0.3.4 // indirect
+	github.com/breml/errchkjson v0.3.5 // indirect
-	github.com/gogo/protobuf v1.2.1 // indirect
+	github.com/chavacava/garif v0.1.0 // indirect
-	github.com/golang/mock v1.2.0 // indirect
+	github.com/daixiang0/gci v0.11.1 // indirect
-	github.com/golang/protobuf v1.2.0
+	github.com/golangci/golangci-lint v1.54.2
-	github.com/golangci/errcheck v0.0.0-20181223084120-ef45e06d44b6 // indirect
+	github.com/golangci/revgrep v0.5.0 // indirect
-	github.com/golangci/go-tools v0.0.0-20190124090046-35a9f45a5db0 // indirect
+	github.com/hashicorp/errwrap v1.1.0 // indirect
-	github.com/golangci/gocyclo v0.0.0-20180528144436-0a533e8fa43d // indirect
+	github.com/jgautheron/goconst v1.6.0 // indirect
-	github.com/golangci/gofmt v0.0.0-20181222123516-0b8337e80d98 // indirect
+	github.com/mattn/go-isatty v0.0.19 // indirect
-	github.com/golangci/golangci-lint v1.17.1
+	github.com/mattn/go-runewidth v0.0.15 // indirect
-	github.com/golangci/gosec v0.0.0-20180901114220-8afd9cbb6cfb // indirect
+	github.com/mgechev/revive v1.3.3 // indirect
-	github.com/golangci/lint-1 v0.0.0-20181222135242-d2cdd8c08219 // indirect
+	github.com/pelletier/go-toml/v2 v2.1.0 // indirect
-	github.com/golangci/misspell v0.3.4 // indirect
+	github.com/polyfloyd/go-errorlint v1.4.5 // indirect
-	github.com/golangci/revgrep v0.0.0-20180812185044-276a5c0a1039 // indirect
+	github.com/prometheus/client_golang v1.16.0 // indirect
-	github.com/google/go-cmp v0.3.0 // indirect
+	github.com/prometheus/common v0.44.0 // indirect
-	github.com/konsorten/go-windows-terminal-sequences v1.0.2 // indirect
+	github.com/prometheus/procfs v0.11.1 // indirect
-	github.com/mitchellh/go-homedir v1.1.0 // indirect
+	github.com/rivo/uniseg v0.4.4 // indirect
-	github.com/pkg/errors v0.8.1 // indirect
+	github.com/ryanrolds/sqlclosecheck v0.5.1 // indirect
-	github.com/sirupsen/logrus v1.4.2 // indirect
+	github.com/spf13/viper v1.16.0 // indirect
-	github.com/spf13/afero v1.2.2 // indirect
+	github.com/stretchr/objx v0.5.1 // indirect
-	github.com/spf13/jwalterweatherman v1.1.0 // indirect
+	github.com/subosito/gotenv v1.6.0 // indirect
-	github.com/spf13/viper v1.3.2 // indirect
+	github.com/tetafro/godot v1.4.15 // indirect
-	github.com/stretchr/testify v1.4.0 // indirect
+	github.com/wadey/gocovmerge v0.0.0-20160331181800-b5bfa59ec0ad
-	golang.org/x/net v0.0.0-20190613194153-d28f0bde5980 // indirect
+	github.com/xen0n/gosmopolitan v1.2.2 // indirect
-	golang.org/x/sys v0.0.0-20191010194322-b09406accb47 // indirect
+	gitlab.com/bosi/decorder v0.4.1 // indirect
-	golang.org/x/tools v0.0.0-20190524210228-3d17549cdc6b
+	go.uber.org/multierr v1.11.0 // indirect
-	mvdan.cc/unparam v0.0.0-20190310220240-1b9ccfa71afe // indirect
+	go.uber.org/zap v1.25.0 // indirect
-	sourcegraph.com/sqs/pbtypes v1.0.0 // indirect
+	golang.org/x/exp v0.0.0-20230905200255-921286631fa9 // indirect
-)
+	golang.org/x/exp/typeparams v0.0.0-20230905200255-921286631fa9 // indirect
-
+	golang.org/x/tools v0.13.0
-// invalid dates in transitive dependencies (first validated in Go 1.13, didn't fail in earlier Go versions)
+	google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.1.0
-replace (
+	google.golang.org/protobuf v1.31.0
-	github.com/go-critic/go-critic v0.0.0-20181204210945-1df300866540 => github.com/go-critic/go-critic v0.3.5-0.20190526074819-1df300866540
+	mvdan.cc/unparam v0.0.0-20230815095028-f7c6fb1088f0 // indirect
-	github.com/go-critic/go-critic v0.0.0-20181204210945-ee9bf5809ead => github.com/go-critic/go-critic v0.3.5-0.20190210220443-ee9bf5809ead
-	github.com/golangci/errcheck v0.0.0-20181003203344-ef45e06d44b6 => github.com/golangci/errcheck v0.0.0-20181223084120-ef45e06d44b6
-	github.com/golangci/go-tools v0.0.0-20180109140146-35a9f45a5db0 => github.com/golangci/go-tools v0.0.0-20190124090046-35a9f45a5db0
-	github.com/golangci/go-tools v0.0.0-20180109140146-af6baa5dc196 => github.com/golangci/go-tools v0.0.0-20190318060251-af6baa5dc196
-	github.com/golangci/gofmt v0.0.0-20181105071733-0b8337e80d98 => github.com/golangci/gofmt v0.0.0-20181222123516-0b8337e80d98
-	github.com/golangci/gosec v0.0.0-20180901114220-66fb7fc33547 => github.com/golangci/gosec v0.0.0-20190211064107-66fb7fc33547
-	github.com/golangci/ineffassign v0.0.0-20180808204949-42439a7714cc => github.com/golangci/ineffassign v0.0.0-20190609212857-42439a7714cc
-	github.com/golangci/lint-1 v0.0.0-20180610141402-ee948d087217 => github.com/golangci/lint-1 v0.0.0-20190420132249-ee948d087217
-	golang.org/x/tools v0.0.0-20190125232054-379209517ffe => golang.org/x/tools v0.0.0-20190205201329-379209517ffe
-	mvdan.cc/unparam v0.0.0-20190124213536-fbb59629db34 => mvdan.cc/unparam v0.0.0-20190209190245-fbb59629db34
 )

build/tools.go

@@ -1,11 +1,15 @@
+//go:build tools
 // +build tools
 
 package main
 
+// the lines are parsed by lazy.sh, do not edit
 import (
 	_ "github.com/alvaroloes/enumer"
-	_ "github.com/golang/protobuf/protoc-gen-go"
 	_ "github.com/golangci/golangci-lint/cmd/golangci-lint"
+	_ "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"
 )

cli/cli.go

@@ -1,12 +1,15 @@
 package cli
 
 import (
+	"context"
 	"fmt"
 	"os"
 
 	"github.com/spf13/cobra"
 	"github.com/spf13/pflag"
 
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+
 	"github.com/zrepl/zrepl/config"
 )
 
@@ -75,7 +78,7 @@ type Subcommand struct {
 	Short            string
 	Example          string
 	NoRequireConfig  bool
-	Run              func(subcommand *Subcommand, args []string) error
+	Run              func(ctx context.Context, subcommand *Subcommand, args []string) error
 	SetupFlags       func(f *pflag.FlagSet)
 	SetupSubcommands func() []*Subcommand
 
@@ -96,7 +99,11 @@ func (s *Subcommand) Config() *config.Config {
 
 func (s *Subcommand) run(cmd *cobra.Command, args []string) {
 	s.tryParseConfig()
-	err := s.Run(s, args)
+	ctx := context.Background()
+	endTask := trace.WithTaskFromStackUpdateCtx(&ctx)
+	defer endTask()
+	err := s.Run(ctx, s, args)
+	endTask()
 	if err != nil {
 		fmt.Fprintf(os.Stderr, "%s\n", err)
 		os.Exit(1)

client/configcheck.go

@@ -1,6 +1,7 @@
 package client
 
 import (
+	"context"
 	"encoding/json"
 	"fmt"
 	"os"
@@ -18,8 +19,9 @@ import (
 )
 
 var configcheckArgs struct {
-	format string
+	format        string
-	what   string
+	what          string
+	skipCertCheck bool
 }
 
 var ConfigcheckCmd = &cli.Subcommand{
@@ -28,8 +30,9 @@ var ConfigcheckCmd = &cli.Subcommand{
 	SetupFlags: func(f *pflag.FlagSet) {
 		f.StringVar(&configcheckArgs.format, "format", "", "dump parsed config object [pretty|yaml|json]")
 		f.StringVar(&configcheckArgs.what, "what", "all", "what to print [all|config|jobs|logging]")
+		f.BoolVar(&configcheckArgs.skipCertCheck, "skip-cert-check", false, "skip checking cert files")
 	},
-	Run: func(subcommand *cli.Subcommand, args []string) error {
+	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
 		formatMap := map[string]func(interface{}){
 			"": func(i interface{}) {},
 			"pretty": func(i interface{}) {
@@ -55,8 +58,16 @@ var ConfigcheckCmd = &cli.Subcommand{
 		}
 
 		var hadErr bool
+
+		parseFlags := config.ParseFlagsNone
+
+		if configcheckArgs.skipCertCheck {
+			parseFlags |= config.ParseFlagsNoCertCheck
+		}
+
 		// further: try to build jobs
-		confJobs, err := job.JobsFromConfig(subcommand.Config())
+		confJobs, err := job.JobsFromConfig(subcommand.Config(), parseFlags)
+
 		if err != nil {
 			err := errors.Wrap(err, "cannot build jobs from config")
 			if configcheckArgs.what == "jobs" {

client/migrate.go

@@ -48,14 +48,13 @@ var migratePlaceholder0_1Args struct {
 	dryRun bool
 }
 
-func doMigratePlaceholder0_1(sc *cli.Subcommand, args []string) error {
+func doMigratePlaceholder0_1(ctx context.Context, sc *cli.Subcommand, args []string) error {
 	if len(args) != 0 {
 		return fmt.Errorf("migration does not take arguments, got %v", args)
 	}
 
 	cfg := sc.Config()
 
-	ctx := context.Background()
 	allFSS, err := zfs.ZFSListMapping(ctx, zfs.NoFilter())
 	if err != nil {
 		return errors.Wrap(err, "cannot list filesystems")
@@ -124,21 +123,19 @@ var fail = color.New(color.FgRed)
 
 var migrateReplicationCursorSkipSentinel = fmt.Errorf("skipping this filesystem")
 
-func doMigrateReplicationCursor(sc *cli.Subcommand, args []string) error {
+func doMigrateReplicationCursor(ctx context.Context, sc *cli.Subcommand, args []string) error {
 	if len(args) != 0 {
 		return fmt.Errorf("migration does not take arguments, got %v", args)
 	}
 
 	cfg := sc.Config()
-	jobs, err := job.JobsFromConfig(cfg)
+	jobs, err := job.JobsFromConfig(cfg, config.ParseFlagsNone)
 	if err != nil {
 		fmt.Printf("cannot parse config:\n%s\n\n", err)
 		fmt.Printf("NOTE: this migration was released together with a change in job name requirements.\n")
 		return fmt.Errorf("exiting migration after error")
 	}
 
-	ctx := context.Background()
-
 	v1cursorJobs := make([]job.Job, 0, len(cfg.Jobs))
 	for i, j := range cfg.Jobs {
 		if jobs[i].Name() != j.Name() {

client/pprof.go

@@ -1,67 +1,10 @@
 package client
 
-import (
+import "github.com/zrepl/zrepl/cli"
-	"errors"
-	"log"
-	"os"
-
-	"github.com/zrepl/zrepl/cli"
-	"github.com/zrepl/zrepl/config"
-	"github.com/zrepl/zrepl/daemon"
-)
-
-var pprofArgs struct {
-	daemon.PprofServerControlMsg
-}
 
 var PprofCmd = &cli.Subcommand{
-	Use:   "pprof off | [on TCP_LISTEN_ADDRESS]",
+	Use: "pprof",
-	Short: "start a http server exposing go-tool-compatible profiling endpoints at TCP_LISTEN_ADDRESS",
+	SetupSubcommands: func() []*cli.Subcommand {
-	Run: func(subcommand *cli.Subcommand, args []string) error {
+		return []*cli.Subcommand{PprofListenCmd, pprofActivityTraceCmd}
-		if len(args) < 1 {
-			goto enargs
-		}
-		switch args[0] {
-		case "on":
-			pprofArgs.Run = true
-			if len(args) != 2 {
-				return errors.New("must specify TCP_LISTEN_ADDRESS as second positional argument")
-			}
-			pprofArgs.HttpListenAddress = args[1]
-		case "off":
-			if len(args) != 1 {
-				goto enargs
-			}
-			pprofArgs.Run = false
-		}
-
-		RunPProf(subcommand.Config())
-		return nil
-	enargs:
-		return errors.New("invalid number of positional arguments")
-
 	},
 }
-
-func RunPProf(conf *config.Config) {
-	log := log.New(os.Stderr, "", 0)
-
-	die := func() {
-		log.Printf("exiting after error")
-		os.Exit(1)
-	}
-
-	log.Printf("connecting to zrepl daemon")
-
-	httpc, err := controlHttpClient(conf.Global.Control.SockPath)
-	if err != nil {
-		log.Printf("error creating http client: %s", err)
-		die()
-	}
-	err = jsonRequestResponse(httpc, daemon.ControlJobEndpointPProf, pprofArgs.PprofServerControlMsg, struct{}{})
-	if err != nil {
-		log.Printf("error sending control message: %s", err)
-		die()
-	}
-	log.Printf("finished")
-}

client/pprof_activity_trace.go

@@ -0,0 +1,44 @@
+package client
+
+import (
+	"context"
+	"io"
+	"log"
+	"os"
+
+	"golang.org/x/net/websocket"
+
+	"github.com/zrepl/zrepl/cli"
+)
+
+var pprofActivityTraceCmd = &cli.Subcommand{
+	Use:   "activity-trace ZREPL_PPROF_HOST:ZREPL_PPROF_PORT",
+	Short: "attach to zrepl daemon with activated pprof listener and dump an activity-trace to stdout",
+	Run:   runPProfActivityTrace,
+}
+
+func runPProfActivityTrace(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
+	log := log.New(os.Stderr, "", 0)
+
+	die := func() {
+		log.Printf("exiting after error")
+		os.Exit(1)
+	}
+
+	if len(args) != 1 {
+		log.Printf("exactly one positional argument is required")
+		die()
+	}
+
+	url := "ws://" + args[0] + "/debug/zrepl/activity-trace" // FIXME dont' repeat that
+
+	log.Printf("attaching to activity trace stream %s", url)
+	ws, err := websocket.Dial(url, "", url)
+	if err != nil {
+		log.Printf("error: %s", err)
+		die()
+	}
+
+	_, err = io.Copy(os.Stdout, ws)
+	return err
+}

client/pprof_listen.go

@@ -0,0 +1,68 @@
+package client
+
+import (
+	"context"
+	"errors"
+	"log"
+	"os"
+
+	"github.com/zrepl/zrepl/cli"
+	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/daemon"
+)
+
+var pprofListenCmd struct {
+	daemon.PprofServerControlMsg
+}
+
+var PprofListenCmd = &cli.Subcommand{
+	Use:   "listen off | [on TCP_LISTEN_ADDRESS]",
+	Short: "start a http server exposing go-tool-compatible profiling endpoints at TCP_LISTEN_ADDRESS",
+	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
+		if len(args) < 1 {
+			goto enargs
+		}
+		switch args[0] {
+		case "on":
+			pprofListenCmd.Run = true
+			if len(args) != 2 {
+				return errors.New("must specify TCP_LISTEN_ADDRESS as second positional argument")
+			}
+			pprofListenCmd.HttpListenAddress = args[1]
+		case "off":
+			if len(args) != 1 {
+				goto enargs
+			}
+			pprofListenCmd.Run = false
+		}
+
+		RunPProf(subcommand.Config())
+		return nil
+	enargs:
+		return errors.New("invalid number of positional arguments")
+
+	},
+}
+
+func RunPProf(conf *config.Config) {
+	log := log.New(os.Stderr, "", 0)
+
+	die := func() {
+		log.Printf("exiting after error")
+		os.Exit(1)
+	}
+
+	log.Printf("connecting to zrepl daemon")
+
+	httpc, err := controlHttpClient(conf.Global.Control.SockPath)
+	if err != nil {
+		log.Printf("error creating http client: %s", err)
+		die()
+	}
+	err = jsonRequestResponse(httpc, daemon.ControlJobEndpointPProf, pprofListenCmd.PprofServerControlMsg, struct{}{})
+	if err != nil {
+		log.Printf("error sending control message: %s", err)
+		die()
+	}
+	log.Printf("finished")
+}

client/signal.go

@@ -1,6 +1,8 @@
 package client
 
 import (
+	"context"
+
 	"github.com/pkg/errors"
 
 	"github.com/zrepl/zrepl/cli"
@@ -11,7 +13,7 @@ import (
 var SignalCmd = &cli.Subcommand{
 	Use:   "signal [wakeup|reset] JOB",
 	Short: "wake up a job from wait state or abort its current invocation",
-	Run: func(subcommand *cli.Subcommand, args []string) error {
+	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
 		return runSignalCmd(subcommand.Config(), args)
 	},
 }

client/status.go

@@ -1,810 +0,0 @@
-package client
-
-import (
-	"fmt"
-	"io"
-	"math"
-	"net/http"
-	"os"
-	"sort"
-	"strings"
-	"sync"
-	"time"
-
-	// tcell is the termbox-compatible library for abstracting away escape sequences, etc.
-	// as of tcell#252, the number of default distributed terminals is relatively limited
-	// additional terminal definitions can be included via side-effect import
-	// See https://github.com/gdamore/tcell/blob/master/terminfo/base/base.go
-	// See https://github.com/gdamore/tcell/issues/252#issuecomment-533836078
-	"github.com/gdamore/tcell/termbox"
-	_ "github.com/gdamore/tcell/terminfo/s/screen" // tmux on FreeBSD 11 & 12 without ncurses
-
-	"github.com/pkg/errors"
-	"github.com/spf13/pflag"
-	"github.com/zrepl/yaml-config"
-
-	"github.com/zrepl/zrepl/cli"
-	"github.com/zrepl/zrepl/daemon"
-	"github.com/zrepl/zrepl/daemon/job"
-	"github.com/zrepl/zrepl/daemon/pruner"
-	"github.com/zrepl/zrepl/daemon/snapper"
-	"github.com/zrepl/zrepl/replication/report"
-)
-
-type byteProgressMeasurement struct {
-	time time.Time
-	val  int64
-}
-
-type bytesProgressHistory struct {
-	last        *byteProgressMeasurement // pointer as poor man's optional
-	changeCount int
-	lastChange  time.Time
-	bpsAvg      float64
-}
-
-func (p *bytesProgressHistory) Update(currentVal int64) (bytesPerSecondAvg int64, changeCount int) {
-
-	if p.last == nil {
-		p.last = &byteProgressMeasurement{
-			time: time.Now(),
-			val:  currentVal,
-		}
-		return 0, 0
-	}
-
-	if p.last.val != currentVal {
-		p.changeCount++
-		p.lastChange = time.Now()
-	}
-
-	if time.Since(p.lastChange) > 3*time.Second {
-		p.last = nil
-		return 0, 0
-	}
-
-	deltaV := currentVal - p.last.val
-	deltaT := time.Since(p.last.time)
-	rate := float64(deltaV) / deltaT.Seconds()
-
-	factor := 0.3
-	p.bpsAvg = (1-factor)*p.bpsAvg + factor*rate
-
-	p.last.time = time.Now()
-	p.last.val = currentVal
-
-	return int64(p.bpsAvg), p.changeCount
-}
-
-type tui struct {
-	x, y   int
-	indent int
-
-	lock   sync.Mutex //For report and error
-	report map[string]*job.Status
-	err    error
-
-	jobFilter string
-
-	replicationProgress map[string]*bytesProgressHistory // by job name
-}
-
-func newTui() tui {
-	return tui{
-		replicationProgress: make(map[string]*bytesProgressHistory),
-	}
-}
-
-const INDENT_MULTIPLIER = 4
-
-func (t *tui) moveLine(dl int, col int) {
-	t.y += dl
-	t.x = t.indent*INDENT_MULTIPLIER + col
-}
-
-func (t *tui) write(text string) {
-	for _, c := range text {
-		if c == '\n' {
-			t.newline()
-			continue
-		}
-		termbox.SetCell(t.x, t.y, c, termbox.ColorDefault, termbox.ColorDefault)
-		t.x += 1
-	}
-}
-
-func (t *tui) printf(text string, a ...interface{}) {
-	t.write(fmt.Sprintf(text, a...))
-}
-
-func wrap(s string, width int) string {
-	var b strings.Builder
-	for len(s) > 0 {
-		rem := width
-		if rem > len(s) {
-			rem = len(s)
-		}
-		if idx := strings.IndexAny(s, "\n\r"); idx != -1 && idx < rem {
-			rem = idx + 1
-		}
-		untilNewline := strings.TrimRight(s[:rem], "\n\r")
-		s = s[rem:]
-		if len(untilNewline) == 0 {
-			continue
-		}
-		b.WriteString(untilNewline)
-		b.WriteString("\n")
-	}
-	return strings.TrimRight(b.String(), "\n\r")
-}
-
-func (t *tui) printfDrawIndentedAndWrappedIfMultiline(format string, a ...interface{}) {
-	whole := fmt.Sprintf(format, a...)
-	width, _ := termbox.Size()
-	if !strings.ContainsAny(whole, "\n\r") && t.x+len(whole) <= width {
-		t.printf(format, a...)
-	} else {
-		t.addIndent(1)
-		t.newline()
-		t.write(wrap(whole, width-INDENT_MULTIPLIER*t.indent))
-		t.addIndent(-1)
-	}
-}
-
-func (t *tui) newline() {
-	t.moveLine(1, 0)
-}
-
-func (t *tui) setIndent(indent int) {
-	t.indent = indent
-	t.moveLine(0, 0)
-}
-
-func (t *tui) addIndent(indent int) {
-	t.indent += indent
-	t.moveLine(0, 0)
-}
-
-var statusFlags struct {
-	Raw bool
-	Job string
-}
-
-var StatusCmd = &cli.Subcommand{
-	Use:   "status",
-	Short: "show job activity or dump as JSON for monitoring",
-	SetupFlags: func(f *pflag.FlagSet) {
-		f.BoolVar(&statusFlags.Raw, "raw", false, "dump raw status description from zrepl daemon")
-		f.StringVar(&statusFlags.Job, "job", "", "only dump specified job")
-	},
-	Run: runStatus,
-}
-
-func runStatus(s *cli.Subcommand, args []string) error {
-	httpc, err := controlHttpClient(s.Config().Global.Control.SockPath)
-	if err != nil {
-		return err
-	}
-
-	if statusFlags.Raw {
-		resp, err := httpc.Get("http://unix" + daemon.ControlJobEndpointStatus)
-		if err != nil {
-			return err
-		}
-		defer resp.Body.Close()
-		if resp.StatusCode != http.StatusOK {
-			fmt.Fprintf(os.Stderr, "Received error response:\n")
-			_, err := io.CopyN(os.Stderr, resp.Body, 4096)
-			if err != nil {
-				return err
-			}
-			return errors.Errorf("exit")
-		}
-		if _, err := io.Copy(os.Stdout, resp.Body); err != nil {
-			return err
-		}
-		return nil
-	}
-
-	t := newTui()
-	t.lock.Lock()
-	t.err = errors.New("Got no report yet")
-	t.lock.Unlock()
-	t.jobFilter = statusFlags.Job
-
-	err = termbox.Init()
-	if err != nil {
-		return err
-	}
-	defer termbox.Close()
-
-	update := func() {
-		var m daemon.Status
-
-		err2 := jsonRequestResponse(httpc, daemon.ControlJobEndpointStatus,
-			struct{}{},
-			&m,
-		)
-
-		t.lock.Lock()
-		t.err = err2
-		t.report = m.Jobs
-		t.lock.Unlock()
-		t.draw()
-	}
-	update()
-
-	ticker := time.NewTicker(500 * time.Millisecond)
-	defer ticker.Stop()
-	go func() {
-		for range ticker.C {
-			update()
-		}
-	}()
-
-	termbox.HideCursor()
-	termbox.Clear(termbox.ColorDefault, termbox.ColorDefault)
-
-loop:
-	for {
-		switch ev := termbox.PollEvent(); ev.Type {
-		case termbox.EventKey:
-			switch ev.Key {
-			case termbox.KeyEsc:
-				break loop
-			case termbox.KeyCtrlC:
-				break loop
-			}
-		case termbox.EventResize:
-			t.draw()
-		}
-	}
-
-	return nil
-
-}
-
-func (t *tui) getReplicationProgressHistory(jobName string) *bytesProgressHistory {
-	p, ok := t.replicationProgress[jobName]
-	if !ok {
-		p = &bytesProgressHistory{}
-		t.replicationProgress[jobName] = p
-	}
-	return p
-}
-
-func (t *tui) draw() {
-	t.lock.Lock()
-	defer t.lock.Unlock()
-
-	termbox.Clear(termbox.ColorDefault, termbox.ColorDefault)
-	t.x = 0
-	t.y = 0
-	t.indent = 0
-
-	if t.err != nil {
-		t.write(t.err.Error())
-	} else {
-		//Iterate over map in alphabetical order
-		keys := make([]string, 0, len(t.report))
-		for k := range t.report {
-			if len(k) == 0 || daemon.IsInternalJobName(k) { //Internal job
-				continue
-			}
-			if t.jobFilter != "" && k != t.jobFilter {
-				continue
-			}
-			keys = append(keys, k)
-		}
-		sort.Strings(keys)
-
-		if len(keys) == 0 {
-			t.setIndent(0)
-			t.printf("no jobs to display")
-			t.newline()
-			termbox.Flush()
-			return
-		}
-
-		for _, k := range keys {
-			v := t.report[k]
-
-			t.setIndent(0)
-
-			t.printf("Job: %s", k)
-			t.setIndent(1)
-			t.newline()
-			t.printf("Type: %s", v.Type)
-			t.setIndent(1)
-			t.newline()
-
-			if v.Type == job.TypePush || v.Type == job.TypePull {
-				activeStatus, ok := v.JobSpecific.(*job.ActiveSideStatus)
-				if !ok || activeStatus == nil {
-					t.printf("ActiveSideStatus is null")
-					t.newline()
-					continue
-				}
-
-				t.printf("Replication:")
-				t.newline()
-				t.addIndent(1)
-				t.renderReplicationReport(activeStatus.Replication, t.getReplicationProgressHistory(k))
-				t.addIndent(-1)
-
-				t.printf("Pruning Sender:")
-				t.newline()
-				t.addIndent(1)
-				t.renderPrunerReport(activeStatus.PruningSender)
-				t.addIndent(-1)
-
-				t.printf("Pruning Receiver:")
-				t.newline()
-				t.addIndent(1)
-				t.renderPrunerReport(activeStatus.PruningReceiver)
-				t.addIndent(-1)
-
-				if v.Type == job.TypePush {
-					t.printf("Snapshotting:")
-					t.newline()
-					t.addIndent(1)
-					t.renderSnapperReport(activeStatus.Snapshotting)
-					t.addIndent(-1)
-				}
-
-			} else if v.Type == job.TypeSnap {
-				snapStatus, ok := v.JobSpecific.(*job.SnapJobStatus)
-				if !ok || snapStatus == nil {
-					t.printf("SnapJobStatus is null")
-					t.newline()
-					continue
-				}
-				t.printf("Pruning snapshots:")
-				t.newline()
-				t.addIndent(1)
-				t.renderPrunerReport(snapStatus.Pruning)
-				t.addIndent(-1)
-				t.printf("Snapshotting:")
-				t.newline()
-				t.addIndent(1)
-				t.renderSnapperReport(snapStatus.Snapshotting)
-				t.addIndent(-1)
-			} else if v.Type == job.TypeSource {
-
-				st := v.JobSpecific.(*job.PassiveStatus)
-				t.printf("Snapshotting:\n")
-				t.addIndent(1)
-				t.renderSnapperReport(st.Snapper)
-				t.addIndent(-1)
-
-			} else {
-				t.printf("No status representation for job type '%s', dumping as YAML", v.Type)
-				t.newline()
-				asYaml, err := yaml.Marshal(v.JobSpecific)
-				if err != nil {
-					t.printf("Error marshaling status to YAML: %s", err)
-					t.newline()
-					continue
-				}
-				t.write(string(asYaml))
-				t.newline()
-				continue
-			}
-		}
-	}
-	termbox.Flush()
-}
-
-func (t *tui) renderReplicationReport(rep *report.Report, history *bytesProgressHistory) {
-	if rep == nil {
-		t.printf("...\n")
-		return
-	}
-
-	if rep.WaitReconnectError != nil {
-		t.printfDrawIndentedAndWrappedIfMultiline("Connectivity: %s", rep.WaitReconnectError)
-		t.newline()
-	}
-	if !rep.WaitReconnectSince.IsZero() {
-		delta := time.Until(rep.WaitReconnectUntil).Round(time.Second)
-		if rep.WaitReconnectUntil.IsZero() || delta > 0 {
-			var until string
-			if rep.WaitReconnectUntil.IsZero() {
-				until = "waiting indefinitely"
-			} else {
-				until = fmt.Sprintf("hard fail in %s @ %s", delta, rep.WaitReconnectUntil)
-			}
-			t.printfDrawIndentedAndWrappedIfMultiline("Connectivity: reconnecting with exponential backoff (since %s) (%s)",
-				rep.WaitReconnectSince, until)
-		} else {
-			t.printfDrawIndentedAndWrappedIfMultiline("Connectivity: reconnects reached hard-fail timeout @ %s", rep.WaitReconnectUntil)
-		}
-		t.newline()
-	}
-
-	// TODO visualize more than the latest attempt by folding all attempts into one
-	if len(rep.Attempts) == 0 {
-		t.printf("no attempts made yet")
-		return
-	} else {
-		t.printf("Attempt #%d", len(rep.Attempts))
-		if len(rep.Attempts) > 1 {
-			t.printf(". Previous attempts failed with the following statuses:")
-			t.newline()
-			t.addIndent(1)
-			for i, a := range rep.Attempts[:len(rep.Attempts)-1] {
-				t.printfDrawIndentedAndWrappedIfMultiline("#%d: %s (failed at %s) (ran %s)", i+1, a.State, a.FinishAt, a.FinishAt.Sub(a.StartAt))
-				t.newline()
-			}
-			t.addIndent(-1)
-		} else {
-			t.newline()
-		}
-	}
-
-	latest := rep.Attempts[len(rep.Attempts)-1]
-	sort.Slice(latest.Filesystems, func(i, j int) bool {
-		return latest.Filesystems[i].Info.Name < latest.Filesystems[j].Info.Name
-	})
-
-	t.printf("Status: %s", latest.State)
-	t.newline()
-	if latest.State == report.AttemptPlanningError {
-		t.printf("Problem: ")
-		t.printfDrawIndentedAndWrappedIfMultiline("%s", latest.PlanError)
-		t.newline()
-	} else if latest.State == report.AttemptFanOutError {
-		t.printf("Problem: one or more of the filesystems encountered errors")
-		t.newline()
-	}
-
-	if latest.State != report.AttemptPlanning && latest.State != report.AttemptPlanningError {
-		// Draw global progress bar
-		// Progress: [---------------]
-		expected, replicated, containsInvalidSizeEstimates := latest.BytesSum()
-		rate, changeCount := history.Update(replicated)
-		eta := time.Duration(0)
-		if rate > 0 {
-			eta = time.Duration((expected-replicated)/rate) * time.Second
-		}
-		t.write("Progress: ")
-		t.drawBar(50, replicated, expected, changeCount)
-		t.write(fmt.Sprintf(" %s / %s @ %s/s", ByteCountBinary(replicated), ByteCountBinary(expected), ByteCountBinary(rate)))
-		if eta != 0 {
-			t.write(fmt.Sprintf(" (%s remaining)", humanizeDuration(eta)))
-		}
-		t.newline()
-		if containsInvalidSizeEstimates {
-			t.write("NOTE: not all steps could be size-estimated, total estimate is likely imprecise!")
-			t.newline()
-		}
-
-		var maxFSLen int
-		for _, fs := range latest.Filesystems {
-			if len(fs.Info.Name) > maxFSLen {
-				maxFSLen = len(fs.Info.Name)
-			}
-		}
-		for _, fs := range latest.Filesystems {
-			t.printFilesystemStatus(fs, false, maxFSLen) // FIXME bring 'active' flag back
-		}
-
-	}
-
-}
-
-func (t *tui) renderPrunerReport(r *pruner.Report) {
-	if r == nil {
-		t.printf("...\n")
-		return
-	}
-
-	state, err := pruner.StateString(r.State)
-	if err != nil {
-		t.printf("Status: %q (parse error: %q)\n", r.State, err)
-		return
-	}
-
-	t.printf("Status: %s", state)
-	t.newline()
-
-	if r.Error != "" {
-		t.printf("Error: %s\n", r.Error)
-	}
-
-	type commonFS struct {
-		*pruner.FSReport
-		completed bool
-	}
-	all := make([]commonFS, 0, len(r.Pending)+len(r.Completed))
-	for i := range r.Pending {
-		all = append(all, commonFS{&r.Pending[i], false})
-	}
-	for i := range r.Completed {
-		all = append(all, commonFS{&r.Completed[i], true})
-	}
-
-	switch state {
-	case pruner.Plan:
-		fallthrough
-	case pruner.PlanErr:
-		return
-	}
-
-	if len(all) == 0 {
-		t.printf("nothing to do\n")
-		return
-	}
-
-	var totalDestroyCount, completedDestroyCount int
-	var maxFSname int
-	for _, fs := range all {
-		totalDestroyCount += len(fs.DestroyList)
-		if fs.completed {
-			completedDestroyCount += len(fs.DestroyList)
-		}
-		if maxFSname < len(fs.Filesystem) {
-			maxFSname = len(fs.Filesystem)
-		}
-	}
-
-	// global progress bar
-	progress := int(math.Round(80 * float64(completedDestroyCount) / float64(totalDestroyCount)))
-	t.write("Progress: ")
-	t.write("[")
-	t.write(times("=", progress))
-	t.write(">")
-	t.write(times("-", 80-progress))
-	t.write("]")
-	t.printf(" %d/%d snapshots", completedDestroyCount, totalDestroyCount)
-	t.newline()
-
-	sort.SliceStable(all, func(i, j int) bool {
-		return strings.Compare(all[i].Filesystem, all[j].Filesystem) == -1
-	})
-
-	// Draw a table-like representation of 'all'
-	for _, fs := range all {
-		t.write(rightPad(fs.Filesystem, maxFSname, " "))
-		t.write(" ")
-		if !fs.SkipReason.NotSkipped() {
-			t.printf("skipped: %s\n", fs.SkipReason)
-			continue
-		}
-		if fs.LastError != "" {
-			if strings.ContainsAny(fs.LastError, "\r\n") {
-				t.printf("ERROR:")
-				t.printfDrawIndentedAndWrappedIfMultiline("%s\n", fs.LastError)
-			} else {
-				t.printfDrawIndentedAndWrappedIfMultiline("ERROR: %s\n", fs.LastError)
-			}
-			t.newline()
-			continue
-		}
-
-		pruneRuleActionStr := fmt.Sprintf("(destroy %d of %d snapshots)",
-			len(fs.DestroyList), len(fs.SnapshotList))
-
-		if fs.completed {
-			t.printf("Completed  %s\n", pruneRuleActionStr)
-			continue
-		}
-
-		t.write("Pending    ") // whitespace is padding 10
-		if len(fs.DestroyList) == 1 {
-			t.write(fs.DestroyList[0].Name)
-		} else {
-			t.write(pruneRuleActionStr)
-		}
-		t.newline()
-	}
-
-}
-
-func (t *tui) renderSnapperReport(r *snapper.Report) {
-	if r == nil {
-		t.printf("<snapshot type does not have a report>\n")
-		return
-	}
-
-	t.printf("Status: %s", r.State)
-	t.newline()
-
-	if r.Error != "" {
-		t.printf("Error: %s\n", r.Error)
-	}
-	if !r.SleepUntil.IsZero() {
-		t.printf("Sleep until: %s\n", r.SleepUntil)
-	}
-
-	sort.Slice(r.Progress, func(i, j int) bool {
-		return strings.Compare(r.Progress[i].Path, r.Progress[j].Path) == -1
-	})
-
-	t.addIndent(1)
-	defer t.addIndent(-1)
-	dur := func(d time.Duration) string {
-		return d.Round(100 * time.Millisecond).String()
-	}
-
-	type row struct {
-		path, state, duration, remainder, hookReport string
-	}
-	var widths struct {
-		path, state, duration int
-	}
-	rows := make([]*row, len(r.Progress))
-	for i, fs := range r.Progress {
-		r := &row{
-			path:  fs.Path,
-			state: fs.State.String(),
-		}
-		if fs.HooksHadError {
-			r.hookReport = fs.Hooks // FIXME render here, not in daemon
-		}
-		switch fs.State {
-		case snapper.SnapPending:
-			r.duration = "..."
-			r.remainder = ""
-		case snapper.SnapStarted:
-			r.duration = dur(time.Since(fs.StartAt))
-			r.remainder = fmt.Sprintf("snap name: %q", fs.SnapName)
-		case snapper.SnapDone:
-			fallthrough
-		case snapper.SnapError:
-			r.duration = dur(fs.DoneAt.Sub(fs.StartAt))
-			r.remainder = fmt.Sprintf("snap name: %q", fs.SnapName)
-		}
-		rows[i] = r
-		if len(r.path) > widths.path {
-			widths.path = len(r.path)
-		}
-		if len(r.state) > widths.state {
-			widths.state = len(r.state)
-		}
-		if len(r.duration) > widths.duration {
-			widths.duration = len(r.duration)
-		}
-	}
-
-	for _, r := range rows {
-		path := rightPad(r.path, widths.path, " ")
-		state := rightPad(r.state, widths.state, " ")
-		duration := rightPad(r.duration, widths.duration, " ")
-		t.printf("%s %s %s", path, state, duration)
-		t.printfDrawIndentedAndWrappedIfMultiline(" %s", r.remainder)
-		if r.hookReport != "" {
-			t.printfDrawIndentedAndWrappedIfMultiline("%s", r.hookReport)
-		}
-		t.newline()
-	}
-
-}
-
-func times(str string, n int) (out string) {
-	for i := 0; i < n; i++ {
-		out += str
-	}
-	return
-}
-
-func rightPad(str string, length int, pad string) string {
-	if len(str) > length {
-		return str[:length]
-	}
-	return str + strings.Repeat(pad, length-len(str))
-}
-
-var arrowPositions = `>\|/`
-
-// changeCount = 0 indicates stall / no progress
-func (t *tui) drawBar(length int, bytes, totalBytes int64, changeCount int) {
-	var completedLength int
-	if totalBytes > 0 {
-		completedLength = int(int64(length) * bytes / totalBytes)
-		if completedLength > length {
-			completedLength = length
-		}
-	} else if totalBytes == bytes {
-		completedLength = length
-	}
-
-	t.write("[")
-	t.write(times("=", completedLength))
-	t.write(string(arrowPositions[changeCount%len(arrowPositions)]))
-	t.write(times("-", length-completedLength))
-	t.write("]")
-}
-
-func (t *tui) printFilesystemStatus(rep *report.FilesystemReport, active bool, maxFS int) {
-
-	expected, replicated, containsInvalidSizeEstimates := rep.BytesSum()
-	sizeEstimationImpreciseNotice := ""
-	if containsInvalidSizeEstimates {
-		sizeEstimationImpreciseNotice = " (some steps lack size estimation)"
-	}
-	if rep.CurrentStep < len(rep.Steps) && rep.Steps[rep.CurrentStep].Info.BytesExpected == 0 {
-		sizeEstimationImpreciseNotice = " (step lacks size estimation)"
-	}
-
-	status := fmt.Sprintf("%s (step %d/%d, %s/%s)%s",
-		strings.ToUpper(string(rep.State)),
-		rep.CurrentStep, len(rep.Steps),
-		ByteCountBinary(replicated), ByteCountBinary(expected),
-		sizeEstimationImpreciseNotice,
-	)
-
-	activeIndicator := " "
-	if active {
-		activeIndicator = "*"
-	}
-	t.printf("%s %s %s ",
-		activeIndicator,
-		rightPad(rep.Info.Name, maxFS, " "),
-		status)
-
-	next := ""
-	if err := rep.Error(); err != nil {
-		next = err.Err
-	} else if rep.State != report.FilesystemDone {
-		if nextStep := rep.NextStep(); nextStep != nil {
-			if nextStep.IsIncremental() {
-				next = fmt.Sprintf("next: %s => %s", nextStep.Info.From, nextStep.Info.To)
-			} else {
-				next = fmt.Sprintf("next: full send %s", nextStep.Info.To)
-			}
-			attribs := []string{}
-
-			if nextStep.Info.Resumed {
-				attribs = append(attribs, "resumed")
-			}
-
-			attribs = append(attribs, fmt.Sprintf("encrypted=%s", nextStep.Info.Encrypted))
-
-			next += fmt.Sprintf(" (%s)", strings.Join(attribs, ", "))
-		} else {
-			next = "" // individual FSes may still be in planning state
-		}
-
-	}
-	t.printfDrawIndentedAndWrappedIfMultiline("%s", next)
-
-	t.newline()
-}
-
-func ByteCountBinary(b int64) string {
-	const unit = 1024
-	if b < unit {
-		return fmt.Sprintf("%d B", b)
-	}
-	div, exp := int64(unit), 0
-	for n := b / unit; n >= unit; n /= unit {
-		div *= unit
-		exp++
-	}
-	return fmt.Sprintf("%.1f %ciB", float64(b)/float64(div), "KMGTPE"[exp])
-}
-
-func humanizeDuration(duration time.Duration) string {
-	days := int64(duration.Hours() / 24)
-	hours := int64(math.Mod(duration.Hours(), 24))
-	minutes := int64(math.Mod(duration.Minutes(), 60))
-	seconds := int64(math.Mod(duration.Seconds(), 60))
-
-	var parts []string
-
-	force := false
-	chunks := []int64{days, hours, minutes, seconds}
-	for i, chunk := range chunks {
-		if force || chunk > 0 {
-			padding := 0
-			if force {
-				padding = 2
-			}
-			parts = append(parts, fmt.Sprintf("%*d%c", padding, chunk, "dhms"[i]))
-			force = true
-		}
-	}
-
-	return strings.Join(parts, " ")
-}

client/status/client/client.go

@@ -0,0 +1,111 @@
+package client
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"io"
+	"net"
+	"net/http"
+
+	"github.com/pkg/errors"
+
+	"github.com/zrepl/zrepl/daemon"
+)
+
+type Client struct {
+	h *http.Client
+}
+
+func New(network, addr string) (*Client, error) {
+	httpc, err := makeControlHttpClient(func(_ context.Context) (net.Conn, error) { return net.Dial(network, addr) })
+	if err != nil {
+		return nil, err
+	}
+	return &Client{httpc}, nil
+}
+
+func (c *Client) Status() (s daemon.Status, _ error) {
+	err := jsonRequestResponse(c.h, daemon.ControlJobEndpointStatus,
+		struct{}{},
+		&s,
+	)
+	return s, err
+}
+
+func (c *Client) StatusRaw() ([]byte, error) {
+	var r json.RawMessage
+	err := jsonRequestResponse(c.h, daemon.ControlJobEndpointStatus, struct{}{}, &r)
+	if err != nil {
+		return nil, err
+	}
+	return r, nil
+}
+
+func (c *Client) signal(job, sig string) error {
+	return jsonRequestResponse(c.h, daemon.ControlJobEndpointSignal,
+		struct {
+			Name string
+			Op   string
+		}{
+			Name: job,
+			Op:   sig,
+		},
+		struct{}{},
+	)
+}
+
+func (c *Client) SignalWakeup(job string) error {
+	return c.signal(job, "wakeup")
+}
+
+func (c *Client) SignalReset(job string) error {
+	return c.signal(job, "reset")
+}
+
+func makeControlHttpClient(dialfunc func(context.Context) (net.Conn, error)) (client *http.Client, err error) {
+	return &http.Client{
+		Transport: &http.Transport{
+			DialContext: func(ctx context.Context, _, _ string) (net.Conn, error) {
+				return dialfunc(ctx)
+			},
+		},
+	}, nil
+}
+
+func jsonRequestResponse(c *http.Client, endpoint string, req interface{}, res interface{}) error {
+	var buf bytes.Buffer
+	encodeErr := json.NewEncoder(&buf).Encode(req)
+	if encodeErr != nil {
+		return encodeErr
+	}
+
+	hreq, err := http.NewRequest("POST", "http://unix"+endpoint, &buf)
+	if err != nil {
+		return err
+	}
+	hreq.Header.Set("Content-Type", "application/json")
+	// Prevent EOF errors when client request frequency and server keepalive close are at the same time.
+	// Found this by watching http.Server.ConnState changes, then found
+	// https://stackoverflow.com/questions/17714494/golang-http-request-results-in-eof-errors-when-making-multiple-requests-successi
+	// Note: The issue seems even more prounounced with local TCP sockets than unix domain sockets. So, I used that for debugging.
+	hreq.Close = true
+	resp, err := c.Do(hreq)
+	if err != nil {
+		return err
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode != http.StatusOK {
+		var msg bytes.Buffer
+		_, _ = io.CopyN(&msg, resp.Body, 4096) // ignore error, just display what we got
+		return errors.Errorf("%s", msg.String())
+	}
+
+	decodeError := json.NewDecoder(resp.Body).Decode(&res)
+	if decodeError != nil {
+		return decodeError
+	}
+
+	return nil
+}

client/status/status.go

@@ -0,0 +1,97 @@
+package status
+
+import (
+	"context"
+	"os"
+	"time"
+
+	"github.com/mattn/go-isatty"
+	"github.com/pkg/errors"
+	"github.com/spf13/pflag"
+
+	"github.com/zrepl/zrepl/cli"
+	"github.com/zrepl/zrepl/client/status/client"
+	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/daemon"
+	"github.com/zrepl/zrepl/util/choices"
+)
+
+type Client interface {
+	Status() (daemon.Status, error)
+	StatusRaw() ([]byte, error)
+	SignalWakeup(job string) error
+	SignalReset(job string) error
+}
+
+type statusFlags struct {
+	Mode  choices.Choices
+	Job   string
+	Delay time.Duration
+}
+
+var statusv2Flags statusFlags
+
+type statusv2Mode int
+
+const (
+	StatusV2ModeInteractive statusv2Mode = 1 + iota
+	StatusV2ModeDump
+	StatusV2ModeRaw
+	StatusV2ModeLegacy
+)
+
+var Subcommand = &cli.Subcommand{
+	Use:   "status",
+	Short: "retrieve & display daemon status information",
+	SetupFlags: func(f *pflag.FlagSet) {
+		statusv2Flags.Mode.Init(
+			"interactive", StatusV2ModeInteractive,
+			"dump", StatusV2ModeDump,
+			"raw", StatusV2ModeRaw,
+			"legacy", StatusV2ModeLegacy,
+		)
+		statusv2Flags.Mode.SetTypeString("mode")
+		statusv2Flags.Mode.SetDefaultValue(StatusV2ModeInteractive)
+		f.Var(&statusv2Flags.Mode, "mode", statusv2Flags.Mode.Usage())
+		f.StringVar(&statusv2Flags.Job, "job", "", "only show specified job (works in \"dump\" and \"interactive\" mode)")
+		f.DurationVarP(&statusv2Flags.Delay, "delay", "d", 1*time.Second, "use -d 3s for 3 seconds delay (minimum delay is 1s)")
+	},
+	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
+		return runStatusV2Command(ctx, subcommand.Config(), args)
+	},
+}
+
+func runStatusV2Command(ctx context.Context, config *config.Config, args []string) error {
+
+	c, err := client.New("unix", config.Global.Control.SockPath)
+	if err != nil {
+		return errors.Wrapf(err, "connect to daemon socket at %q", config.Global.Control.SockPath)
+	}
+
+	mode := statusv2Flags.Mode.Value().(statusv2Mode)
+
+	if !isatty.IsTerminal(os.Stdout.Fd()) && mode != StatusV2ModeDump && mode != StatusV2ModeRaw {
+		dumpmode, err := statusv2Flags.Mode.InputForChoice(StatusV2ModeDump)
+		if err != nil {
+			panic(err)
+		}
+		rawmode, err := statusv2Flags.Mode.InputForChoice(StatusV2ModeRaw)
+		if err != nil {
+			panic(err)
+		}
+		return errors.Errorf("error: stdout is not a tty, please use --mode %s or --mode %s", dumpmode, rawmode)
+	}
+
+	switch mode {
+	case StatusV2ModeInteractive:
+		return interactive(c, statusv2Flags)
+	case StatusV2ModeDump:
+		return dump(c, statusv2Flags.Job)
+	case StatusV2ModeRaw:
+		return raw(c)
+	case StatusV2ModeLegacy:
+		return legacy(c, statusv2Flags)
+	default:
+		panic("unreachable")
+	}
+}

client/status/status_dump.go

@@ -0,0 +1,70 @@
+package status
+
+import (
+	"fmt"
+	"os"
+	"strings"
+
+	"github.com/gdamore/tcell"
+	"github.com/mattn/go-isatty"
+	"github.com/pkg/errors"
+
+	"github.com/zrepl/zrepl/client/status/viewmodel"
+)
+
+func dump(c Client, job string) error {
+	s, err := c.Status()
+	if err != nil {
+		return err
+	}
+
+	if job != "" {
+		if _, ok := s.Jobs[job]; !ok {
+			return errors.Errorf("job %q not found", job)
+		}
+	}
+
+	width := (1 << 31) - 1
+	wrap := false
+	hline := strings.Repeat("-", 80)
+	if isatty.IsTerminal(os.Stdout.Fd()) {
+		wrap = true
+		screen, err := tcell.NewScreen()
+		if err != nil {
+			return errors.Wrap(err, "get terminal dimensions")
+		}
+		if err := screen.Init(); err != nil {
+			return errors.Wrap(err, "init screen")
+		}
+		width, _ = screen.Size()
+		screen.Fini()
+		hline = strings.Repeat("-", width)
+	}
+
+	m := viewmodel.New()
+	params := viewmodel.Params{
+		Report:                  s.Jobs,
+		ReportFetchError:        nil,
+		SelectedJob:             nil,
+		FSFilter:                func(s string) bool { return true },
+		DetailViewWidth:         width,
+		DetailViewWrap:          wrap,
+		ShortKeybindingOverview: "",
+	}
+	m.Update(params)
+	for _, j := range m.Jobs() {
+		if job != "" && j.Name() != job {
+			continue
+		}
+		params.SelectedJob = j
+		m.Update(params)
+		fmt.Println(m.SelectedJob().FullDescription())
+		if job != "" {
+			return nil
+		} else {
+			fmt.Println(hline)
+		}
+	}
+
+	return nil
+}

client/status/status_interactive.go

@@ -0,0 +1,347 @@
+package status
+
+import (
+	"fmt"
+	"regexp"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/gdamore/tcell/v2"
+	tview "gitlab.com/tslocum/cview"
+
+	"github.com/zrepl/zrepl/client/status/viewmodel"
+)
+
+func interactive(c Client, flag statusFlags) error {
+
+	// Set this so we don't overwrite the default terminal colors
+	// See https://github.com/rivo/tview/blob/master/styles.go
+	tview.Styles.PrimitiveBackgroundColor = tcell.ColorDefault
+	tview.Styles.ContrastBackgroundColor = tcell.ColorDefault
+	tview.Styles.PrimaryTextColor = tcell.ColorDefault
+	tview.Styles.BorderColor = tcell.ColorDefault
+	app := tview.NewApplication()
+
+	jobDetailSplit := tview.NewFlex()
+	jobMenu := tview.NewTreeView()
+	jobMenuRoot := tview.NewTreeNode("jobs")
+	jobMenuRoot.SetSelectable(true)
+	jobMenu.SetRoot(jobMenuRoot)
+	jobMenu.SetCurrentNode(jobMenuRoot)
+	jobMenu.SetSelectedTextColor(tcell.ColorGreen)
+	jobTextDetail := tview.NewTextView()
+	jobTextDetail.SetWrap(false)
+
+	jobMenu.SetBorder(true)
+	jobTextDetail.SetBorder(true)
+
+	toolbarSplit := tview.NewFlex()
+	toolbarSplit.SetDirection(tview.FlexRow)
+	inputBarContainer := tview.NewFlex()
+	fsFilterInput := tview.NewInputField()
+	fsFilterInput.SetBorder(false)
+	fsFilterInput.SetFieldBackgroundColor(tcell.ColorDefault)
+	inputBarLabel := tview.NewTextView()
+	inputBarLabel.SetText("[::b]FILTER ")
+	inputBarLabel.SetDynamicColors(true)
+	inputBarContainer.AddItem(inputBarLabel, 7, 1, false)
+	inputBarContainer.AddItem(fsFilterInput, 0, 10, false)
+	toolbarSplit.AddItem(inputBarContainer, 1, 0, false)
+	toolbarSplit.AddItem(jobDetailSplit, 0, 10, false)
+
+	bottombar := tview.NewFlex()
+	bottombar.SetDirection(tview.FlexColumn)
+	bottombarDateView := tview.NewTextView()
+	bottombar.AddItem(bottombarDateView, len(time.Now().String()), 0, false)
+	bottomBarStatus := tview.NewTextView()
+	bottomBarStatus.SetDynamicColors(true)
+	bottomBarStatus.SetTextAlign(tview.AlignRight)
+	bottombar.AddItem(bottomBarStatus, 0, 10, false)
+	toolbarSplit.AddItem(bottombar, 1, 0, false)
+
+	tabbableWithJobMenu := []tview.Primitive{jobMenu, jobTextDetail, fsFilterInput}
+	tabbableWithoutJobMenu := []tview.Primitive{jobTextDetail, fsFilterInput}
+	var tabbable []tview.Primitive
+	tabbableActiveIndex := 0
+	tabbableRedraw := func() {
+		if len(tabbable) == 0 {
+			app.SetFocus(nil)
+			return
+		}
+		if tabbableActiveIndex >= len(tabbable) {
+			app.SetFocus(tabbable[0])
+			return
+		}
+		app.SetFocus(tabbable[tabbableActiveIndex])
+	}
+	tabbableCycle := func() {
+		if len(tabbable) == 0 {
+			return
+		}
+		tabbableActiveIndex = (tabbableActiveIndex + 1) % len(tabbable)
+		app.SetFocus(tabbable[tabbableActiveIndex])
+		tabbableRedraw()
+	}
+
+	jobMenuVisisble := false
+	reconfigureJobDetailSplit := func(setJobMenuVisible bool) {
+		if jobMenuVisisble == setJobMenuVisible {
+			return
+		}
+		jobMenuVisisble = setJobMenuVisible
+		if setJobMenuVisible {
+			jobDetailSplit.RemoveItem(jobTextDetail)
+			jobDetailSplit.AddItem(jobMenu, 0, 1, true)
+			jobDetailSplit.AddItem(jobTextDetail, 0, 5, false)
+			tabbable = tabbableWithJobMenu
+		} else {
+			jobDetailSplit.RemoveItem(jobMenu)
+			tabbable = tabbableWithoutJobMenu
+		}
+		tabbableRedraw()
+	}
+
+	showModal := func(m *tview.Modal, modalDoneFunc func(idx int, label string)) {
+		preModalFocus := app.GetFocus()
+		m.SetDoneFunc(func(idx int, label string) {
+			if modalDoneFunc != nil {
+				modalDoneFunc(idx, label)
+			}
+			app.SetRoot(toolbarSplit, true)
+			app.SetFocus(preModalFocus)
+			app.Draw()
+		})
+		app.SetRoot(m, true)
+		app.Draw()
+	}
+
+	app.SetRoot(toolbarSplit, true)
+	// initial focus
+	tabbableActiveIndex = len(tabbable)
+	tabbableCycle()
+	reconfigureJobDetailSplit(true)
+
+	m := viewmodel.New()
+	params := &viewmodel.Params{
+		Report:                  nil,
+		SelectedJob:             nil,
+		FSFilter:                func(_ string) bool { return true },
+		DetailViewWidth:         100,
+		DetailViewWrap:          false,
+		ShortKeybindingOverview: "[::b]Q[::-] quit  [::b]<TAB>[::-] switch panes [::b]W[::-] wrap lines  [::b]Shift+M[::-] toggle navbar  [::b]Shift+S[::-] signal job [::b]</>[::-] filter filesystems",
+	}
+	paramsMtx := &sync.Mutex{}
+	var redraw func()
+	viewmodelupdate := func(cb func(*viewmodel.Params)) {
+		paramsMtx.Lock()
+		defer paramsMtx.Unlock()
+		cb(params)
+		m.Update(*params)
+	}
+	redraw = func() {
+		jobs := m.Jobs()
+		if flag.Job != "" {
+			job_found := false
+			for _, job := range jobs {
+				if strings.Compare(flag.Job, job.Name()) == 0 {
+					jobs = []*viewmodel.Job{job}
+					job_found = true
+					break
+				}
+			}
+			if !job_found {
+				jobs = nil
+			}
+		}
+		redrawJobsList := false
+		var selectedJobN *tview.TreeNode
+		if len(jobMenuRoot.GetChildren()) == len(jobs) {
+			for i, jobN := range jobMenuRoot.GetChildren() {
+				if jobN.GetReference().(*viewmodel.Job) != jobs[i] {
+					redrawJobsList = true
+					break
+				}
+				if jobN.GetReference().(*viewmodel.Job) == m.SelectedJob() {
+					selectedJobN = jobN
+				}
+			}
+		} else {
+			redrawJobsList = true
+		}
+		if redrawJobsList {
+			selectedJobN = nil
+			children := make([]*tview.TreeNode, len(jobs))
+			for i := range jobs {
+				jobN := tview.NewTreeNode(jobs[i].JobTreeTitle())
+				jobN.SetReference(jobs[i])
+				jobN.SetSelectable(true)
+				children[i] = jobN
+				jobN.SetSelectedFunc(func() {
+					viewmodelupdate(func(p *viewmodel.Params) {
+						p.SelectedJob = jobN.GetReference().(*viewmodel.Job)
+					})
+				})
+				if jobs[i] == m.SelectedJob() {
+					selectedJobN = jobN
+				}
+			}
+			jobMenuRoot.SetChildren(children)
+		}
+
+		if selectedJobN != nil && jobMenu.GetCurrentNode() != selectedJobN {
+			jobMenu.SetCurrentNode(selectedJobN)
+		} else if selectedJobN == nil {
+			// select something, otherwise selection breaks (likely bug in tview)
+			jobMenu.SetCurrentNode(jobMenuRoot)
+		}
+
+		if selJ := m.SelectedJob(); selJ != nil {
+			jobTextDetail.SetText(selJ.FullDescription())
+		} else {
+			jobTextDetail.SetText("please select a job")
+		}
+
+		bottombardatestring := m.DateString()
+		bottombarDateView.SetText(bottombardatestring)
+		bottombar.ResizeItem(bottombarDateView, len(bottombardatestring), 0)
+
+		bottomBarStatus.SetText(m.BottomBarStatus())
+
+		app.Draw()
+
+	}
+
+	go func() {
+		defer func() {
+			if err := recover(); err != nil {
+				app.Suspend(func() {
+					panic(err)
+				})
+			}
+		}()
+		for {
+			st, err := c.Status()
+			viewmodelupdate(func(p *viewmodel.Params) {
+				p.Report = st.Jobs
+				p.ReportFetchError = err
+			})
+			app.QueueUpdateDraw(redraw)
+
+			time.Sleep(flag.Delay)
+		}
+	}()
+
+	jobMenu.SetChangedFunc(func(jobN *tview.TreeNode) {
+		viewmodelupdate(func(p *viewmodel.Params) {
+			p.SelectedJob, _ = jobN.GetReference().(*viewmodel.Job)
+		})
+		redraw()
+		jobTextDetail.ScrollToBeginning()
+	})
+	jobMenu.SetSelectedFunc(func(jobN *tview.TreeNode) {
+		app.SetFocus(jobTextDetail)
+	})
+
+	app.SetBeforeDrawFunc(func(screen tcell.Screen) bool {
+		viewmodelupdate(func(p *viewmodel.Params) {
+			_, _, p.DetailViewWidth, _ = jobTextDetail.GetInnerRect()
+		})
+		return false
+	})
+
+	app.SetInputCapture(func(e *tcell.EventKey) *tcell.EventKey {
+		if e.Key() == tcell.KeyTab {
+			tabbableCycle()
+			return nil
+		}
+
+		if e.Key() == tcell.KeyRune && app.GetFocus() == fsFilterInput {
+			return e
+		}
+
+		if e.Key() == tcell.KeyRune && e.Rune() == '/' {
+			if app.GetFocus() != fsFilterInput {
+				app.SetFocus(fsFilterInput)
+			}
+			return e
+		}
+
+		if e.Key() == tcell.KeyRune && e.Rune() == 'M' {
+			reconfigureJobDetailSplit(!jobMenuVisisble)
+			return nil
+		}
+
+		if e.Key() == tcell.KeyRune && e.Rune() == 'q' {
+			app.Stop()
+		}
+
+		if e.Key() == tcell.KeyRune && e.Rune() == 'S' {
+			job, ok := jobMenu.GetCurrentNode().GetReference().(*viewmodel.Job)
+			if !ok {
+				return nil
+			}
+			signals := []string{"wakeup", "reset"}
+			clientFuncs := []func(job string) error{c.SignalWakeup, c.SignalReset}
+			sigMod := tview.NewModal()
+			sigMod.SetBackgroundColor(tcell.ColorDefault)
+			sigMod.SetBorder(true)
+			sigMod.GetForm().SetButtonTextColorFocused(tcell.ColorGreen)
+			sigMod.AddButtons(signals)
+			sigMod.SetText(fmt.Sprintf("Send a signal to job %q", job.Name()))
+			showModal(sigMod, func(idx int, _ string) {
+				go func() {
+					if idx == -1 {
+						return
+					}
+					err := clientFuncs[idx](job.Name())
+					if err != nil {
+						app.QueueUpdate(func() {
+							me := tview.NewModal()
+							me.SetText(fmt.Sprintf("signal error: %s", err))
+							me.AddButtons([]string{"Close"})
+							showModal(me, nil)
+						})
+					}
+				}()
+			})
+		}
+
+		return e
+	})
+
+	fsFilterInput.SetChangedFunc(func(searchterm string) {
+		viewmodelupdate(func(p *viewmodel.Params) {
+			p.FSFilter = func(fs string) bool {
+				r, err := regexp.Compile(searchterm)
+				if err != nil {
+					return true
+				}
+				return r.MatchString(fs)
+			}
+		})
+		redraw()
+		jobTextDetail.ScrollToBeginning()
+	})
+	fsFilterInput.SetInputCapture(func(event *tcell.EventKey) *tcell.EventKey {
+		if event.Key() == tcell.KeyEnter {
+			app.SetFocus(jobTextDetail)
+			return nil
+		}
+		return event
+	})
+
+	jobTextDetail.SetInputCapture(func(event *tcell.EventKey) *tcell.EventKey {
+		if event.Key() == tcell.KeyRune && event.Rune() == 'w' {
+			// toggle wrapping
+			viewmodelupdate(func(p *viewmodel.Params) {
+				p.DetailViewWrap = !p.DetailViewWrap
+			})
+			redraw()
+			return nil
+		}
+		return event
+	})
+
+	return app.Run()
+}

client/status/status_legacy.go

@@ -0,0 +1,128 @@
+package status
+
+import (
+	"fmt"
+	"os"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/gdamore/tcell/v2"
+	"github.com/mattn/go-isatty"
+	"github.com/pkg/errors"
+	tview "gitlab.com/tslocum/cview"
+
+	"github.com/zrepl/zrepl/client/status/viewmodel"
+)
+
+func legacy(c Client, flag statusFlags) error {
+
+	// Set this so we don't overwrite the default terminal colors
+	// See https://github.com/rivo/tview/blob/master/styles.go
+	tview.Styles.PrimitiveBackgroundColor = tcell.ColorDefault
+	tview.Styles.ContrastBackgroundColor = tcell.ColorDefault
+	tview.Styles.PrimaryTextColor = tcell.ColorDefault
+	tview.Styles.BorderColor = tcell.ColorDefault
+	app := tview.NewApplication()
+
+	textView := tview.NewTextView()
+	textView.SetWrap(true)
+	textView.SetScrollable(true) // so that it allows us to set scroll position
+	textView.SetScrollBarVisibility(tview.ScrollBarNever)
+
+	app.SetRoot(textView, true)
+
+	width := (1 << 31) - 1
+	wrap := false
+	if isatty.IsTerminal(os.Stdout.Fd()) {
+		wrap = true
+		screen, err := tcell.NewScreen()
+		if err != nil {
+			return errors.Wrap(err, "get terminal dimensions")
+		}
+		if err := screen.Init(); err != nil {
+			return errors.Wrap(err, "init screen")
+		}
+		width, _ = screen.Size()
+		screen.Fini()
+	}
+
+	paramsMtx := &sync.Mutex{}
+	params := viewmodel.Params{
+		Report:                  nil,
+		ReportFetchError:        nil,
+		SelectedJob:             nil,
+		FSFilter:                func(s string) bool { return true },
+		DetailViewWidth:         width,
+		DetailViewWrap:          wrap,
+		ShortKeybindingOverview: "",
+	}
+
+	redraw := func() {
+		textView.Clear()
+
+		paramsMtx.Lock()
+		defer paramsMtx.Unlock()
+
+		if params.ReportFetchError != nil {
+			fmt.Fprintln(textView, params.ReportFetchError.Error())
+		} else if params.Report != nil {
+			m := viewmodel.New()
+			m.Update(params)
+			for _, j := range m.Jobs() {
+				if flag.Job != "" && j.Name() != flag.Job {
+					continue
+				}
+				params.SelectedJob = j
+				m.Update(params)
+				fmt.Fprintln(textView, m.SelectedJob().FullDescription())
+				if flag.Job != "" {
+					break
+				} else {
+					hline := strings.Repeat("-", params.DetailViewWidth)
+					fmt.Fprintln(textView, hline)
+				}
+			}
+		} else {
+			fmt.Fprintln(textView, "waiting for request results")
+		}
+		textView.ScrollToBeginning()
+	}
+
+	app.SetBeforeDrawFunc(func(screen tcell.Screen) bool {
+		// sync resizes to `params`
+		paramsMtx.Lock()
+		_, _, newWidth, _ := textView.GetInnerRect()
+		if newWidth != params.DetailViewWidth {
+			params.DetailViewWidth = newWidth
+			app.QueueUpdateDraw(redraw)
+		}
+		paramsMtx.Unlock()
+
+		textView.ScrollToBeginning() // has the effect of inhibiting user scrolls
+		return false
+	})
+
+	go func() {
+		defer func() {
+			if err := recover(); err != nil {
+				app.Suspend(func() {
+					panic(err)
+				})
+			}
+		}()
+		for {
+			st, err := c.Status()
+			paramsMtx.Lock()
+			params.Report = st.Jobs
+			params.ReportFetchError = err
+			paramsMtx.Unlock()
+
+			app.QueueUpdateDraw(redraw)
+
+			time.Sleep(flag.Delay)
+		}
+	}()
+
+	return app.Run()
+}

client/status/status_raw.go

@@ -0,0 +1,18 @@
+package status
+
+import (
+	"bytes"
+	"io"
+	"os"
+)
+
+func raw(c Client) error {
+	b, err := c.StatusRaw()
+	if err != nil {
+		return err
+	}
+	if _, err := io.Copy(os.Stdout, bytes.NewReader(b)); err != nil {
+		return err
+	}
+	return nil
+}

client/status/viewmodel/bytecountbinary.go

@@ -0,0 +1,26 @@
+package viewmodel
+
+import (
+	"fmt"
+	"math"
+)
+
+func ByteCountBinaryUint(b uint64) string {
+	if b > math.MaxInt64 {
+		panic(b)
+	}
+	return ByteCountBinary(int64(b))
+}
+
+func ByteCountBinary(b int64) string {
+	const unit = 1024
+	if b < unit {
+		return fmt.Sprintf("%d B", b)
+	}
+	div, exp := unit, 0
+	for n := b / unit; n >= unit; n /= unit {
+		div *= unit
+		exp++
+	}
+	return fmt.Sprintf("%.1f %ciB", float64(b)/float64(div), "KMGTPE"[exp])
+}

client/status/viewmodel/bytesprogresshistory.go

@@ -0,0 +1,49 @@
+package viewmodel
+
+import "time"
+
+type byteProgressMeasurement struct {
+	time time.Time
+	val  uint64
+}
+
+type bytesProgressHistory struct {
+	last        *byteProgressMeasurement // pointer as poor man's optional
+	changeCount int
+	lastChange  time.Time
+}
+
+func (p *bytesProgressHistory) Update(currentVal uint64) (bytesPerSecondAvg int64, changeCount int) {
+
+	if p.last == nil {
+		p.last = &byteProgressMeasurement{
+			time: time.Now(),
+			val:  currentVal,
+		}
+		return 0, 0
+	}
+
+	if p.last.val != currentVal {
+		p.changeCount++
+		p.lastChange = time.Now()
+	}
+
+	if time.Since(p.lastChange) > 3*time.Second {
+		p.last = nil
+		return 0, 0
+	}
+
+	var deltaV int64
+	if currentVal >= p.last.val {
+		deltaV = int64(currentVal - p.last.val)
+	} else {
+		deltaV = -int64(p.last.val - currentVal)
+	}
+	deltaT := time.Since(p.last.time)
+	rate := float64(deltaV) / deltaT.Seconds()
+
+	p.last.time = time.Now()
+	p.last.val = currentVal
+
+	return int64(rate), p.changeCount
+}

client/status/viewmodel/render.go

@@ -0,0 +1,660 @@
+package viewmodel
+
+import (
+	"fmt"
+	"math"
+	"sort"
+	"strings"
+	"time"
+
+	"github.com/go-playground/validator/v10"
+	yaml "github.com/zrepl/yaml-config"
+
+	"github.com/zrepl/zrepl/client/status/viewmodel/stringbuilder"
+	"github.com/zrepl/zrepl/daemon"
+	"github.com/zrepl/zrepl/daemon/job"
+	"github.com/zrepl/zrepl/daemon/pruner"
+	"github.com/zrepl/zrepl/daemon/snapper"
+	"github.com/zrepl/zrepl/replication/report"
+)
+
+type M struct {
+	jobs            map[string]*Job
+	jobsList        []*Job
+	selectedJob     *Job
+	dateString      string
+	bottomBarStatus string
+}
+
+type Job struct {
+	// long-lived
+	name         string
+	byteProgress *bytesProgressHistory
+
+	lastStatus      *job.Status
+	fulldescription string
+}
+
+func New() *M {
+	return &M{
+		jobs:        make(map[string]*Job),
+		jobsList:    make([]*Job, 0),
+		selectedJob: nil,
+	}
+}
+
+type FilterFunc func(string) bool
+
+type Params struct {
+	Report                  map[string]*job.Status
+	ReportFetchError        error
+	SelectedJob             *Job
+	FSFilter                FilterFunc `validate:"required"`
+	DetailViewWidth         int        `validate:"gte=1"`
+	DetailViewWrap          bool
+	ShortKeybindingOverview string
+}
+
+var validate = validator.New()
+
+func (m *M) Update(p Params) {
+
+	if err := validate.Struct(p); err != nil {
+		panic(err)
+	}
+
+	if p.ReportFetchError != nil {
+		m.bottomBarStatus = fmt.Sprintf("[red::]status fetch: %s", p.ReportFetchError)
+	} else {
+		m.bottomBarStatus = p.ShortKeybindingOverview
+		for jobname, st := range p.Report {
+			// TODO handle job renames & deletions
+			j, ok := m.jobs[jobname]
+			if !ok {
+				j = &Job{
+					name:         jobname,
+					byteProgress: &bytesProgressHistory{},
+				}
+				m.jobs[jobname] = j
+				m.jobsList = append(m.jobsList, j)
+			}
+			j.lastStatus = st
+		}
+	}
+
+	// filter out internal jobs
+	var jobsList []*Job
+	for _, j := range m.jobsList {
+		if daemon.IsInternalJobName(j.name) {
+			continue
+		}
+		jobsList = append(jobsList, j)
+	}
+	m.jobsList = jobsList
+
+	// determinism!
+	sort.Slice(m.jobsList, func(i, j int) bool {
+		return strings.Compare(m.jobsList[i].name, m.jobsList[j].name) < 0
+	})
+
+	// try to not lose the selected job
+	m.selectedJob = nil
+	for _, j := range m.jobsList {
+		j.updateFullDescription(p)
+		if j == p.SelectedJob {
+			m.selectedJob = j
+		}
+	}
+
+	m.dateString = time.Now().Format(time.RFC3339)
+
+}
+
+func (m *M) BottomBarStatus() string { return m.bottomBarStatus }
+
+func (m *M) Jobs() []*Job { return m.jobsList }
+
+// may be nil
+func (m *M) SelectedJob() *Job { return m.selectedJob }
+
+func (m *M) DateString() string { return m.dateString }
+
+func (j *Job) updateFullDescription(p Params) {
+	width := p.DetailViewWidth
+	if !p.DetailViewWrap {
+		width = 10000000 // FIXME
+	}
+	b := stringbuilder.New(stringbuilder.Config{
+		IndentMultiplier: 3,
+		Width:            width,
+	})
+	drawJob(b, j.name, j.lastStatus, j.byteProgress, p.FSFilter)
+	j.fulldescription = b.String()
+}
+
+func (j *Job) JobTreeTitle() string {
+	return j.name
+}
+
+func (j *Job) FullDescription() string {
+	return j.fulldescription
+}
+
+func (j *Job) Name() string {
+	return j.name
+}
+
+func drawJob(t *stringbuilder.B, name string, v *job.Status, history *bytesProgressHistory, fsfilter FilterFunc) {
+
+	t.Printf("Job: %s\n", name)
+	t.Printf("Type: %s\n\n", v.Type)
+
+	if v.Type == job.TypePush || v.Type == job.TypePull {
+		activeStatus, ok := v.JobSpecific.(*job.ActiveSideStatus)
+		if !ok || activeStatus == nil {
+			t.Printf("ActiveSideStatus is null")
+			t.Newline()
+			return
+		}
+
+		t.Printf("Replication:")
+		t.AddIndentAndNewline(1)
+		renderReplicationReport(t, activeStatus.Replication, history, fsfilter)
+		t.AddIndentAndNewline(-1)
+
+		t.Printf("Pruning Sender:")
+		t.AddIndentAndNewline(1)
+		renderPrunerReport(t, activeStatus.PruningSender, fsfilter)
+		t.AddIndentAndNewline(-1)
+
+		t.Printf("Pruning Receiver:")
+		t.AddIndentAndNewline(1)
+		renderPrunerReport(t, activeStatus.PruningReceiver, fsfilter)
+		t.AddIndentAndNewline(-1)
+
+		if v.Type == job.TypePush {
+			t.Printf("Snapshotting:")
+			t.AddIndentAndNewline(1)
+			renderSnapperReport(t, activeStatus.Snapshotting, fsfilter)
+			t.AddIndentAndNewline(-1)
+		}
+
+	} else if v.Type == job.TypeSnap {
+		snapStatus, ok := v.JobSpecific.(*job.SnapJobStatus)
+		if !ok || snapStatus == nil {
+			t.Printf("SnapJobStatus is null")
+			t.Newline()
+			return
+		}
+		t.Printf("Pruning snapshots:")
+		t.AddIndentAndNewline(1)
+		renderPrunerReport(t, snapStatus.Pruning, fsfilter)
+		t.AddIndentAndNewline(-1)
+		t.Printf("Snapshotting:")
+		t.AddIndentAndNewline(1)
+		renderSnapperReport(t, snapStatus.Snapshotting, fsfilter)
+		t.AddIndentAndNewline(-1)
+	} else if v.Type == job.TypeSource {
+
+		st := v.JobSpecific.(*job.PassiveStatus)
+		t.Printf("Snapshotting:\n")
+		t.AddIndent(1)
+		renderSnapperReport(t, st.Snapper, fsfilter)
+		t.AddIndentAndNewline(-1)
+
+	} else {
+		t.Printf("No status representation for job type '%s', dumping as YAML", v.Type)
+		t.Newline()
+		asYaml, err := yaml.Marshal(v.JobSpecific)
+		if err != nil {
+			t.Printf("Error marshaling status to YAML: %s", err)
+			t.Newline()
+			return
+		}
+		t.Write(string(asYaml))
+		t.Newline()
+	}
+}
+
+func printFilesystemStatus(t *stringbuilder.B, rep *report.FilesystemReport, maxFS int) {
+
+	expected, replicated, containsInvalidSizeEstimates := rep.BytesSum()
+	sizeEstimationImpreciseNotice := ""
+	if containsInvalidSizeEstimates {
+		sizeEstimationImpreciseNotice = " (some steps lack size estimation)"
+	}
+	if rep.CurrentStep < len(rep.Steps) && rep.Steps[rep.CurrentStep].Info.BytesExpected == 0 {
+		sizeEstimationImpreciseNotice = " (step lacks size estimation)"
+	}
+
+	userVisisbleCurrentStep, userVisibleTotalSteps := rep.CurrentStep, len(rep.Steps)
+	// `.CurrentStep` is == len(rep.Steps) if all steps are done.
+	// Until then, it's an index into .Steps that starts at 0.
+	// For the user, we want it to start at 1.
+	if rep.CurrentStep >= len(rep.Steps) {
+		// rep.CurrentStep is what we want to show.
+		// We check for >= and not == for robustness.
+	} else {
+		// We're not done yet, so, make step count start at 1
+		// (The `.State` is included in the output, indicating we're not done yet)
+		userVisisbleCurrentStep = rep.CurrentStep + 1
+	}
+	status := fmt.Sprintf("%s (step %d/%d, %s/%s)%s",
+		strings.ToUpper(string(rep.State)),
+		userVisisbleCurrentStep, userVisibleTotalSteps,
+		ByteCountBinaryUint(replicated), ByteCountBinaryUint(expected),
+		sizeEstimationImpreciseNotice,
+	)
+
+	activeIndicator := " "
+	if rep.BlockedOn == report.FsBlockedOnNothing &&
+		(rep.State == report.FilesystemPlanning || rep.State == report.FilesystemStepping) {
+		activeIndicator = "*"
+	}
+	t.AddIndent(1)
+	t.Printf("%s %s %s ",
+		activeIndicator,
+		stringbuilder.RightPad(rep.Info.Name, maxFS, " "),
+		status)
+
+	next := ""
+	if err := rep.Error(); err != nil {
+		next = err.Err
+	} else if rep.State != report.FilesystemDone {
+		if nextStep := rep.NextStep(); nextStep != nil {
+			if nextStep.IsIncremental() {
+				next = fmt.Sprintf("next: %s => %s", nextStep.Info.From, nextStep.Info.To)
+			} else {
+				next = fmt.Sprintf("next: full send %s", nextStep.Info.To)
+			}
+			attribs := []string{}
+
+			if nextStep.Info.Resumed {
+				attribs = append(attribs, "resumed")
+			}
+
+			if len(attribs) > 0 {
+				next += fmt.Sprintf(" (%s)", strings.Join(attribs, ", "))
+			}
+		} else {
+			next = "" // individual FSes may still be in planning state
+		}
+
+	}
+	t.Printf("%s", next)
+
+	t.AddIndent(-1)
+	t.Newline()
+}
+
+func renderReplicationReport(t *stringbuilder.B, rep *report.Report, history *bytesProgressHistory, fsfilter FilterFunc) {
+	if rep == nil {
+		t.Printf("...\n")
+		return
+	}
+
+	if rep.WaitReconnectError != nil {
+		t.PrintfDrawIndentedAndWrappedIfMultiline("Connectivity: %s", rep.WaitReconnectError)
+		t.Newline()
+	}
+	if !rep.WaitReconnectSince.IsZero() {
+		delta := time.Until(rep.WaitReconnectUntil).Round(time.Second)
+		if rep.WaitReconnectUntil.IsZero() || delta > 0 {
+			var until string
+			if rep.WaitReconnectUntil.IsZero() {
+				until = "waiting indefinitely"
+			} else {
+				until = fmt.Sprintf("hard fail in %s @ %s", delta, rep.WaitReconnectUntil)
+			}
+			t.PrintfDrawIndentedAndWrappedIfMultiline("Connectivity: reconnecting with exponential backoff (since %s) (%s)",
+				rep.WaitReconnectSince, until)
+		} else {
+			t.PrintfDrawIndentedAndWrappedIfMultiline("Connectivity: reconnects reached hard-fail timeout @ %s", rep.WaitReconnectUntil)
+		}
+		t.Newline()
+	}
+
+	// TODO visualize more than the latest attempt by folding all attempts into one
+	if len(rep.Attempts) == 0 {
+		t.Printf("no attempts made yet")
+		return
+	} else {
+		t.Printf("Attempt #%d", len(rep.Attempts))
+		if len(rep.Attempts) > 1 {
+			t.Printf(". Previous attempts failed with the following statuses:")
+			t.AddIndentAndNewline(1)
+			for i, a := range rep.Attempts[:len(rep.Attempts)-1] {
+				t.PrintfDrawIndentedAndWrappedIfMultiline("#%d: %s (failed at %s) (ran %s)\n", i+1, a.State, a.FinishAt, a.FinishAt.Sub(a.StartAt))
+			}
+			t.AddIndentAndNewline(-1)
+		} else {
+			t.Newline()
+		}
+	}
+
+	latest := rep.Attempts[len(rep.Attempts)-1]
+	sort.Slice(latest.Filesystems, func(i, j int) bool {
+		return latest.Filesystems[i].Info.Name < latest.Filesystems[j].Info.Name
+	})
+
+	// apply filter
+	filtered := make([]*report.FilesystemReport, 0, len(latest.Filesystems))
+	for _, fs := range latest.Filesystems {
+		if !fsfilter(fs.Info.Name) {
+			continue
+		}
+		filtered = append(filtered, fs)
+	}
+	latest.Filesystems = filtered
+
+	t.Printf("Status: %s", latest.State)
+	t.Newline()
+	if !latest.FinishAt.IsZero() {
+		t.Printf("Last Run: %s (lasted %s)\n", latest.FinishAt.Round(time.Second), latest.FinishAt.Sub(latest.StartAt).Round(time.Second))
+	} else {
+		t.Printf("Started: %s (lasting %s)\n", latest.StartAt.Round(time.Second), time.Since(latest.StartAt).Round(time.Second))
+	}
+
+	if latest.State == report.AttemptPlanningError {
+		t.Printf("Problem: ")
+		t.PrintfDrawIndentedAndWrappedIfMultiline("%s", latest.PlanError)
+		t.Newline()
+	} else if latest.State == report.AttemptFanOutError {
+		t.Printf("Problem: one or more of the filesystems encountered errors")
+		t.Newline()
+	}
+
+	if latest.State != report.AttemptPlanning && latest.State != report.AttemptPlanningError {
+		// Draw global progress bar
+		// Progress: [---------------]
+		expected, replicated, containsInvalidSizeEstimates := latest.BytesSum()
+		rate, changeCount := history.Update(replicated)
+		eta := time.Duration(0)
+		if rate > 0 {
+			eta = time.Duration((float64(expected)-float64(replicated))/float64(rate)) * time.Second
+		}
+
+		if !latest.State.IsTerminal() {
+			t.Write("Progress: ")
+			t.DrawBar(50, replicated, expected, changeCount)
+			t.Write(fmt.Sprintf(" %s / %s @ %s/s", ByteCountBinaryUint(replicated), ByteCountBinaryUint(expected), ByteCountBinary(rate)))
+			if eta != 0 {
+				t.Write(fmt.Sprintf(" (%s remaining)", humanizeDuration(eta)))
+			}
+			t.Newline()
+		}
+		if containsInvalidSizeEstimates {
+			t.Write("NOTE: not all steps could be size-estimated, total estimate is likely imprecise!")
+			t.Newline()
+		}
+
+		if len(latest.Filesystems) == 0 {
+			t.Write("NOTE: no filesystems were considered for replication!")
+			t.Newline()
+		}
+
+		var maxFSLen int
+		for _, fs := range latest.Filesystems {
+			if len(fs.Info.Name) > maxFSLen {
+				maxFSLen = len(fs.Info.Name)
+			}
+		}
+		for _, fs := range latest.Filesystems {
+			printFilesystemStatus(t, fs, maxFSLen)
+		}
+
+	}
+}
+
+func humanizeDuration(duration time.Duration) string {
+	days := int64(duration.Hours() / 24)
+	hours := int64(math.Mod(duration.Hours(), 24))
+	minutes := int64(math.Mod(duration.Minutes(), 60))
+	seconds := int64(math.Mod(duration.Seconds(), 60))
+
+	var parts []string
+
+	force := false
+	chunks := []int64{days, hours, minutes, seconds}
+	for i, chunk := range chunks {
+		if force || chunk > 0 {
+			padding := 0
+			if force {
+				padding = 2
+			}
+			parts = append(parts, fmt.Sprintf("%*d%c", padding, chunk, "dhms"[i]))
+			force = true
+		}
+	}
+
+	return strings.Join(parts, " ")
+}
+
+func renderPrunerReport(t *stringbuilder.B, r *pruner.Report, fsfilter FilterFunc) {
+	if r == nil {
+		t.Printf("...\n")
+		return
+	}
+
+	state, err := pruner.StateString(r.State)
+	if err != nil {
+		t.Printf("Status: %q (parse error: %q)\n", r.State, err)
+		return
+	}
+
+	t.Printf("Status: %s", state)
+	t.Newline()
+
+	if r.Error != "" {
+		t.Printf("Error: %s\n", r.Error)
+	}
+
+	type commonFS struct {
+		*pruner.FSReport
+		completed bool
+	}
+	all := make([]commonFS, 0, len(r.Pending)+len(r.Completed))
+	for i := range r.Pending {
+		all = append(all, commonFS{&r.Pending[i], false})
+	}
+	for i := range r.Completed {
+		all = append(all, commonFS{&r.Completed[i], true})
+	}
+
+	// filter all
+	filtered := make([]commonFS, 0, len(all))
+	for _, fs := range all {
+		if fsfilter(fs.FSReport.Filesystem) {
+			filtered = append(filtered, fs)
+		}
+	}
+	all = filtered
+
+	switch state {
+	case pruner.Plan:
+		fallthrough
+	case pruner.PlanErr:
+		return
+	}
+
+	if len(all) == 0 {
+		t.Printf("nothing to do\n")
+		return
+	}
+
+	var totalDestroyCount, completedDestroyCount int
+	var maxFSname int
+	for _, fs := range all {
+		totalDestroyCount += len(fs.DestroyList)
+		if fs.completed {
+			completedDestroyCount += len(fs.DestroyList)
+		}
+		if maxFSname < len(fs.Filesystem) {
+			maxFSname = len(fs.Filesystem)
+		}
+	}
+
+	// global progress bar
+	if !state.IsTerminal() {
+		progress := int(math.Round(80 * float64(completedDestroyCount) / float64(totalDestroyCount)))
+		t.Write("Progress: ")
+		t.Write("[")
+		t.Write(stringbuilder.Times("=", progress))
+		t.Write(">")
+		t.Write(stringbuilder.Times("-", 80-progress))
+		t.Write("]")
+		t.Printf(" %d/%d snapshots", completedDestroyCount, totalDestroyCount)
+		t.Newline()
+	}
+
+	sort.SliceStable(all, func(i, j int) bool {
+		return strings.Compare(all[i].Filesystem, all[j].Filesystem) == -1
+	})
+
+	// Draw a table-like representation of 'all'
+	for _, fs := range all {
+		t.Write(stringbuilder.RightPad(fs.Filesystem, maxFSname, " "))
+		t.Write(" ")
+		if !fs.SkipReason.NotSkipped() {
+			t.Printf("skipped: %s\n", fs.SkipReason)
+			continue
+		}
+		if fs.LastError != "" {
+			if strings.ContainsAny(fs.LastError, "\r\n") {
+				t.Printf("ERROR:")
+				t.PrintfDrawIndentedAndWrappedIfMultiline("%s\n", fs.LastError)
+			} else {
+				t.PrintfDrawIndentedAndWrappedIfMultiline("ERROR: %s\n", fs.LastError)
+			}
+			t.Newline()
+			continue
+		}
+
+		pruneRuleActionStr := fmt.Sprintf("(destroy %d of %d snapshots)",
+			len(fs.DestroyList), len(fs.SnapshotList))
+
+		if fs.completed {
+			t.Printf("Completed  %s\n", pruneRuleActionStr)
+			continue
+		}
+
+		t.Write("Pending    ") // whitespace is padding 10
+		if len(fs.DestroyList) == 1 {
+			t.Write(fs.DestroyList[0].Name)
+		} else {
+			t.Write(pruneRuleActionStr)
+		}
+		t.Newline()
+	}
+
+}
+
+func renderSnapperReport(t *stringbuilder.B, r *snapper.Report, fsfilter FilterFunc) {
+	if r == nil {
+		t.Printf("<no snapshotting report available>\n")
+		return
+	}
+	t.Printf("Type: %s\n", r.Type)
+	if r.Periodic != nil {
+		renderSnapperReportPeriodic(t, r.Periodic, fsfilter)
+	} else if r.Cron != nil {
+		renderSnapperReportCron(t, r.Cron, fsfilter)
+	} else {
+		t.Printf("<no details available>")
+	}
+}
+
+func renderSnapperReportPeriodic(t *stringbuilder.B, r *snapper.PeriodicReport, fsfilter FilterFunc) {
+	t.Printf("Status: %s", r.State)
+	t.Newline()
+
+	if r.Error != "" {
+		t.Printf("Error: %s\n", r.Error)
+	}
+	if !r.SleepUntil.IsZero() {
+		t.Printf("Sleep until: %s\n", r.SleepUntil)
+	}
+
+	renderSnapperPlanReportFilesystem(t, r.Progress, fsfilter)
+}
+
+func renderSnapperReportCron(t *stringbuilder.B, r *snapper.CronReport, fsfilter FilterFunc) {
+	t.Printf("State: %s\n", r.State)
+
+	now := time.Now()
+	if r.WakeupTime.After(now) {
+		t.Printf("Sleep until: %s (%s remaining)\n", r.WakeupTime, r.WakeupTime.Sub(now).Round(time.Second))
+	} else {
+		t.Printf("Started: %s (lasting %s)\n", r.WakeupTime, now.Sub(r.WakeupTime).Round(time.Second))
+	}
+
+	renderSnapperPlanReportFilesystem(t, r.Progress, fsfilter)
+}
+
+func renderSnapperPlanReportFilesystem(t *stringbuilder.B, fss []*snapper.ReportFilesystem, fsfilter FilterFunc) {
+	sort.Slice(fss, func(i, j int) bool {
+		return strings.Compare(fss[i].Path, fss[j].Path) == -1
+	})
+
+	dur := func(d time.Duration) string {
+		return d.Round(100 * time.Millisecond).String()
+	}
+
+	type row struct {
+		path, state, duration, remainder, hookReport string
+	}
+	var widths struct {
+		path, state, duration int
+	}
+	rows := make([]*row, 0, len(fss))
+	for _, fs := range fss {
+		if !fsfilter(fs.Path) {
+			continue
+		}
+		r := &row{
+			path:  fs.Path,
+			state: fs.State.String(),
+		}
+		if fs.HooksHadError {
+			r.hookReport = fs.Hooks // FIXME render here, not in daemon
+		}
+		switch fs.State {
+		case snapper.SnapPending:
+			r.duration = "..."
+			r.remainder = ""
+		case snapper.SnapStarted:
+			r.duration = dur(time.Since(fs.StartAt))
+			r.remainder = fmt.Sprintf("snap name: %q", fs.SnapName)
+		case snapper.SnapDone:
+			fallthrough
+		case snapper.SnapError:
+			r.duration = dur(fs.DoneAt.Sub(fs.StartAt))
+			r.remainder = fmt.Sprintf("snap name: %q", fs.SnapName)
+		}
+		rows = append(rows, r)
+		if len(r.path) > widths.path {
+			widths.path = len(r.path)
+		}
+		if len(r.state) > widths.state {
+			widths.state = len(r.state)
+		}
+		if len(r.duration) > widths.duration {
+			widths.duration = len(r.duration)
+		}
+	}
+
+	for _, r := range rows {
+		path := stringbuilder.RightPad(r.path, widths.path, " ")
+		state := stringbuilder.RightPad(r.state, widths.state, " ")
+		duration := stringbuilder.RightPad(r.duration, widths.duration, " ")
+		t.Printf("%s %s %s", path, state, duration)
+		t.PrintfDrawIndentedAndWrappedIfMultiline(" %s", r.remainder)
+		if r.hookReport != "" {
+			t.AddIndent(1)
+			t.Newline()
+			t.Printf("%s", r.hookReport)
+			t.AddIndent(-1)
+		}
+		t.Newline()
+	}
+}

client/status/viewmodel/stringbuilder/stringbuilder.go

@@ -0,0 +1,119 @@
+package stringbuilder
+
+import (
+	"fmt"
+	"strings"
+
+	"github.com/go-playground/validator/v10"
+)
+
+type B struct {
+	// const
+	indentMultiplier int
+
+	// mut
+	sb     *strings.Builder
+	indent int
+	width  int
+	x, y   int
+}
+
+type Config struct {
+	IndentMultiplier int `validate:"gte=1"`
+	Width            int `validate:"gte=1"`
+}
+
+var validate = validator.New()
+
+func New(config Config) *B {
+
+	if err := validate.Struct(config); err != nil {
+		panic(err)
+	}
+
+	return &B{sb: &strings.Builder{}, width: config.Width, indentMultiplier: config.IndentMultiplier}
+}
+
+func (b *B) String() string { return b.sb.String() }
+
+func (w *B) Newline() {
+	w.Write("\n")
+}
+
+func (w *B) PrintfDrawIndentedAndWrappedIfMultiline(format string, args ...interface{}) {
+	whole := fmt.Sprintf(format, args...)
+	if strings.ContainsAny(whole, "\n\r") {
+		w.AddIndent(1)
+		defer w.AddIndent(-1)
+	}
+	w.Write(whole)
+}
+
+func (w *B) Printf(format string, args ...interface{}) {
+	whole := fmt.Sprintf(format, args...)
+	w.Write(whole)
+}
+
+func (t *B) AddIndent(delta int) {
+	t.indent += delta * t.indentMultiplier
+}
+
+func (t *B) AddIndentAndNewline(delta int) {
+	t.indent += delta * t.indentMultiplier
+	t.Write("\n")
+}
+
+func (w *B) Write(s string) {
+	for _, c := range s {
+		if c == '\n' {
+			fmt.Fprint(w.sb, "\n")
+			w.x = 0
+			fmt.Fprint(w.sb, Times(" ", w.indent-w.x))
+			w.x = w.indent
+			w.y++
+			continue
+		}
+		if w.x >= w.width {
+			fmt.Fprint(w.sb, "\n")
+			w.x = 0
+			fmt.Fprint(w.sb, Times(" ", w.indent-w.x))
+			w.x = w.indent
+		}
+		fmt.Fprintf(w.sb, "%c", c)
+		w.x++
+	}
+}
+
+func Times(str string, n int) (out string) {
+	for i := 0; i < n; i++ {
+		out += str
+	}
+	return
+}
+
+func RightPad(str string, length int, pad string) string {
+	if len(str) > length {
+		return str[:length]
+	}
+	return str + strings.Repeat(pad, length-len(str))
+}
+
+// changeCount = 0 indicates stall / no progress
+func (w *B) DrawBar(length int, bytes, totalBytes uint64, changeCount int) {
+	const arrowPositions = `>\|/`
+	var completedLength int
+	if totalBytes > 0 {
+		completedLength = int(uint64(length) * bytes / totalBytes)
+		if completedLength > length {
+			completedLength = length
+		}
+	} else if totalBytes == bytes {
+		completedLength = length
+	}
+
+	w.Write("[")
+	w.Write(Times("=", completedLength))
+	w.Write(string(arrowPositions[changeCount%len(arrowPositions)]))
+	w.Write(Times("-", length-completedLength))
+	w.Write("]")
+}

client/stdinserver.go

@@ -17,7 +17,7 @@ import (
 var StdinserverCmd = &cli.Subcommand{
 	Use:   "stdinserver CLIENT_IDENTITY",
 	Short: "stdinserver transport mode (started from authorized_keys file as forced command)",
-	Run: func(subcommand *cli.Subcommand, args []string) error {
+	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
 		return runStdinserver(subcommand.Config(), args)
 	},
 }

client/testcmd.go

@@ -39,7 +39,7 @@ var testFilter = &cli.Subcommand{
 	Run: runTestFilterCmd,
 }
 
-func runTestFilterCmd(subcommand *cli.Subcommand, args []string) error {
+func runTestFilterCmd(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
 
 	if testFilterArgs.job == "" {
 		return fmt.Errorf("must specify --job flag")
@@ -49,7 +49,6 @@ func runTestFilterCmd(subcommand *cli.Subcommand, args []string) error {
 	}
 
 	conf := subcommand.Config()
-	ctx := context.Background()
 
 	var confFilter config.FilesystemsFilter
 	job, err := conf.Job(testFilterArgs.job)
@@ -137,13 +136,14 @@ var testPlaceholder = &cli.Subcommand{
 	Run: runTestPlaceholder,
 }
 
-func runTestPlaceholder(subcommand *cli.Subcommand, args []string) error {
+func runTestPlaceholder(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
 
 	var checkDPs []*zfs.DatasetPath
-	ctx := context.Background()
+	var datasetWasExplicitArgument bool
 
 	// all actions first
 	if testPlaceholderArgs.all {
+		datasetWasExplicitArgument = false
 		out, err := zfs.ZFSList(ctx, []string{"name"})
 		if err != nil {
 			return errors.Wrap(err, "could not list ZFS filesystems")
@@ -156,6 +156,7 @@ func runTestPlaceholder(subcommand *cli.Subcommand, args []string) error {
 			checkDPs = append(checkDPs, dp)
 		}
 	} else {
+		datasetWasExplicitArgument = true
 		dp, err := zfs.NewDatasetPath(testPlaceholderArgs.ds)
 		if err != nil {
 			return err
@@ -173,7 +174,12 @@ func runTestPlaceholder(subcommand *cli.Subcommand, args []string) error {
 			return errors.Wrap(err, "cannot get placeholder state")
 		}
 		if !ph.FSExists {
-			panic("placeholder state inconsistent: filesystem " + ph.FS + " must exist in this context")
+			if datasetWasExplicitArgument {
+				return errors.Errorf("filesystem %q does not exist", ph.FS)
+			} else {
+				// got deleted between ZFSList and ZFSGetFilesystemPlaceholderState
+				continue
+			}
 		}
 		is := "yes"
 		if !ph.IsPlaceholder {
@@ -197,11 +203,11 @@ var testDecodeResumeToken = &cli.Subcommand{
 	Run: runTestDecodeResumeTokenCmd,
 }
 
-func runTestDecodeResumeTokenCmd(subcommand *cli.Subcommand, args []string) error {
+func runTestDecodeResumeTokenCmd(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
 	if testDecodeResumeTokenArgs.token == "" {
 		return fmt.Errorf("token argument must be specified")
 	}
-	token, err := zfs.ParseResumeToken(context.Background(), testDecodeResumeTokenArgs.token)
+	token, err := zfs.ParseResumeToken(ctx, testDecodeResumeTokenArgs.token)
 	if err != nil {
 		return err
 	}

client/version.go

@@ -1,6 +1,7 @@
 package client
 
 import (
+	"context"
 	"fmt"
 	"os"
 
@@ -25,7 +26,7 @@ var VersionCmd = &cli.Subcommand{
 	SetupFlags: func(f *pflag.FlagSet) {
 		f.StringVar(&versionArgs.Show, "show", "", "version info to show (client|daemon)")
 	},
-	Run: func(subcommand *cli.Subcommand, args []string) error {
+	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
 		versionArgs.Config = subcommand.Config()
 		versionArgs.ConfigErr = subcommand.ConfigParsingError()
 		return runVersionCmd()

client/zfsabstractions.go

@@ -57,7 +57,7 @@ func (f *zabsFilterFlags) registerZabsFilterFlags(s *pflag.FlagSet, verb string)
 	for v := range endpoint.AbstractionTypesAll {
 		variants = append(variants, string(v))
 	}
-	variants = sort.StringSlice(variants)
+	sort.Strings(variants)
 	variantsJoined := strings.Join(variants, "|")
 	s.Var(&f.Types, "type", fmt.Sprintf("only %s holds of the specified type [default: all] [comma-separated list of %s]", verb, variantsJoined))
 

client/zfsabstractions_create_step_hold.go

@@ -28,7 +28,7 @@ var zabsCmdCreateStepHold = &cli.Subcommand{
 	},
 }
 
-func doZabsCreateStep(sc *cli.Subcommand, args []string) error {
+func doZabsCreateStep(ctx context.Context, sc *cli.Subcommand, args []string) error {
 	if len(args) > 0 {
 		return errors.New("subcommand takes no arguments")
 	}
@@ -44,8 +44,6 @@ func doZabsCreateStep(sc *cli.Subcommand, args []string) error {
 		return errors.Errorf("jobid must be set")
 	}
 
-	ctx := context.Background()
-
 	v, err := zfs.ZFSGetFilesystemVersion(ctx, f.target)
 	if err != nil {
 		return errors.Wrapf(err, "get info about target %q", f.target)

client/zfsabstractions_list.go

@@ -32,9 +32,8 @@ var zabsCmdList = &cli.Subcommand{
 	},
 }
 
-func doZabsList(sc *cli.Subcommand, args []string) error {
+func doZabsList(ctx context.Context, sc *cli.Subcommand, args []string) error {
 	var err error
-	ctx := context.Background()
 
 	if len(args) > 0 {
 		return errors.New("this subcommand takes no positional arguments")
@@ -45,17 +44,18 @@ func doZabsList(sc *cli.Subcommand, args []string) error {
 		return errors.Wrap(err, "invalid filter specification on command line")
 	}
 
-	abstractions, errors, err := endpoint.ListAbstractionsStreamed(ctx, q)
+	abstractions, errors, drainDone, err := endpoint.ListAbstractionsStreamed(ctx, q)
 	if err != nil {
 		return err // context clear by invocation of command
 	}
+	defer drainDone()
 
 	var line chainlock.L
 	var wg sync.WaitGroup
 	defer wg.Wait()
-	wg.Add(1)
 
 	// print results
+	wg.Add(1)
 	go func() {
 		defer wg.Done()
 		enc := json.NewEncoder(os.Stdout)
@@ -64,7 +64,7 @@ func doZabsList(sc *cli.Subcommand, args []string) error {
 				defer line.Lock().Unlock()
 				if zabsListFlags.Json {
 					enc.SetIndent("", "  ")
-					if err := enc.Encode(abstractions); err != nil {
+					if err := enc.Encode(a); err != nil {
 						panic(err)
 					}
 					fmt.Println()

client/zfsabstractions_release.go

@@ -43,9 +43,8 @@ var zabsCmdReleaseStale = &cli.Subcommand{
 	SetupFlags:      registerZabsReleaseFlags,
 }
 
-func doZabsReleaseAll(sc *cli.Subcommand, args []string) error {
+func doZabsReleaseAll(ctx context.Context, sc *cli.Subcommand, args []string) error {
 	var err error
-	ctx := context.Background()
 
 	if len(args) > 0 {
 		return errors.New("this subcommand takes no positional arguments")
@@ -68,10 +67,9 @@ func doZabsReleaseAll(sc *cli.Subcommand, args []string) error {
 	return doZabsRelease_Common(ctx, abstractions)
 }
 
-func doZabsReleaseStale(sc *cli.Subcommand, args []string) error {
+func doZabsReleaseStale(ctx context.Context, sc *cli.Subcommand, args []string) error {
 
 	var err error
-	ctx := context.Background()
 
 	if len(args) > 0 {
 		return errors.New("this subcommand takes no positional arguments")

config/config.go

@@ -6,12 +6,21 @@ import (
 	"log/syslog"
 	"os"
 	"reflect"
-	"regexp"
-	"strconv"
 	"time"
 
 	"github.com/pkg/errors"
+	"github.com/robfig/cron/v3"
 	"github.com/zrepl/yaml-config"
+
+	"github.com/zrepl/zrepl/util/datasizeunit"
+	zfsprop "github.com/zrepl/zrepl/zfs/property"
+)
+
+type ParseFlags uint
+
+const (
+	ParseFlagsNone        ParseFlags = 0
+	ParseFlagsNoCertCheck ParseFlags = 1 << iota
 )
 
 type Config struct {
@@ -52,51 +61,87 @@ func (j JobEnum) Name() string {
 }
 
 type ActiveJob struct {
-	Type    string                `yaml:"type"`
+	Type               string                `yaml:"type"`
-	Name    string                `yaml:"name"`
+	Name               string                `yaml:"name"`
-	Connect ConnectEnum           `yaml:"connect"`
+	Connect            ConnectEnum           `yaml:"connect"`
-	Pruning PruningSenderReceiver `yaml:"pruning"`
+	Pruning            PruningSenderReceiver `yaml:"pruning"`
-	Debug   JobDebugSettings      `yaml:"debug,optional"`
+	Replication        *Replication          `yaml:"replication,optional,fromdefaults"`
+	ConflictResolution *ConflictResolution   `yaml:"conflict_resolution,optional,fromdefaults"`
+}
+
+type ConflictResolution struct {
+	InitialReplication string `yaml:"initial_replication,optional,default=most_recent"`
 }
 
 type PassiveJob struct {
-	Type  string           `yaml:"type"`
+	Type  string    `yaml:"type"`
-	Name  string           `yaml:"name"`
+	Name  string    `yaml:"name"`
-	Serve ServeEnum        `yaml:"serve"`
+	Serve ServeEnum `yaml:"serve"`
-	Debug JobDebugSettings `yaml:"debug,optional"`
 }
 
 type SnapJob struct {
 	Type         string            `yaml:"type"`
 	Name         string            `yaml:"name"`
 	Pruning      PruningLocal      `yaml:"pruning"`
-	Debug        JobDebugSettings  `yaml:"debug,optional"`
 	Snapshotting SnapshottingEnum  `yaml:"snapshotting"`
 	Filesystems  FilesystemsFilter `yaml:"filesystems"`
 }
 
 type SendOptions struct {
-	Encrypted bool `yaml:"encrypted"`
+	Encrypted        bool `yaml:"encrypted,optional,default=false"`
-}
+	Raw              bool `yaml:"raw,optional,default=false"`
+	SendProperties   bool `yaml:"send_properties,optional,default=false"`
+	BackupProperties bool `yaml:"backup_properties,optional,default=false"`
+	LargeBlocks      bool `yaml:"large_blocks,optional,default=false"`
+	Compressed       bool `yaml:"compressed,optional,default=false"`
+	EmbeddedData     bool `yaml:"embedded_data,optional,default=false"`
+	Saved            bool `yaml:"saved,optional,default=false"`
 
-var _ yaml.Defaulter = (*SendOptions)(nil)
+	BandwidthLimit *BandwidthLimit `yaml:"bandwidth_limit,optional,fromdefaults"`
-
-func (l *SendOptions) SetDefault() {
-	*l = SendOptions{Encrypted: false}
 }
 
 type RecvOptions struct {
 	// Note: we cannot enforce encrypted recv as the ZFS cli doesn't provide a mechanism for it
 	// Encrypted bool `yaml:"may_encrypted"`
-
 	// Future:
 	// Reencrypt bool `yaml:"reencrypt"`
+
+	Properties *PropertyRecvOptions `yaml:"properties,fromdefaults"`
+
+	BandwidthLimit *BandwidthLimit `yaml:"bandwidth_limit,optional,fromdefaults"`
+
+	Placeholder *PlaceholderRecvOptions `yaml:"placeholder,fromdefaults"`
 }
 
-var _ yaml.Defaulter = (*RecvOptions)(nil)
+var _ yaml.Unmarshaler = &datasizeunit.Bits{}
 
-func (l *RecvOptions) SetDefault() {
+type BandwidthLimit struct {
-	*l = RecvOptions{}
+	Max            datasizeunit.Bits `yaml:"max,default=-1 B"`
+	BucketCapacity datasizeunit.Bits `yaml:"bucket_capacity,default=128 KiB"`
+}
+
+type Replication struct {
+	Protection  *ReplicationOptionsProtection  `yaml:"protection,optional,fromdefaults"`
+	Concurrency *ReplicationOptionsConcurrency `yaml:"concurrency,optional,fromdefaults"`
+}
+
+type ReplicationOptionsProtection struct {
+	Initial     string `yaml:"initial,optional,default=guarantee_resumability"`
+	Incremental string `yaml:"incremental,optional,default=guarantee_resumability"`
+}
+
+type ReplicationOptionsConcurrency struct {
+	Steps         int `yaml:"steps,optional,default=1"`
+	SizeEstimates int `yaml:"size_estimates,optional,default=4"`
+}
+
+type PropertyRecvOptions struct {
+	Inherit  []zfsprop.Property          `yaml:"inherit,optional"`
+	Override map[zfsprop.Property]string `yaml:"override,optional"`
+}
+
+type PlaceholderRecvOptions struct {
+	Encryption string `yaml:"encryption,default=unspecified"`
 }
 
 type PushJob struct {
@@ -106,6 +151,9 @@ type PushJob struct {
 	Send         *SendOptions      `yaml:"send,fromdefaults,optional"`
 }
 
+func (j *PushJob) GetFilesystems() FilesystemsFilter { return j.Filesystems }
+func (j *PushJob) GetSendOptions() *SendOptions      { return j.Send }
+
 type PullJob struct {
 	ActiveJob `yaml:",inline"`
 	RootFS    string                   `yaml:"root_fs"`
@@ -113,6 +161,10 @@ type PullJob struct {
 	Recv      *RecvOptions             `yaml:"recv,fromdefaults,optional"`
 }
 
+func (j *PullJob) GetRootFS() string             { return j.RootFS }
+func (j *PullJob) GetAppendClientIdentity() bool { return false }
+func (j *PullJob) GetRecvOptions() *RecvOptions  { return j.Recv }
+
 type PositiveDurationOrManual struct {
 	Interval time.Duration
 	Manual   bool
@@ -133,13 +185,10 @@ func (i *PositiveDurationOrManual) UnmarshalYAML(u func(interface{}, bool) error
 		return fmt.Errorf("value must not be empty")
 	default:
 		i.Manual = false
-		i.Interval, err = time.ParseDuration(s)
+		i.Interval, err = parsePositiveDuration(s)
 		if err != nil {
 			return err
 		}
-		if i.Interval <= 0 {
-			return fmt.Errorf("value must be a positive duration, got %q", s)
-		}
 	}
 	return nil
 }
@@ -150,6 +199,10 @@ type SinkJob struct {
 	Recv       *RecvOptions `yaml:"recv,optional,fromdefaults"`
 }
 
+func (j *SinkJob) GetRootFS() string             { return j.RootFS }
+func (j *SinkJob) GetAppendClientIdentity() bool { return true }
+func (j *SinkJob) GetRecvOptions() *RecvOptions  { return j.Recv }
+
 type SourceJob struct {
 	PassiveJob   `yaml:",inline"`
 	Snapshotting SnapshottingEnum  `yaml:"snapshotting"`
@@ -157,6 +210,9 @@ type SourceJob struct {
 	Send         *SendOptions      `yaml:"send,optional,fromdefaults"`
 }
 
+func (j *SourceJob) GetFilesystems() FilesystemsFilter { return j.Filesystems }
+func (j *SourceJob) GetSendOptions() *SendOptions      { return j.Send }
+
 type FilesystemsFilter map[string]bool
 
 type SnapshottingEnum struct {
@@ -164,10 +220,44 @@ type SnapshottingEnum struct {
 }
 
 type SnapshottingPeriodic struct {
-	Type     string        `yaml:"type"`
+	Type            string            `yaml:"type"`
-	Prefix   string        `yaml:"prefix"`
+	Prefix          string            `yaml:"prefix"`
-	Interval time.Duration `yaml:"interval,positive"`
+	Interval        *PositiveDuration `yaml:"interval"`
-	Hooks    HookList      `yaml:"hooks,optional"`
+	Hooks           HookList          `yaml:"hooks,optional"`
+	TimestampFormat string            `yaml:"timestamp_format,optional,default=dense"`
+}
+
+type CronSpec struct {
+	Schedule cron.Schedule
+}
+
+var _ yaml.Unmarshaler = &CronSpec{}
+
+func (s *CronSpec) UnmarshalYAML(unmarshal func(v interface{}, not_strict bool) error) error {
+	var specString string
+	if err := unmarshal(&specString, false); err != nil {
+		return err
+	}
+
+	// Use standard cron format.
+	// Disable the various "descriptors" (@daily, etc)
+	// They are just aliases to "top of hour", "midnight", etc.
+	parser := cron.NewParser(cron.Minute | cron.Hour | cron.Dom | cron.Month | cron.Dow | cron.SecondOptional)
+
+	sched, err := parser.Parse(specString)
+	if err != nil {
+		return errors.Wrap(err, "cron syntax invalid")
+	}
+	s.Schedule = sched
+	return nil
+}
+
+type SnapshottingCron struct {
+	Type            string   `yaml:"type"`
+	Prefix          string   `yaml:"prefix"`
+	Cron            CronSpec `yaml:"cron"`
+	Hooks           HookList `yaml:"hooks,optional"`
+	TimestampFormat string   `yaml:"timestamp_format,optional,default=dense"`
 }
 
 type SnapshottingManual struct {
@@ -312,6 +402,7 @@ type PruneKeepNotReplicated struct {
 type PruneKeepLastN struct {
 	Type  string `yaml:"type"`
 	Count int    `yaml:"count"`
+	Regex string `yaml:"regex,optional"`
 }
 
 type PruneKeepRegex struct { // FIXME rename to KeepRegex
@@ -386,14 +477,6 @@ type GlobalStdinServer struct {
 	SockDir string `yaml:"sockdir,default=/var/run/zrepl/stdinserver"`
 }
 
-type JobDebugSettings struct {
-	Conn *struct {
-		ReadDump  string `yaml:"read_dump"`
-		WriteDump string `yaml:"write_dump"`
-	} `yaml:"conn,optional"`
-	RPCLog bool `yaml:"rpc_log,optional,default=false"`
-}
-
 type HookList []HookEnum
 
 type HookEnum struct {
@@ -492,6 +575,7 @@ func (t *SnapshottingEnum) UnmarshalYAML(u func(interface{}, bool) error) (err e
 	t.Ret, err = enumUnmarshal(u, map[string]interface{}{
 		"periodic": &SnapshottingPeriodic{},
 		"manual":   &SnapshottingManual{},
+		"cron":     &SnapshottingCron{},
 	})
 	return
 }
@@ -617,41 +701,3 @@ func ParseConfigBytes(bytes []byte) (*Config, error) {
 	}
 	return c, nil
 }
-
-var durationStringRegex *regexp.Regexp = regexp.MustCompile(`^\s*(\d+)\s*(s|m|h|d|w)\s*$`)
-
-func parsePositiveDuration(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
-}

config/config_dep_test.go

@@ -0,0 +1,90 @@
+package config
+
+import (
+	"fmt"
+	"testing"
+
+	"github.com/stretchr/testify/require"
+	"github.com/zrepl/yaml-config"
+)
+
+type A struct {
+	B  *B     `yaml:"b,optional,fromdefaults"`
+	A1 string `yaml:"a1,optional"`
+}
+
+type B struct {
+	C *C     `yaml:"c,optional,fromdefaults"`
+	D string `yaml:"d,default=ddd"`
+	E string `yaml:"e,optional"`
+}
+
+type C struct {
+	Q string `yaml:"q,optional"`
+	R string `yaml:"r,default=r"`
+}
+
+func TestDepFromDefaults(t *testing.T) {
+
+	type testcase struct {
+		name   string
+		yaml   string
+		expect *A
+	}
+
+	tcs := []testcase{
+		{
+			name: "empty",
+			yaml: `{}`,
+			expect: &A{
+				B: &B{
+					C: &C{
+						R: "r",
+					},
+					D: "ddd",
+				},
+			},
+		},
+		{
+			name: "a1 set",
+			yaml: `{"a1":"blah"}`,
+			expect: &A{
+				A1: "blah",
+				B: &B{
+					C: &C{
+						R: "r",
+					},
+					D: "ddd",
+				},
+			},
+		},
+		{
+			name: "D set",
+			yaml: `
+b:
+  d: 4d
+`,
+			expect: &A{
+				B: &B{
+					D: "4d",
+					C: &C{
+						R: "r",
+					},
+				},
+			},
+		},
+	}
+
+	for tci := range tcs {
+		t.Run(fmt.Sprintf("%d-%s", tci, tcs[tci].name), func(t *testing.T) {
+			tc := tcs[tci]
+
+			var a A
+			err := yaml.UnmarshalStrict([]byte(tc.yaml), &a)
+			require.NoError(t, err)
+
+			require.Equal(t, tc.expect, &a)
+		})
+	}
+
+}

config/config_duration.go

@@ -0,0 +1,106 @@
+package config
+
+import (
+	"errors"
+	"fmt"
+	"regexp"
+	"strconv"
+	"time"
+
+	"github.com/kr/pretty"
+	"github.com/zrepl/yaml-config"
+)
+
+type Duration struct{ d time.Duration }
+
+func (d Duration) Duration() time.Duration { return d.d }
+
+var _ yaml.Unmarshaler = &Duration{}
+
+func (d *Duration) UnmarshalYAML(unmarshal func(v interface{}, not_strict bool) error) error {
+	var s string
+	err := unmarshal(&s, false)
+	if err != nil {
+		return err
+	}
+	d.d, err = parseDuration(s)
+	if err != nil {
+		d.d = 0
+		return &yaml.TypeError{Errors: []string{fmt.Sprintf("cannot parse value %q: %s", s, err)}}
+	}
+	return nil
+}
+
+type PositiveDuration struct{ d Duration }
+
+var _ yaml.Unmarshaler = &PositiveDuration{}
+
+func (d PositiveDuration) Duration() time.Duration { return d.d.Duration() }
+
+func (d *PositiveDuration) UnmarshalYAML(unmarshal func(v interface{}, not_strict bool) error) error {
+	err := d.d.UnmarshalYAML(unmarshal)
+	if err != nil {
+		return err
+	}
+	if d.d.Duration() <= 0 {
+		return fmt.Errorf("duration must be positive, got %s", d.d.Duration())
+	}
+	return nil
+}
+
+func parsePositiveDuration(e string) (time.Duration, error) {
+	d, err := parseDuration(e)
+	if err != nil {
+		return d, err
+	}
+	if d <= 0 {
+		return 0, errors.New("duration must be positive integer")
+	}
+	return d, err
+}
+
+var durationStringRegex *regexp.Regexp = regexp.MustCompile(`^\s*([\+-]?\d+)\s*(|s|m|h|d|w)\s*$`)
+
+func parseDuration(e string) (d time.Duration, err error) {
+	comps := durationStringRegex.FindStringSubmatch(e)
+	if comps == nil {
+		err = fmt.Errorf("must match %s", durationStringRegex)
+		return
+	}
+	if len(comps) != 3 {
+		panic(pretty.Sprint(comps))
+	}
+
+	durationFactor, err := strconv.ParseInt(comps[1], 10, 64)
+	if err != nil {
+		return 0, err
+	}
+
+	var durationUnit time.Duration
+	switch comps[2] {
+	case "":
+		if durationFactor != 0 {
+			err = fmt.Errorf("missing time unit")
+			return
+		} else {
+			// It's the case where user specified '0'.
+			// We want to allow this, just like time.ParseDuration.
+		}
+	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
+}

config/config_global_test.go

@@ -70,9 +70,9 @@ func TestPrometheusMonitoring(t *testing.T) {
 global:
   monitoring:
     - type: prometheus
-      listen: ':9091'
+      listen: ':9811'
 `)
-	assert.Equal(t, ":9091", conf.Global.Monitoring[0].Ret.(*PrometheusMonitoring).Listen)
+	assert.Equal(t, ":9811", conf.Global.Monitoring[0].Ret.(*PrometheusMonitoring).Listen)
 }
 
 func TestSyslogLoggingOutletFacility(t *testing.T) {

config/config_recv_test.go

@@ -0,0 +1,134 @@
+package config
+
+import (
+	"fmt"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	zfsprop "github.com/zrepl/zrepl/zfs/property"
+)
+
+func TestRecvOptions(t *testing.T) {
+	tmpl := `
+jobs:
+- name: foo
+  type: pull
+  connect:
+    type: local
+    listener_name: foo
+    client_identity: bar
+  root_fs: "zreplplatformtest"
+  %s
+  interval: manual
+  pruning:
+    keep_sender:
+    - type: last_n
+      count: 10
+    keep_receiver:
+    - type: last_n
+      count: 10
+
+`
+
+	recv_properties_empty := `
+  recv:
+    properties:
+`
+
+	recv_inherit_empty := `
+  recv:
+    properties:
+      inherit:
+`
+
+	recv_inherit := `
+  recv:
+    properties:
+      inherit:
+        - testprop
+`
+
+	recv_override_empty := `
+  recv:
+    properties:
+      override:
+`
+
+	recv_override := `
+  recv:
+    properties:
+      override:
+        testprop2: "test123"
+`
+
+	recv_override_and_inherit := `
+  recv:
+    properties:
+      inherit:
+        - testprop
+      override:
+        testprop2: "test123"
+`
+
+	recv_empty := `
+  recv: {}
+`
+
+	recv_not_specified := `
+`
+
+	fill := func(s string) string { return fmt.Sprintf(tmpl, s) }
+
+	t.Run("recv_inherit_empty", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_inherit_empty))
+		assert.NotNil(t, c)
+	})
+
+	t.Run("recv_inherit", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_inherit))
+		inherit := c.Jobs[0].Ret.(*PullJob).Recv.Properties.Inherit
+		assert.NotEmpty(t, inherit)
+		assert.Contains(t, inherit, zfsprop.Property("testprop"))
+	})
+
+	t.Run("recv_override_empty", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_override_empty))
+		assert.NotNil(t, c)
+	})
+
+	t.Run("recv_override", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_override))
+		override := c.Jobs[0].Ret.(*PullJob).Recv.Properties.Override
+		require.Len(t, override, 1)
+		require.Equal(t, "test123", override["testprop2"])
+
+	})
+
+	t.Run("recv_override_and_inherit", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_override_and_inherit))
+		inherit := c.Jobs[0].Ret.(*PullJob).Recv.Properties.Inherit
+		override := c.Jobs[0].Ret.(*PullJob).Recv.Properties.Override
+		assert.NotEmpty(t, inherit)
+		assert.Contains(t, inherit, zfsprop.Property("testprop"))
+		assert.NotEmpty(t, override)
+		assert.Equal(t, "test123", override["testprop2"])
+	})
+
+	t.Run("recv_properties_empty", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_properties_empty))
+		assert.NotNil(t, c)
+	})
+
+	t.Run("recv_empty", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_empty))
+		assert.NotNil(t, c)
+	})
+
+	t.Run("send_not_specified", func(t *testing.T) {
+		c := testValidConfig(t, fill(recv_not_specified))
+		assert.NotNil(t, c)
+	})
+
+}

config/config_send_test.go

@@ -38,7 +38,41 @@ jobs:
     encrypted: true
 `
 
-	encrypted_unspecified := `
+	raw_true := `
+  send:
+    raw: true
+
+`
+
+	raw_false := `
+  send:
+    raw: false
+
+`
+
+	raw_and_encrypted := `
+  send:
+    encrypted: true
+    raw: true
+`
+
+	properties_and_encrypted := `
+  send:
+    encrypted: true
+    send_properties: true
+`
+
+	properties_true := `
+  send:
+    send_properties: true
+`
+
+	properties_false := `
+  send:
+    send_properties: false
+`
+
+	send_empty := `
   send: {}
 `
 
@@ -46,28 +80,66 @@ jobs:
 `
 
 	fill := func(s string) string { return fmt.Sprintf(tmpl, s) }
-	var c *Config
 
 	t.Run("encrypted_false", func(t *testing.T) {
-		c = testValidConfig(t, fill(encrypted_false))
+		c := testValidConfig(t, fill(encrypted_false))
 		encrypted := c.Jobs[0].Ret.(*PushJob).Send.Encrypted
 		assert.Equal(t, false, encrypted)
 	})
 	t.Run("encrypted_true", func(t *testing.T) {
-		c = testValidConfig(t, fill(encrypted_true))
+		c := testValidConfig(t, fill(encrypted_true))
 		encrypted := c.Jobs[0].Ret.(*PushJob).Send.Encrypted
 		assert.Equal(t, true, encrypted)
 	})
 
-	t.Run("encrypted_unspecified", func(t *testing.T) {
+	t.Run("send_empty", func(t *testing.T) {
-		c, err := testConfig(t, fill(encrypted_unspecified))
+		c := testValidConfig(t, fill(send_empty))
-		assert.Error(t, err)
+		encrypted := c.Jobs[0].Ret.(*PushJob).Send.Encrypted
-		assert.Nil(t, c)
+		assert.Equal(t, false, encrypted)
+	})
+
+	t.Run("properties_and_encrypted", func(t *testing.T) {
+		c := testValidConfig(t, fill(properties_and_encrypted))
+		encrypted := c.Jobs[0].Ret.(*PushJob).Send.Encrypted
+		properties := c.Jobs[0].Ret.(*PushJob).Send.SendProperties
+		assert.Equal(t, true, encrypted)
+		assert.Equal(t, true, properties)
+	})
+
+	t.Run("properties_false", func(t *testing.T) {
+		c := testValidConfig(t, fill(properties_false))
+		properties := c.Jobs[0].Ret.(*PushJob).Send.SendProperties
+		assert.Equal(t, false, properties)
+	})
+
+	t.Run("properties_true", func(t *testing.T) {
+		c := testValidConfig(t, fill(properties_true))
+		properties := c.Jobs[0].Ret.(*PushJob).Send.SendProperties
+		assert.Equal(t, true, properties)
+	})
+
+	t.Run("raw_true", func(t *testing.T) {
+		c := testValidConfig(t, fill(raw_true))
+		raw := c.Jobs[0].Ret.(*PushJob).Send.Raw
+		assert.Equal(t, true, raw)
+	})
+
+	t.Run("raw_false", func(t *testing.T) {
+		c := testValidConfig(t, fill(raw_false))
+		raw := c.Jobs[0].Ret.(*PushJob).Send.Raw
+		assert.Equal(t, false, raw)
+	})
+
+	t.Run("raw_and_encrypted", func(t *testing.T) {
+		c := testValidConfig(t, fill(raw_and_encrypted))
+		raw := c.Jobs[0].Ret.(*PushJob).Send.Raw
+		encrypted := c.Jobs[0].Ret.(*PushJob).Send.Encrypted
+		assert.Equal(t, true, raw)
+		assert.Equal(t, true, encrypted)
 	})
 
 	t.Run("send_not_specified", func(t *testing.T) {
-		c, err := testConfig(t, fill(send_not_specified))
+		c := testValidConfig(t, fill(send_not_specified))
-		assert.NoError(t, err)
 		assert.NotNil(t, c)
 	})
 

config/config_snapshotting_test.go

@@ -35,8 +35,23 @@ jobs:
   snapshotting:
     type: periodic
     prefix: zrepl_
+    timestamp_format: dense
     interval: 10m
 `
+	cron := `
+  snapshotting:
+    type: cron
+    prefix: zrepl_
+    timestamp_format: human
+    cron: "10 * * * *"
+`
+
+	periodicDaily := `
+  snapshotting:
+    type: periodic
+    prefix: zrepl_
+    interval: 1d
+`
 
 	hooks := `
   snapshotting:
@@ -74,10 +89,27 @@ jobs:
 		c = testValidConfig(t, fillSnapshotting(periodic))
 		snp := c.Jobs[0].Ret.(*PushJob).Snapshotting.Ret.(*SnapshottingPeriodic)
 		assert.Equal(t, "periodic", snp.Type)
-		assert.Equal(t, 10*time.Minute, snp.Interval)
+		assert.Equal(t, 10*time.Minute, snp.Interval.Duration())
 		assert.Equal(t, "zrepl_", snp.Prefix)
 	})
 
+	t.Run("periodicDaily", func(t *testing.T) {
+		c = testValidConfig(t, fillSnapshotting(periodicDaily))
+		snp := c.Jobs[0].Ret.(*PushJob).Snapshotting.Ret.(*SnapshottingPeriodic)
+		assert.Equal(t, "periodic", snp.Type)
+		assert.Equal(t, 24*time.Hour, snp.Interval.Duration())
+		assert.Equal(t, "zrepl_", snp.Prefix)
+		assert.Equal(t, "dense", snp.TimestampFormat)
+	})
+
+	t.Run("cron", func(t *testing.T) {
+		c = testValidConfig(t, fillSnapshotting(cron))
+		snp := c.Jobs[0].Ret.(*PushJob).Snapshotting.Ret.(*SnapshottingCron)
+		assert.Equal(t, "cron", snp.Type)
+		assert.Equal(t, "zrepl_", snp.Prefix)
+		assert.Equal(t, "human", snp.TimestampFormat)
+	})
+
 	t.Run("hooks", func(t *testing.T) {
 		c = testValidConfig(t, fillSnapshotting(hooks))
 		hs := c.Jobs[0].Ret.(*PushJob).Snapshotting.Ret.(*SnapshottingPeriodic).Hooks
@@ -88,3 +120,57 @@ jobs:
 	})
 
 }
+
+func TestSnapshottingTimestampDefaults(t *testing.T) {
+	tmpl := `
+jobs:
+- name: foo
+  type: push
+  connect:
+    type: local
+    listener_name: foo
+    client_identity: bar
+  filesystems: {"<": true}
+  %s
+  pruning:
+    keep_sender:
+    - type: last_n
+      count: 10
+    keep_receiver:
+    - type: last_n
+      count: 10
+`
+
+	periodic := `
+  snapshotting:
+    type: periodic
+    prefix: zrepl_
+    interval: 10m
+`
+	cron := `
+  snapshotting:
+    type: cron
+    prefix: zrepl_
+    cron: "10 * * * *"
+`
+
+	fillSnapshotting := func(s string) string { return fmt.Sprintf(tmpl, s) }
+	var c *Config
+
+	t.Run("periodic", func(t *testing.T) {
+		c = testValidConfig(t, fillSnapshotting(periodic))
+		snp := c.Jobs[0].Ret.(*PushJob).Snapshotting.Ret.(*SnapshottingPeriodic)
+		assert.Equal(t, "periodic", snp.Type)
+		assert.Equal(t, 10*time.Minute, snp.Interval.Duration())
+		assert.Equal(t, "zrepl_", snp.Prefix)
+		assert.Equal(t, "dense", snp.TimestampFormat) // default was set correctly
+	})
+
+	t.Run("cron", func(t *testing.T) {
+		c = testValidConfig(t, fillSnapshotting(cron))
+		snp := c.Jobs[0].Ret.(*PushJob).Snapshotting.Ret.(*SnapshottingCron)
+		assert.Equal(t, "cron", snp.Type)
+		assert.Equal(t, "zrepl_", snp.Prefix)
+		assert.Equal(t, "dense", snp.TimestampFormat) // default was set correctly
+	})
+}

config/config_test.go

@@ -13,6 +13,7 @@ import (
 	"github.com/kr/pretty"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
+	"github.com/zrepl/yaml-config"
 )
 
 func TestSampleConfigsAreParsedWithoutErrors(t *testing.T) {
@@ -43,7 +44,8 @@ func TestSampleConfigsAreParsedWithoutErrors(t *testing.T) {
 }
 
 // template must be a template/text template with a single '{{ . }}' as placeholder for val
-//nolint[:deadcode,unused]
+//
+//nolint:deadcode,unused
 func testValidConfigTemplate(t *testing.T, tmpl string, val string) *Config {
 	tmp, err := template.New("master").Parse(tmpl)
 	if err != nil {
@@ -86,3 +88,53 @@ func TestTrimSpaceEachLineAndPad(t *testing.T) {
 	`
 	assert.Equal(t, "  \n  foo\n  bar baz\n  \n", trimSpaceEachLineAndPad(foo, "  "))
 }
+
+func TestCronSpec(t *testing.T) {
+
+	expectAccept := []string{
+		`"* * * * *"`,
+		`"0-10 * * * *"`,
+		`"* 0-5,8,12 * * *"`,
+	}
+
+	expectFail := []string{
+		`* * * *`,
+		``,
+		`23`,
+		`"@reboot"`,
+		`"@every 1h30m"`,
+		`"@daily"`,
+		`* * * * * *`,
+	}
+
+	for _, input := range expectAccept {
+		t.Run(input, func(t *testing.T) {
+			s := fmt.Sprintf("spec: %s\n", input)
+			var v struct {
+				Spec CronSpec
+			}
+			v.Spec.Schedule = nil
+			t.Logf("input:\n%s", s)
+			err := yaml.UnmarshalStrict([]byte(s), &v)
+			t.Logf("error: %T %s", err, err)
+			require.NoError(t, err)
+			require.NotNil(t, v.Spec.Schedule)
+		})
+	}
+
+	for _, input := range expectFail {
+		t.Run(input, func(t *testing.T) {
+			s := fmt.Sprintf("spec: %s\n", input)
+			var v struct {
+				Spec CronSpec
+			}
+			v.Spec.Schedule = nil
+			t.Logf("input: %q", s)
+			err := yaml.UnmarshalStrict([]byte(s), &v)
+			t.Logf("error: %T %s", err, err)
+			require.Error(t, err)
+			require.Nil(t, v.Spec.Schedule)
+		})
+	}
+
+}

config/retentiongrid.go

@@ -37,7 +37,7 @@ func (t *RetentionIntervalList) UnmarshalYAML(u func(interface{}, bool) error) (
 		return err
 	}
 
-	intervals, err := parseRetentionGridIntervalsString(in)
+	intervals, err := ParseRetentionIntervalSpec(in)
 	if err != nil {
 		return err
 	}
@@ -102,7 +102,7 @@ func parseRetentionGridIntervalString(e string) (intervals []RetentionInterval,
 
 }
 
-func parseRetentionGridIntervalsString(s string) (intervals []RetentionInterval, err error) {
+func ParseRetentionIntervalSpec(s string) (intervals []RetentionInterval, err error) {
 
 	ges := strings.Split(s, "|")
 	intervals = make([]RetentionInterval, 0, 7*len(ges))

config/samples/bandwidth_limit.yml

@@ -0,0 +1,41 @@
+
+jobs:
+  - type: sink
+    name: "limited_sink"
+    root_fs: "fs0"
+    recv:
+      bandwidth_limit:
+        max: 12345 B
+    serve:
+      type: local
+      listener_name: localsink
+
+  - type: push
+    name: "limited_push"
+    connect:
+      type: local
+      listener_name: localsink
+      client_identity: local_backup
+    filesystems: {
+      "root<": true,
+    }
+    send:
+      bandwidth_limit:
+        max: 54321 B
+        bucket_capacity: 1024 B
+    snapshotting:
+      type: manual
+    pruning:
+      keep_sender:
+        - type: last_n
+          count: 1
+      keep_receiver:
+        - type: last_n
+          count: 1
+
+  - type: sink
+    name: "nolimit_sink"
+    root_fs: "fs1"
+    serve:
+      type: local
+      listener_name: localsink

config/samples/quickstart_backup_to_external_disk.yml

@@ -0,0 +1,88 @@
+# This config serves as an example for a local zrepl installation that
+# backups the entire zpool `system` to `backuppool/zrepl/sink`
+#
+# The requirements covered by this setup are described in the zrepl documentation's
+# quick start section which inlines this example.
+#
+# CUSTOMIZATIONS YOU WILL LIKELY WANT TO APPLY:
+# - adjust the name of the production pool `system` in the `filesystems` filter of jobs `snapjob` and `push_to_drive`
+# - adjust the name of the backup pool `backuppool` in the `backuppool_sink` job
+# - adjust the occurences of `myhostname` to the name of the system you are backing up (cannot be easily changed once you start replicating)
+# - make sure the `zrepl_` prefix is not being used by any other zfs tools you might have installed (it likely isn't)
+
+jobs:
+
+# this job takes care of snapshot creation + pruning
+- name: snapjob
+  type: snap
+  filesystems: {
+      "system<": true,
+  }
+  # create snapshots with prefix `zrepl_` every 15 minutes
+  snapshotting:
+    type: periodic
+    interval: 15m
+    prefix: zrepl_
+  pruning:
+    keep:
+    # fade-out scheme for snapshots starting with `zrepl_`
+    # - keep all created in the last hour
+    # - then destroy snapshots such that we keep 24 each 1 hour apart
+    # - then destroy snapshots such that we keep 14 each 1 day apart
+    # - then destroy all older snapshots
+    - type: grid
+      grid: 1x1h(keep=all) | 24x1h | 14x1d
+      regex: "^zrepl_.*"
+    # keep all snapshots that don't have the `zrepl_` prefix
+    - type: regex
+      negate: true
+      regex: "^zrepl_.*"
+
+# This job pushes to the local sink defined in job `backuppool_sink`.
+# We trigger replication manually from the command line / udev rules using
+#  `zrepl signal wakeup push_to_drive`
+- type: push
+  name: push_to_drive
+  connect:
+    type: local
+    listener_name: backuppool_sink
+    client_identity: myhostname
+  filesystems: {
+      "system<": true
+  }
+  send:
+    encrypted: true
+  replication:
+    protection:
+      initial: guarantee_resumability
+      # Downgrade protection to guarantee_incremental which uses zfs bookmarks instead of zfs holds.
+      # Thus, when we yank out the backup drive during replication
+      # - we might not be able to resume the interrupted replication step because the partially received `to` snapshot of a `from`->`to` step may be pruned any time
+      # - but in exchange we get back the disk space allocated by `to` when we prune it
+      # - and because we still have the bookmarks created by `guarantee_incremental`, we can still do incremental replication of `from`->`to2` in the future
+      incremental: guarantee_incremental
+  snapshotting:
+    type: manual
+  pruning:
+    # no-op prune rule on sender (keep all snapshots), job `snapshot` takes care of this
+    keep_sender:
+    - type: regex
+      regex: ".*"
+    # retain 
+    keep_receiver:
+    # longer retention on the backup drive, we have more space there
+    - type: grid
+      grid: 1x1h(keep=all) | 24x1h | 360x1d
+      regex: "^zrepl_.*"
+    # retain all non-zrepl snapshots on the backup drive
+    - type: regex
+      negate: true
+      regex: "^zrepl_.*"
+
+# This job receives from job `push_to_drive` into `backuppool/zrepl/sink/myhostname`
+- type: sink
+  name: backuppool_sink
+  root_fs: "backuppool/zrepl/sink"
+  serve:
+    type: local
+    listener_name: backuppool_sink

config/samples/quickstart_continuous_server_backup_receiver.yml

@@ -0,0 +1,12 @@
+jobs:
+- name: sink
+  type: sink
+  serve:
+      type: tls
+      listen: ":8888"
+      ca: "/etc/zrepl/prod.crt"
+      cert: "/etc/zrepl/backups.crt"
+      key: "/etc/zrepl/backups.key"
+      client_cns:
+        - "prod"
+  root_fs: "storage/zrepl/sink"

config/samples/quickstart_continuous_server_backup_sender.yml

@@ -0,0 +1,28 @@
+jobs:
+- name: prod_to_backups
+  type: push
+  connect:
+    type: tls
+    address: "backups.example.com:8888"
+    ca: /etc/zrepl/backups.crt
+    cert: /etc/zrepl/prod.crt
+    key:  /etc/zrepl/prod.key
+    server_cn: "backups"
+  filesystems: {
+    "zroot<": true,
+    "zroot/var/tmp<": false,
+    "zroot/usr/home/paranoid": false
+  }
+  snapshotting:
+    type: periodic
+    prefix: zrepl_
+    interval: 10m
+  pruning:
+    keep_sender:
+    - type: not_replicated
+    - type: last_n
+      count: 10
+    keep_receiver:
+    - type: grid
+      grid: 1x1h(keep=all) | 24x1h | 30x1d | 6x30d
+      regex: "^zrepl_"

config/samples/quickstart_fan_out_replication_source.yml

@@ -0,0 +1,59 @@
+jobs:
+# Separate job for snapshots and pruning
+- name: snapshots
+  type: snap
+  filesystems:
+    'tank<': true # all filesystems
+  snapshotting:
+    type: periodic
+    prefix: zrepl_
+    interval: 10m
+  pruning:
+    keep:
+      # Keep non-zrepl snapshots
+      - type: regex
+        negate: true
+        regex: '^zrepl_'
+      # Time-based snapshot retention
+      - type: grid
+        grid: 1x1h(keep=all) | 24x1h | 30x1d | 12x30d
+        regex: '^zrepl_'
+
+# Source job for target B
+- name: target_b
+  type: source
+  serve:
+    type: tls
+    listen: :8888
+    ca: /etc/zrepl/b.example.com.crt
+    cert: /etc/zrepl/a.example.com.crt
+    key: /etc/zrepl/a.example.com.key
+    client_cns:
+      - b.example.com
+  filesystems:
+    'tank<': true # all filesystems
+  # Snapshots are handled by the separate snap job
+  snapshotting:
+    type: manual
+
+# Source job for target C
+- name: target_c
+  type: source
+  serve:
+    type: tls
+    listen: :8889
+    ca: /etc/zrepl/c.example.com.crt
+    cert: /etc/zrepl/a.example.com.crt
+    key: /etc/zrepl/a.example.com.key
+    client_cns:
+      - c.example.com
+  filesystems:
+    'tank<': true # all filesystems
+  # Snapshots are handled by the separate snap job
+  snapshotting:
+    type: manual
+
+# Source jobs for remaining targets. Each one should listen on a different port
+# and reference the correct certificate and client CN.
+# - name: target_c
+#   ...

config/samples/quickstart_fan_out_replication_target.yml

@@ -0,0 +1,30 @@
+jobs:
+# Pull from source server A
+- name: source_a
+  type: pull
+  connect:
+    type: tls
+    # Use the correct port for this specific client (eg. B is 8888, C is 8889, etc.)
+    address: a.example.com:8888
+    ca: /etc/zrepl/a.example.com.crt
+    # Use the correct key pair for this specific client
+    cert: /etc/zrepl/b.example.com.crt
+    key: /etc/zrepl/b.example.com.key
+    server_cn: a.example.com
+  root_fs: pool0/backup
+  interval: 10m
+  pruning:
+    keep_sender:
+      # Source does the pruning in its snap job
+      - type: regex
+        regex: '.*'
+    # Receiver-side pruning can be configured as desired on each target server
+    keep_receiver:
+      # Keep non-zrepl snapshots
+      - type: regex
+        negate: true
+        regex: '^zrepl_'
+      # Time-based snapshot retention
+      - type: grid
+        grid: 1x1h(keep=all) | 24x1h | 30x1d | 12x30d
+        regex: '^zrepl_'

config/samples/snap_cron.yml

@@ -0,0 +1,14 @@
+jobs:
+- name: snapjob
+  type: snap
+  filesystems: {
+    "tank<": true,
+  }
+  snapshotting:
+    type: cron
+    prefix: zrepl_snapjob_
+    cron: "*/5 * * * *"
+  pruning:
+    keep:
+      - type: last_n
+        count: 60

daemon/control.go

@@ -8,6 +8,7 @@ import (
 	"io"
 	"net"
 	"net/http"
+	"os"
 	"time"
 
 	"github.com/pkg/errors"
@@ -120,7 +121,14 @@ func (j *controlJob) Run(ctx context.Context) {
 		jsonResponder{log, func() (interface{}, error) {
 			jobs := j.jobs.status()
 			globalZFS := zfscmd.GetReport()
-			s := Status{Jobs: jobs, Global: GlobalStatus{ZFSCmds: globalZFS}}
+			envconstReport := envconst.GetReport()
+			s := Status{
+				Jobs: jobs,
+				Global: GlobalStatus{
+					ZFSCmds:   globalZFS,
+					Envconst:  envconstReport,
+					OsEnviron: os.Environ(),
+				}}
 			return s, nil
 		}})
 
@@ -150,8 +158,8 @@ func (j *controlJob) Run(ctx context.Context) {
 	server := http.Server{
 		Handler: mux,
 		// control socket is local, 1s timeout should be more than sufficient, even on a loaded system
-		WriteTimeout: 1 * time.Second,
+		WriteTimeout: envconst.Duration("ZREPL_DAEMON_CONTROL_SERVER_WRITE_TIMEOUT", 1*time.Second),
-		ReadTimeout:  1 * time.Second,
+		ReadTimeout:  envconst.Duration("ZREPL_DAEMON_CONTROL_SERVER_READ_TIMEOUT", 1*time.Second),
 	}
 
 outer:

daemon/daemon.go

@@ -3,6 +3,7 @@ package daemon
 import (
 	"context"
 	"fmt"
+	"math/rand"
 	"os"
 	"os/signal"
 	"strings"
@@ -13,6 +14,10 @@ import (
 	"github.com/pkg/errors"
 	"github.com/prometheus/client_golang/prometheus"
 
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+	"github.com/zrepl/zrepl/endpoint"
+	"github.com/zrepl/zrepl/util/envconst"
+
 	"github.com/zrepl/zrepl/config"
 	"github.com/zrepl/zrepl/daemon/job"
 	"github.com/zrepl/zrepl/daemon/job/reset"
@@ -23,9 +28,8 @@ import (
 	"github.com/zrepl/zrepl/zfs/zfscmd"
 )
 
-func Run(conf *config.Config) error {
+func Run(ctx context.Context, conf *config.Config) error {
-
+	ctx, cancel := context.WithCancel(ctx)
-	ctx, cancel := context.WithCancel(context.Background())
 
 	defer cancel()
 	sigChan := make(chan os.Signal, 1)
@@ -35,12 +39,19 @@ func Run(conf *config.Config) error {
 		cancel()
 	}()
 
+	// The math/rand package is used presently for generating trace IDs, we
+	// seed it with the current time and pid so that the IDs are mostly
+	// unique.
+	rand.Seed(time.Now().UnixNano())
+	rand.Seed(int64(os.Getpid()))
+
 	outlets, err := logging.OutletsFromConfig(*conf.Global.Logging)
 	if err != nil {
 		return errors.Wrap(err, "cannot build logging from config")
 	}
+	outlets.Add(newPrometheusLogOutlet(), logger.Debug)
 
-	confJobs, err := job.JobsFromConfig(conf)
+	confJobs, err := job.JobsFromConfig(conf, config.ParseFlagsNone)
 	if err != nil {
 		return errors.Wrap(err, "cannot build jobs from config")
 	}
@@ -48,14 +59,23 @@ func Run(conf *config.Config) error {
 	log := logger.NewLogger(outlets, 1*time.Second)
 	log.Info(version.NewZreplVersionInformation().String())
 
+	ctx = logging.WithLoggers(ctx, logging.SubsystemLoggersWithUniversalLogger(log))
+	trace.RegisterCallback(trace.Callback{
+		OnBegin: func(ctx context.Context) { logging.GetLogger(ctx, logging.SubsysTraceData).Debug("begin span") },
+		OnEnd: func(ctx context.Context, spanInfo trace.SpanInfo) {
+			logging.
+				GetLogger(ctx, logging.SubsysTraceData).
+				WithField("duration_s", spanInfo.EndedAt().Sub(spanInfo.StartedAt()).Seconds()).
+				Debug("finished span " + spanInfo.TaskAndSpanStack(trace.SpanStackKindAnnotation))
+		},
+	})
+
 	for _, job := range confJobs {
 		if IsInternalJobName(job.Name()) {
 			panic(fmt.Sprintf("internal job name used for config job '%s'", job.Name())) //FIXME
 		}
 	}
 
-	ctx = job.WithLogger(ctx, log)
-
 	jobs := newJobs()
 
 	// start control socket
@@ -83,7 +103,10 @@ func Run(conf *config.Config) error {
 	}
 
 	// register global (=non job-local) metrics
+	version.PrometheusRegister(prometheus.DefaultRegisterer)
 	zfscmd.RegisterMetrics(prometheus.DefaultRegisterer)
+	trace.RegisterMetrics(prometheus.DefaultRegisterer)
+	endpoint.RegisterMetrics(prometheus.DefaultRegisterer)
 
 	log.Info("starting daemon")
 
@@ -98,6 +121,8 @@ func Run(conf *config.Config) error {
 	case <-ctx.Done():
 		log.WithError(ctx.Err()).Info("context finished")
 	}
+	log.Info("waiting for jobs to finish")
+	<-jobs.wait()
 	log.Info("daemon exiting")
 	return nil
 }
@@ -120,14 +145,11 @@ func newJobs() *jobs {
 	}
 }
 
-const (
-	logJobField string = "job"
-)
-
 func (s *jobs) wait() <-chan struct{} {
 	ch := make(chan struct{})
 	go func() {
 		s.wg.Wait()
+		close(ch)
 	}()
 	return ch
 }
@@ -138,7 +160,9 @@ type Status struct {
 }
 
 type GlobalStatus struct {
-	ZFSCmds *zfscmd.Report
+	ZFSCmds   *zfscmd.Report
+	Envconst  *envconst.Report
+	OsEnviron []string
 }
 
 func (s *jobs) status() map[string]*job.Status {
@@ -202,9 +226,8 @@ func (s *jobs) start(ctx context.Context, j job.Job, internal bool) {
 	s.m.Lock()
 	defer s.m.Unlock()
 
-	jobLog := job.GetLogger(ctx).
+	ctx = logging.WithInjectedField(ctx, logging.JobField, j.Name())
-		WithField(logJobField, j.Name()).
+
-		WithOutlet(newPrometheusLogOutlet(j.Name()), logger.Debug)
 	jobName := j.Name()
 	if !internal && IsInternalJobName(jobName) {
 		panic(fmt.Sprintf("internal job name used for non-internal job %s", jobName))
@@ -219,7 +242,6 @@ func (s *jobs) start(ctx context.Context, j job.Job, internal bool) {
 	j.RegisterMetrics(prometheus.DefaultRegisterer)
 
 	s.jobs[jobName] = j
-	ctx = job.WithLogger(ctx, jobLog)
 	ctx = zfscmd.WithJobID(ctx, j.Name())
 	ctx, wakeup := wakeup.Context(ctx)
 	ctx, resetFunc := reset.Context(ctx)
@@ -229,8 +251,8 @@ func (s *jobs) start(ctx context.Context, j job.Job, internal bool) {
 	s.wg.Add(1)
 	go func() {
 		defer s.wg.Done()
-		jobLog.Info("starting job")
+		job.GetLogger(ctx).Info("starting job")
-		defer jobLog.Info("job exited")
+		defer job.GetLogger(ctx).Info("job exited")
 		j.Run(ctx)
 	}()
 }

daemon/filters/fsmapfilter.go

@@ -160,6 +160,14 @@ 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.
 func (m DatasetMapFilter) InvertedFilter() (inv *DatasetMapFilter, err error) {

daemon/hooks/hook_docs.go

@@ -5,7 +5,7 @@
 //
 // This package also provides all supported hook type implementations and abstractions around them.
 //
-// Use For Other Kinds Of ExpectStepReports
+// # Use For Other Kinds Of ExpectStepReports
 //
 // This package REQUIRES REFACTORING before it can be used for other activities than snapshots, e.g. pre- and post-replication:
 //
@@ -15,7 +15,7 @@
 // The hook implementations should move out of this package.
 // However, there is a lot of tight coupling which to untangle isn't worth it ATM.
 //
-// How This Package Is Used By Package Snapper
+// # How This Package Is Used By Package Snapper
 //
 // Deserialize a config.List using ListFromConfig().
 // Then it MUST filter the list to only contain hooks for a particular filesystem using
@@ -30,5 +30,4 @@
 // Command hooks make it available in the environment variable ZREPL_DRYRUN.
 //
 // Plan.Report() can be called while Plan.Run() is executing to give an overview of plan execution progress (future use in "zrepl status").
-//
 package hooks

daemon/hooks/hook_logging.go

@@ -6,29 +6,17 @@ import (
 	"context"
 	"sync"
 
+	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/logger"
 	"github.com/zrepl/zrepl/util/envconst"
 )
 
-type contextKey int
-
-const (
-	contextKeyLog contextKey = 0
-)
-
 type Logger = logger.Logger
 
-func WithLogger(ctx context.Context, log Logger) context.Context {
-	return context.WithValue(ctx, contextKeyLog, log)
-}
-
 func GetLogger(ctx context.Context) Logger { return getLogger(ctx) }
 
 func getLogger(ctx context.Context) Logger {
-	if log, ok := ctx.Value(contextKeyLog).(Logger); ok {
+	return logging.GetLogger(ctx, logging.SubsysHooks)
-		return log
-	}
-	return logger.NewNullLogger()
 }
 
 const MAX_HOOK_LOG_SIZE_DEFAULT int = 1 << 20

daemon/hooks/hook_type_command.go

@@ -93,7 +93,14 @@ func (r *CommandHookReport) String() string {
 		cmdLine.WriteString(fmt.Sprintf("%s'%s'", sep, a))
 	}
 
-	return fmt.Sprintf("command hook invocation: \"%s\"", cmdLine.String()) // no %q to make copy-pastable
+	var msg string
+	if r.Err == nil {
+		msg = "command hook"
+	} else {
+		msg = fmt.Sprintf("command hook failed with %q", r.Err)
+	}
+
+	return fmt.Sprintf("%s: \"%s\"", msg, cmdLine.String()) // no %q to make copy-pastable
 }
 func (r *CommandHookReport) Error() string {
 	if r.Err == nil {

daemon/hooks/hook_type_mysql_lock_tables.go

@@ -17,19 +17,22 @@ import (
 	"github.com/zrepl/zrepl/zfs"
 )
 
+// Hook to implement the following recommmendation from MySQL docs
 // https://dev.mysql.com/doc/mysql-backup-excerpt/5.7/en/backup-methods.html
 //
-// Making Backups Using a File System Snapshot:
+//	Making Backups Using a File System Snapshot:
 //
-//  If you are using a Veritas file system, you can make a backup like this:
+//	If you are using a Veritas file system, you can make a backup like this:
 //
-//    	From a client program, execute FLUSH TABLES WITH READ LOCK.
+//	   	From a client program, execute FLUSH TABLES WITH READ LOCK.
-//    	From another shell, execute mount vxfs snapshot.
+//	   	From another shell, execute mount vxfs snapshot.
-//    	From the first client, execute UNLOCK TABLES.
+//	   	From the first client, execute UNLOCK TABLES.
-//    	Copy files from the snapshot.
+//	   	Copy files from the snapshot.
-//    	Unmount the snapshot.
+//	   	Unmount the snapshot.
 //
-//	Similar snapshot capabilities may be available in other file systems, such as LVM or ZFS.
+//		Similar snapshot capabilities may be available in other file systems, such as LVM or ZFS.
+//
+
 type MySQLLockTables struct {
 	errIsFatal  bool
 	connector   sqldriver.Connector

daemon/hooks/hooks_test.go

@@ -11,8 +11,11 @@ import (
 
 	"github.com/stretchr/testify/require"
 
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+
 	"github.com/zrepl/zrepl/config"
 	"github.com/zrepl/zrepl/daemon/hooks"
+	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/logger"
 	"github.com/zrepl/zrepl/zfs"
 )
@@ -69,6 +72,9 @@ func curry(f comparisonAssertionFunc, expected interface{}, right bool) (ret val
 }
 
 func TestHooks(t *testing.T) {
+	ctx, end := trace.WithTaskFromStack(context.Background())
+	defer end()
+
 	testFSName := "testpool/testdataset"
 	testSnapshotName := "testsnap"
 
@@ -155,7 +161,7 @@ jobs:
 					ExpectedEdge: hooks.Pre,
 					ExpectStatus: hooks.StepErr,
 					OutputTest:   containsTest(fmt.Sprintf("TEST ERROR pre_testing %s@%s", testFSName, testSnapshotName)),
-					ErrorTest:    regexpTest("^command hook invocation.*exit status 1$"),
+					ErrorTest:    regexpTest("^command hook failed.*exit status 1$"),
 				},
 				expectStep{ExpectedEdge: hooks.Callback, ExpectStatus: hooks.StepOk},
 				expectStep{
@@ -179,7 +185,7 @@ jobs:
 					ExpectedEdge: hooks.Pre,
 					ExpectStatus: hooks.StepErr,
 					OutputTest:   containsTest(fmt.Sprintf("TEST ERROR pre_testing %s@%s", testFSName, testSnapshotName)),
-					ErrorTest:    regexpTest("^command hook invocation.*exit status 1$"),
+					ErrorTest:    regexpTest("^command hook failed.*exit status 1$"),
 				},
 				expectStep{
 					ExpectedEdge: hooks.Pre,
@@ -228,7 +234,7 @@ jobs:
 					ExpectedEdge: hooks.Post,
 					ExpectStatus: hooks.StepErr,
 					OutputTest:   containsTest(fmt.Sprintf("TEST ERROR post_testing %s@%s", testFSName, testSnapshotName)),
-					ErrorTest:    regexpTest("^command hook invocation.*exit status 1$"),
+					ErrorTest:    regexpTest("^command hook failed.*exit status 1$"),
 				},
 			},
 		},
@@ -261,7 +267,7 @@ jobs:
 					ExpectedEdge: hooks.Pre,
 					ExpectStatus: hooks.StepErr,
 					OutputTest:   containsTest(fmt.Sprintf("TEST ERROR pre_testing %s@%s", testFSName, testSnapshotName)),
-					ErrorTest:    regexpTest("^command hook invocation.*exit status 1$"),
+					ErrorTest:    regexpTest("^command hook failed.*exit status 1$"),
 				},
 				expectStep{ExpectedEdge: hooks.Callback, ExpectStatus: hooks.StepOk},
 				expectStep{
@@ -289,7 +295,7 @@ jobs:
 					ExpectedEdge: hooks.Pre,
 					ExpectStatus: hooks.StepErr,
 					OutputTest:   containsTest(fmt.Sprintf("TEST ERROR pre_testing %s@%s", testFSName, testSnapshotName)),
-					ErrorTest:    regexpTest("^command hook invocation.*exit status 1$"),
+					ErrorTest:    regexpTest("^command hook failed.*exit status 1$"),
 				},
 				expectStep{ExpectedEdge: hooks.Callback, ExpectStatus: hooks.StepOk},
 				expectStep{
@@ -418,9 +424,8 @@ jobs:
 
 			cbReached = false
 
-			ctx := context.Background()
 			if testing.Verbose() && !tt.SuppressOutput {
-				ctx = hooks.WithLogger(ctx, log)
+				ctx = logging.WithLoggers(ctx, logging.SubsystemLoggersWithUniversalLogger(log))
 			}
 			plan.Run(ctx, false)
 			report := plan.Report()

daemon/job/active.go

@@ -8,12 +8,14 @@ import (
 
 	"github.com/pkg/errors"
 	"github.com/prometheus/client_golang/prometheus"
+	"github.com/prometheus/common/log"
+
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+	"github.com/zrepl/zrepl/util/envconst"
 
 	"github.com/zrepl/zrepl/config"
-	"github.com/zrepl/zrepl/daemon/filters"
 	"github.com/zrepl/zrepl/daemon/job/reset"
 	"github.com/zrepl/zrepl/daemon/job/wakeup"
-	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/daemon/pruner"
 	"github.com/zrepl/zrepl/daemon/snapper"
 	"github.com/zrepl/zrepl/endpoint"
@@ -32,11 +34,15 @@ type ActiveSide struct {
 	name      endpoint.JobID
 	connecter transport.Connecter
 
+	replicationDriverConfig driver.Config
+
 	prunerFactory *pruner.PrunerFactory
 
-	promRepStateSecs    *prometheus.HistogramVec // labels: state
+	promRepStateSecs      *prometheus.HistogramVec // labels: state
-	promPruneSecs       *prometheus.HistogramVec // labels: prune_side
+	promPruneSecs         *prometheus.HistogramVec // labels: prune_side
-	promBytesReplicated *prometheus.CounterVec   // labels: filesystem
+	promBytesReplicated   *prometheus.CounterVec   // labels: filesystem
+	promReplicationErrors prometheus.Gauge
+	promLastSuccessful    prometheus.Gauge
 
 	tasksMtx sync.Mutex
 	tasks    activeSideTasks
@@ -79,7 +85,7 @@ func (a *ActiveSide) updateTasks(u func(*activeSideTasks)) activeSideTasks {
 }
 
 type activeMode interface {
-	ConnectEndpoints(rpcLoggers rpc.Loggers, connecter transport.Connecter)
+	ConnectEndpoints(ctx context.Context, connecter transport.Connecter)
 	DisconnectEndpoints()
 	SenderReceiver() (logic.Sender, logic.Receiver)
 	Type() Type
@@ -95,17 +101,17 @@ type modePush struct {
 	receiver      *rpc.Client
 	senderConfig  *endpoint.SenderConfig
 	plannerPolicy *logic.PlannerPolicy
-	snapper       *snapper.PeriodicOrManual
+	snapper       snapper.Snapper
 }
 
-func (m *modePush) ConnectEndpoints(loggers rpc.Loggers, connecter transport.Connecter) {
+func (m *modePush) ConnectEndpoints(ctx context.Context, connecter transport.Connecter) {
 	m.setupMtx.Lock()
 	defer m.setupMtx.Unlock()
 	if m.receiver != nil || m.sender != nil {
 		panic("inconsistent use of ConnectEndpoints and DisconnectEndpoints")
 	}
 	m.sender = endpoint.NewSender(*m.senderConfig)
-	m.receiver = rpc.NewClient(connecter, loggers)
+	m.receiver = rpc.NewClient(connecter, rpc.GetLoggersOrPanic(ctx))
 }
 
 func (m *modePush) DisconnectEndpoints() {
@@ -131,7 +137,8 @@ func (m *modePush) RunPeriodic(ctx context.Context, wakeUpCommon chan<- struct{}
 }
 
 func (m *modePush) SnapperReport() *snapper.Report {
-	return m.snapper.Report()
+	r := m.snapper.Report()
+	return &r
 }
 
 func (m *modePush) ResetConnectBackoff() {
@@ -144,22 +151,33 @@ func (m *modePush) ResetConnectBackoff() {
 
 func modePushFromConfig(g *config.Global, in *config.PushJob, jobID endpoint.JobID) (*modePush, error) {
 	m := &modePush{}
+	var err error
 
-	fsf, err := filters.DatasetMapFilterFromConfig(in.Filesystems)
+	m.senderConfig, err = buildSenderConfig(in, jobID)
 	if err != nil {
-		return nil, errors.Wrap(err, "cannot build filesystem filter")
+		return nil, errors.Wrap(err, "sender config")
 	}
 
-	m.senderConfig = &endpoint.SenderConfig{
+	replicationConfig, err := logic.ReplicationConfigFromConfig(in.Replication)
-		FSF:     fsf,
+	if err != nil {
-		Encrypt: &zfs.NilBool{B: in.Send.Encrypted},
+		return nil, errors.Wrap(err, "field `replication`")
-		JobID:   jobID,
 	}
+
+	conflictResolution, err := logic.ConflictResolutionFromConfig(in.ConflictResolution)
+	if err != nil {
+		return nil, errors.Wrap(err, "field `conflict_resolution`")
+	}
+
 	m.plannerPolicy = &logic.PlannerPolicy{
-		EncryptedSend: logic.TriFromBool(in.Send.Encrypted),
+		ConflictResolution:        conflictResolution,
+		ReplicationConfig:         replicationConfig,
+		SizeEstimationConcurrency: in.Replication.Concurrency.SizeEstimates,
+	}
+	if err := m.plannerPolicy.Validate(); err != nil {
+		return nil, errors.Wrap(err, "cannot build planner policy")
 	}
 
-	if m.snapper, err = snapper.FromConfig(g, fsf, in.Snapshotting); err != nil {
+	if m.snapper, err = snapper.FromConfig(g, m.senderConfig.FSF, in.Snapshotting); err != nil {
 		return nil, errors.Wrap(err, "cannot build snapper")
 	}
 
@@ -171,19 +189,18 @@ type modePull struct {
 	receiver       *endpoint.Receiver
 	receiverConfig endpoint.ReceiverConfig
 	sender         *rpc.Client
-	rootFS         *zfs.DatasetPath
 	plannerPolicy  *logic.PlannerPolicy
 	interval       config.PositiveDurationOrManual
 }
 
-func (m *modePull) ConnectEndpoints(loggers rpc.Loggers, connecter transport.Connecter) {
+func (m *modePull) ConnectEndpoints(ctx context.Context, connecter transport.Connecter) {
 	m.setupMtx.Lock()
 	defer m.setupMtx.Unlock()
 	if m.receiver != nil || m.sender != nil {
 		panic("inconsistent use of ConnectEndpoints and DisconnectEndpoints")
 	}
 	m.receiver = endpoint.NewReceiver(m.receiverConfig)
-	m.sender = rpc.NewClient(connecter, loggers)
+	m.sender = rpc.NewClient(connecter, rpc.GetLoggersOrPanic(ctx))
 }
 
 func (m *modePull) DisconnectEndpoints() {
@@ -245,32 +262,44 @@ func modePullFromConfig(g *config.Global, in *config.PullJob, jobID endpoint.Job
 	m = &modePull{}
 	m.interval = in.Interval
 
-	m.rootFS, err = zfs.NewDatasetPath(in.RootFS)
+	replicationConfig, err := logic.ReplicationConfigFromConfig(in.Replication)
 	if err != nil {
-		return nil, errors.New("RootFS is not a valid zfs filesystem path")
+		return nil, errors.Wrap(err, "field `replication`")
 	}
-	if m.rootFS.Length() <= 0 {
+
-		return nil, errors.New("RootFS must not be empty") // duplicates error check of receiver
+	conflictResolution, err := logic.ConflictResolutionFromConfig(in.ConflictResolution)
+	if err != nil {
+		return nil, errors.Wrap(err, "field `conflict_resolution`")
 	}
 
 	m.plannerPolicy = &logic.PlannerPolicy{
-		EncryptedSend: logic.DontCare,
+		ConflictResolution:        conflictResolution,
+		ReplicationConfig:         replicationConfig,
+		SizeEstimationConcurrency: in.Replication.Concurrency.SizeEstimates,
+	}
+	if err := m.plannerPolicy.Validate(); err != nil {
+		return nil, errors.Wrap(err, "cannot build planner policy")
 	}
 
-	m.receiverConfig = endpoint.ReceiverConfig{
+	m.receiverConfig, err = buildReceiverConfig(in, jobID)
-		JobID:                      jobID,
+	if err != nil {
-		RootWithoutClientComponent: m.rootFS,
+		return nil, err
-		AppendClientIdentity:       false, // !
-		UpdateLastReceivedHold:     true,
-	}
-	if err := m.receiverConfig.Validate(); err != nil {
-		return nil, errors.Wrap(err, "cannot build receiver config")
 	}
 
 	return m, nil
 }
 
-func activeSide(g *config.Global, in *config.ActiveJob, configJob interface{}) (j *ActiveSide, err error) {
+func replicationDriverConfigFromConfig(in *config.Replication) (c driver.Config, err error) {
+	c = driver.Config{
+		StepQueueConcurrency:     in.Concurrency.Steps,
+		MaxAttempts:              envconst.Int("ZREPL_REPLICATION_MAX_ATTEMPTS", 3),
+		ReconnectHardFailTimeout: envconst.Duration("ZREPL_REPLICATION_RECONNECT_HARD_FAIL_TIMEOUT", 10*time.Minute),
+	}
+	err = c.Validate()
+	return c, err
+}
+
+func activeSide(g *config.Global, in *config.ActiveJob, configJob interface{}, parseFlags config.ParseFlags) (j *ActiveSide, err error) {
 
 	j = &ActiveSide{}
 	j.name, err = endpoint.MakeJobID(in.Name)
@@ -304,8 +333,22 @@ func activeSide(g *config.Global, in *config.ActiveJob, configJob interface{}) (
 		Help:        "number of bytes replicated from sender to receiver per filesystem",
 		ConstLabels: prometheus.Labels{"zrepl_job": j.name.String()},
 	}, []string{"filesystem"})
+	j.promReplicationErrors = prometheus.NewGauge(prometheus.GaugeOpts{
+		Namespace:   "zrepl",
+		Subsystem:   "replication",
+		Name:        "filesystem_errors",
+		Help:        "number of filesystems that failed replication in the latest replication attempt, or -1 if the job failed before enumerating the filesystems",
+		ConstLabels: prometheus.Labels{"zrepl_job": j.name.String()},
+	})
+	j.promLastSuccessful = prometheus.NewGauge(prometheus.GaugeOpts{
+		Namespace:   "zrepl",
+		Subsystem:   "replication",
+		Name:        "last_successful",
+		Help:        "timestamp of last successful replication",
+		ConstLabels: prometheus.Labels{"zrepl_job": j.name.String()},
+	})
 
-	j.connecter, err = fromconfig.ConnecterFromConfig(g, in.Connect)
+	j.connecter, err = fromconfig.ConnecterFromConfig(g, in.Connect, parseFlags)
 	if err != nil {
 		return nil, errors.Wrap(err, "cannot build client")
 	}
@@ -322,6 +365,11 @@ func activeSide(g *config.Global, in *config.ActiveJob, configJob interface{}) (
 		return nil, err
 	}
 
+	j.replicationDriverConfig, err = replicationDriverConfigFromConfig(in.Replication)
+	if err != nil {
+		return nil, errors.Wrap(err, "cannot build replication driver config")
+	}
+
 	return j, nil
 }
 
@@ -329,6 +377,8 @@ func (j *ActiveSide) RegisterMetrics(registerer prometheus.Registerer) {
 	registerer.MustRegister(j.promRepStateSecs)
 	registerer.MustRegister(j.promPruneSecs)
 	registerer.MustRegister(j.promBytesReplicated)
+	registerer.MustRegister(j.promReplicationErrors)
+	registerer.MustRegister(j.promLastSuccessful)
 }
 
 func (j *ActiveSide) Name() string { return j.name.String() }
@@ -363,7 +413,7 @@ func (j *ActiveSide) OwnedDatasetSubtreeRoot() (rfs *zfs.DatasetPath, ok bool) {
 		_ = j.mode.(*modePush) // make sure we didn't introduce a new job type
 		return nil, false
 	}
-	return pull.rootFS.Copy(), true
+	return pull.receiverConfig.RootWithoutClientComponent.Copy(), true
 }
 
 func (j *ActiveSide) SenderConfig() *endpoint.SenderConfig {
@@ -375,16 +425,31 @@ func (j *ActiveSide) SenderConfig() *endpoint.SenderConfig {
 	return push.senderConfig
 }
 
+// The active side of a replication uses one end (sender or receiver)
+// directly by method invocation, without going through a transport that
+// provides a client identity.
+// However, in order to avoid the need to distinguish between direct-method-invocating
+// clients and RPC client, we use an invalid client identity as a sentinel value.
+func FakeActiveSideDirectMethodInvocationClientIdentity(jobId endpoint.JobID) string {
+	return fmt.Sprintf("<local><active><job><client><identity><job=%q>", jobId.String())
+}
+
 func (j *ActiveSide) Run(ctx context.Context) {
+	ctx, endTask := trace.WithTaskAndSpan(ctx, "active-side-job", j.Name())
+	defer endTask()
+
+	ctx = context.WithValue(ctx, endpoint.ClientIdentityKey, FakeActiveSideDirectMethodInvocationClientIdentity(j.name))
+
 	log := GetLogger(ctx)
-	ctx = logging.WithSubsystemLoggers(ctx, log)
 
 	defer log.Info("job exiting")
 
 	periodicDone := make(chan struct{})
 	ctx, cancel := context.WithCancel(ctx)
 	defer cancel()
-	go j.mode.RunPeriodic(ctx, periodicDone)
+	periodicCtx, endTask := trace.WithTask(ctx, "periodic")
+	defer endTask()
+	go j.mode.RunPeriodic(periodicCtx, periodicDone)
 
 	invocationCount := 0
 outer:
@@ -400,17 +465,15 @@ outer:
 		case <-periodicDone:
 		}
 		invocationCount++
-		invLog := log.WithField("invocation", invocationCount)
+		invocationCtx, endSpan := trace.WithSpan(ctx, fmt.Sprintf("invocation-%d", invocationCount))
-		j.do(WithLogger(ctx, invLog))
+		j.do(invocationCtx)
+		endSpan()
 	}
 }
 
 func (j *ActiveSide) do(ctx context.Context) {
 
-	log := GetLogger(ctx)
+	j.mode.ConnectEndpoints(ctx, j.connecter)
-	ctx = logging.WithSubsystemLoggers(ctx, log)
-	loggers := rpc.GetLoggersOrPanic(ctx) // filled by WithSubsystemLoggers
-	j.mode.ConnectEndpoints(loggers, j.connecter)
 	defer j.mode.DisconnectEndpoints()
 
 	// allow cancellation of an invocation (this function)
@@ -433,20 +496,30 @@ func (j *ActiveSide) do(ctx context.Context) {
 			return
 		default:
 		}
+		ctx, endSpan := trace.WithSpan(ctx, "replication")
 		ctx, repCancel := context.WithCancel(ctx)
 		var repWait driver.WaitFunc
 		j.updateTasks(func(tasks *activeSideTasks) {
 			// reset it
 			*tasks = activeSideTasks{}
-			tasks.replicationCancel = repCancel
+			tasks.replicationCancel = func() { repCancel(); endSpan() }
 			tasks.replicationReport, repWait = replication.Do(
-				ctx, logic.NewPlanner(j.promRepStateSecs, j.promBytesReplicated, sender, receiver, j.mode.PlannerPolicy()),
+				ctx, j.replicationDriverConfig, logic.NewPlanner(j.promRepStateSecs, j.promBytesReplicated, sender, receiver, j.mode.PlannerPolicy()),
 			)
 			tasks.state = ActiveSideReplicating
 		})
-		log.Info("start replication")
+		GetLogger(ctx).Info("start replication")
 		repWait(true) // wait blocking
 		repCancel()   // always cancel to free up context resources
+
+		replicationReport := j.tasks.replicationReport()
+		var numErrors = replicationReport.GetFailedFilesystemsCountInLatestAttempt()
+		j.promReplicationErrors.Set(float64(numErrors))
+		if numErrors == 0 {
+			j.promLastSuccessful.SetToCurrentTime()
+		}
+
+		endSpan()
 	}
 
 	{
@@ -455,16 +528,18 @@ func (j *ActiveSide) do(ctx context.Context) {
 			return
 		default:
 		}
+		ctx, endSpan := trace.WithSpan(ctx, "prune_sender")
 		ctx, senderCancel := context.WithCancel(ctx)
 		tasks := j.updateTasks(func(tasks *activeSideTasks) {
 			tasks.prunerSender = j.prunerFactory.BuildSenderPruner(ctx, sender, sender)
-			tasks.prunerSenderCancel = senderCancel
+			tasks.prunerSenderCancel = func() { senderCancel(); endSpan() }
 			tasks.state = ActiveSidePruneSender
 		})
-		log.Info("start pruning sender")
+		GetLogger(ctx).Info("start pruning sender")
 		tasks.prunerSender.Prune()
-		log.Info("finished pruning sender")
+		GetLogger(ctx).Info("finished pruning sender")
 		senderCancel()
+		endSpan()
 	}
 	{
 		select {
@@ -472,16 +547,18 @@ func (j *ActiveSide) do(ctx context.Context) {
 			return
 		default:
 		}
+		ctx, endSpan := trace.WithSpan(ctx, "prune_recever")
 		ctx, receiverCancel := context.WithCancel(ctx)
 		tasks := j.updateTasks(func(tasks *activeSideTasks) {
 			tasks.prunerReceiver = j.prunerFactory.BuildReceiverPruner(ctx, receiver, sender)
-			tasks.prunerReceiverCancel = receiverCancel
+			tasks.prunerReceiverCancel = func() { receiverCancel(); endSpan() }
 			tasks.state = ActiveSidePruneReceiver
 		})
-		log.Info("start pruning receiver")
+		GetLogger(ctx).Info("start pruning receiver")
 		tasks.prunerReceiver.Prune()
-		log.Info("finished pruning receiver")
+		GetLogger(ctx).Info("finished pruning receiver")
 		receiverCancel()
+		endSpan()
 	}
 
 	j.updateTasks(func(tasks *activeSideTasks) {

daemon/job/active_test.go

@@ -0,0 +1,20 @@
+package job
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+
+	"github.com/zrepl/zrepl/endpoint"
+	"github.com/zrepl/zrepl/transport"
+)
+
+func TestFakeActiveSideDirectMethodInvocationClientIdentityDoesNotPassValidityTest(t *testing.T) {
+	jobid, err := endpoint.MakeJobID("validjobname")
+	require.NoError(t, err)
+	clientIdentity := FakeActiveSideDirectMethodInvocationClientIdentity(jobid)
+	t.Logf("%v", clientIdentity)
+	err = transport.ValidateClientIdentity(clientIdentity)
+	assert.Error(t, err)
+}

daemon/job/build_jobs.go

@@ -8,12 +8,13 @@ import (
 	"github.com/pkg/errors"
 
 	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/util/bandwidthlimit"
 )
 
-func JobsFromConfig(c *config.Config) ([]Job, error) {
+func JobsFromConfig(c *config.Config, parseFlags config.ParseFlags) ([]Job, error) {
 	js := make([]Job, len(c.Jobs))
 	for i := range c.Jobs {
-		j, err := buildJob(c.Global, c.Jobs[i])
+		j, err := buildJob(c.Global, c.Jobs[i], parseFlags)
 		if err != nil {
 			return nil, err
 		}
@@ -41,19 +42,19 @@ func JobsFromConfig(c *config.Config) ([]Job, error) {
 	return js, nil
 }
 
-func buildJob(c *config.Global, in config.JobEnum) (j Job, err error) {
+func buildJob(c *config.Global, in config.JobEnum, parseFlags config.ParseFlags) (j Job, err error) {
 	cannotBuildJob := func(e error, name string) (Job, error) {
 		return nil, errors.Wrapf(e, "cannot build job %q", name)
 	}
 	// FIXME prettify this
 	switch v := in.Ret.(type) {
 	case *config.SinkJob:
-		j, err = passiveSideFromConfig(c, &v.PassiveJob, v)
+		j, err = passiveSideFromConfig(c, &v.PassiveJob, v, parseFlags)
 		if err != nil {
 			return cannotBuildJob(err, v.Name)
 		}
 	case *config.SourceJob:
-		j, err = passiveSideFromConfig(c, &v.PassiveJob, v)
+		j, err = passiveSideFromConfig(c, &v.PassiveJob, v, parseFlags)
 		if err != nil {
 			return cannotBuildJob(err, v.Name)
 		}
@@ -63,12 +64,12 @@ func buildJob(c *config.Global, in config.JobEnum) (j Job, err error) {
 			return cannotBuildJob(err, v.Name)
 		}
 	case *config.PushJob:
-		j, err = activeSide(c, &v.ActiveJob, v)
+		j, err = activeSide(c, &v.ActiveJob, v, parseFlags)
 		if err != nil {
 			return cannotBuildJob(err, v.Name)
 		}
 	case *config.PullJob:
-		j, err = activeSide(c, &v.ActiveJob, v)
+		j, err = activeSide(c, &v.ActiveJob, v, parseFlags)
 		if err != nil {
 			return cannotBuildJob(err, v.Name)
 		}
@@ -107,3 +108,13 @@ func validateReceivingSidesDoNotOverlap(receivingRootFSs []string) error {
 	}
 	return nil
 }
+
+func buildBandwidthLimitConfig(in *config.BandwidthLimit) (c bandwidthlimit.Config, _ error) {
+	if in.Max.ToBytes() > 0 && int64(in.Max.ToBytes()) == 0 {
+		return c, fmt.Errorf("bandwidth limit `max` is too small, must at least specify one byte")
+	}
+	return bandwidthlimit.Config{
+		Max:            int64(in.Max.ToBytes()),
+		BucketCapacity: int64(in.BucketCapacity.ToBytes()),
+	}, nil
+}

daemon/job/build_jobs_sendrecvoptions.go

@@ -0,0 +1,102 @@
+package job
+
+import (
+	"github.com/pkg/errors"
+
+	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/daemon/filters"
+	"github.com/zrepl/zrepl/endpoint"
+	"github.com/zrepl/zrepl/util/nodefault"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+type SendingJobConfig interface {
+	GetFilesystems() config.FilesystemsFilter
+	GetSendOptions() *config.SendOptions // must not be nil
+}
+
+func buildSenderConfig(in SendingJobConfig, jobID endpoint.JobID) (*endpoint.SenderConfig, error) {
+
+	fsf, err := filters.DatasetMapFilterFromConfig(in.GetFilesystems())
+	if err != nil {
+		return nil, errors.Wrap(err, "cannot build filesystem filter")
+	}
+	sendOpts := in.GetSendOptions()
+	bwlim, err := buildBandwidthLimitConfig(sendOpts.BandwidthLimit)
+	if err != nil {
+		return nil, errors.Wrap(err, "cannot build bandwith limit config")
+	}
+
+	sc := &endpoint.SenderConfig{
+		FSF:   fsf,
+		JobID: jobID,
+
+		Encrypt:              &nodefault.Bool{B: sendOpts.Encrypted},
+		SendRaw:              sendOpts.Raw,
+		SendProperties:       sendOpts.SendProperties,
+		SendBackupProperties: sendOpts.BackupProperties,
+		SendLargeBlocks:      sendOpts.LargeBlocks,
+		SendCompressed:       sendOpts.Compressed,
+		SendEmbeddedData:     sendOpts.EmbeddedData,
+		SendSaved:            sendOpts.Saved,
+
+		BandwidthLimit: bwlim,
+	}
+
+	if err := sc.Validate(); err != nil {
+		return nil, errors.Wrap(err, "cannot build sender config")
+	}
+
+	return sc, nil
+}
+
+type ReceivingJobConfig interface {
+	GetRootFS() string
+	GetAppendClientIdentity() bool
+	GetRecvOptions() *config.RecvOptions
+}
+
+func buildReceiverConfig(in ReceivingJobConfig, jobID endpoint.JobID) (rc endpoint.ReceiverConfig, err error) {
+	rootFs, err := zfs.NewDatasetPath(in.GetRootFS())
+	if err != nil {
+		return rc, errors.New("root_fs is not a valid zfs filesystem path")
+	}
+	if rootFs.Length() <= 0 {
+		return rc, errors.New("root_fs must not be empty") // duplicates error check of receiver
+	}
+
+	recvOpts := in.GetRecvOptions()
+
+	bwlim, err := buildBandwidthLimitConfig(recvOpts.BandwidthLimit)
+	if err != nil {
+		return rc, errors.Wrap(err, "cannot build bandwith limit config")
+	}
+
+	placeholderEncryption, err := endpoint.PlaceholderCreationEncryptionPropertyString(recvOpts.Placeholder.Encryption)
+	if err != nil {
+		options := []string{}
+		for _, v := range endpoint.PlaceholderCreationEncryptionPropertyValues() {
+			options = append(options, endpoint.PlaceholderCreationEncryptionProperty(v).String())
+		}
+		return rc, errors.Errorf("placeholder encryption value %q is invalid, must be one of %s",
+			recvOpts.Placeholder.Encryption, options)
+	}
+
+	rc = endpoint.ReceiverConfig{
+		JobID:                      jobID,
+		RootWithoutClientComponent: rootFs,
+		AppendClientIdentity:       in.GetAppendClientIdentity(),
+
+		InheritProperties:  recvOpts.Properties.Inherit,
+		OverrideProperties: recvOpts.Properties.Override,
+
+		BandwidthLimit: bwlim,
+
+		PlaceholderEncryption: placeholderEncryption,
+	}
+	if err := rc.Validate(); err != nil {
+		return rc, errors.Wrap(err, "cannot build receiver config")
+	}
+
+	return rc, nil
+}

daemon/job/build_jobs_test.go

@@ -2,8 +2,11 @@ package job
 
 import (
 	"fmt"
+	"path"
+	"path/filepath"
 	"testing"
 
+	"github.com/kr/pretty"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 
@@ -92,7 +95,7 @@ jobs:
 			conf, err := config.ParseConfigBytes([]byte(fill(c.jobName)))
 			require.NoError(t, err, "not expecting yaml-config to know about job ids")
 			require.NotNil(t, conf)
-			jobs, err := JobsFromConfig(conf)
+			jobs, err := JobsFromConfig(conf, config.ParseFlagsNone)
 
 			if c.valid {
 				assert.NoError(t, err)
@@ -108,3 +111,208 @@ jobs:
 	}
 
 }
+
+func TestSampleConfigsAreBuiltWithoutErrors(t *testing.T) {
+	paths, err := filepath.Glob("../../config/samples/*")
+	if err != nil {
+		t.Errorf("glob failed: %+v", err)
+	}
+
+	type additionalCheck struct {
+		state int
+		test  func(t *testing.T, jobs []Job)
+	}
+	additionalChecks := map[string]*additionalCheck{
+		"bandwidth_limit.yml": {test: testSampleConfig_BandwidthLimit},
+	}
+
+	for _, p := range paths {
+
+		if path.Ext(p) != ".yml" {
+			t.Logf("skipping file %s", p)
+			continue
+		}
+
+		filename := path.Base(p)
+		t.Logf("checking for presence additonal checks for file %q", filename)
+		additionalCheck := additionalChecks[filename]
+		if additionalCheck == nil {
+			t.Logf("no additional checks")
+		} else {
+			t.Logf("additional check present")
+			additionalCheck.state = 1
+		}
+
+		t.Run(p, func(t *testing.T) {
+			c, err := config.ParseConfig(p)
+			if err != nil {
+				t.Fatalf("error parsing %s:\n%+v", p, err)
+			}
+
+			t.Logf("file: %s", p)
+			t.Log(pretty.Sprint(c))
+
+			jobs, err := JobsFromConfig(c, config.ParseFlagsNoCertCheck)
+			t.Logf("jobs: %#v", jobs)
+			require.NoError(t, err)
+
+			if additionalCheck != nil {
+				additionalCheck.test(t, jobs)
+				additionalCheck.state = 2
+			}
+		})
+
+	}
+
+	for basename, c := range additionalChecks {
+		if c.state == 0 {
+			panic("univisited additional check " + basename)
+		}
+	}
+
+}
+
+func testSampleConfig_BandwidthLimit(t *testing.T, jobs []Job) {
+	require.Len(t, jobs, 3)
+
+	{
+		limitedSink, ok := jobs[0].(*PassiveSide)
+		require.True(t, ok, "%T", jobs[0])
+		limitedSinkMode, ok := limitedSink.mode.(*modeSink)
+		require.True(t, ok, "%T", limitedSink)
+
+		assert.Equal(t, int64(12345), limitedSinkMode.receiverConfig.BandwidthLimit.Max)
+		assert.Equal(t, int64(1<<17), limitedSinkMode.receiverConfig.BandwidthLimit.BucketCapacity)
+	}
+
+	{
+		limitedPush, ok := jobs[1].(*ActiveSide)
+		require.True(t, ok, "%T", jobs[1])
+		limitedPushMode, ok := limitedPush.mode.(*modePush)
+		require.True(t, ok, "%T", limitedPush)
+
+		assert.Equal(t, int64(54321), limitedPushMode.senderConfig.BandwidthLimit.Max)
+		assert.Equal(t, int64(1024), limitedPushMode.senderConfig.BandwidthLimit.BucketCapacity)
+	}
+
+	{
+		unlimitedSink, ok := jobs[2].(*PassiveSide)
+		require.True(t, ok, "%T", jobs[2])
+		unlimitedSinkMode, ok := unlimitedSink.mode.(*modeSink)
+		require.True(t, ok, "%T", unlimitedSink)
+
+		max := unlimitedSinkMode.receiverConfig.BandwidthLimit.Max
+		assert.Less(t, max, int64(0), max, "unlimited mode <=> negative value for .Max, see bandwidthlimit.Config")
+	}
+
+}
+
+func TestReplicationOptions(t *testing.T) {
+	tmpl := `
+jobs:
+- name: foo
+  type: push
+  connect:
+    type: local
+    listener_name: foo
+    client_identity: bar
+  filesystems: {"<": true}
+  %s
+  snapshotting:
+    type: manual
+  pruning:
+    keep_sender:
+    - type: last_n
+      count: 10
+    keep_receiver:
+    - type: last_n
+      count: 10
+`
+
+	type Test struct {
+		name        string
+		input       string
+		expectOk    func(t *testing.T, a *ActiveSide, m *modePush)
+		expectError bool
+	}
+
+	tests := []Test{
+		{
+			name: "defaults",
+			input: `
+  replication: {}
+`,
+			expectOk: func(t *testing.T, a *ActiveSide, m *modePush) {},
+		},
+		{
+			name: "steps_zero",
+			input: `
+  replication:
+    concurrency:
+      steps: 0
+`,
+			expectError: true,
+		},
+		{
+			name: "size_estimates_zero",
+			input: `
+  replication:
+    concurrency:
+      size_estimates: 0
+`,
+			expectError: true,
+		},
+		{
+			name: "custom_values",
+			input: `
+  replication:
+    concurrency:
+      steps: 23
+      size_estimates: 42
+`,
+			expectOk: func(t *testing.T, a *ActiveSide, m *modePush) {
+				assert.Equal(t, 23, a.replicationDriverConfig.StepQueueConcurrency)
+				assert.Equal(t, 42, m.plannerPolicy.SizeEstimationConcurrency)
+			},
+		},
+		{
+			name: "negative_values_forbidden",
+			input: `
+  replication:
+    concurrency:
+      steps: -23
+      size_estimates: -42
+`,
+			expectError: true,
+		},
+	}
+
+	fill := func(s string) string { return fmt.Sprintf(tmpl, s) }
+
+	for _, ts := range tests {
+		t.Run(ts.name, func(t *testing.T) {
+			assert.True(t, (ts.expectError) != (ts.expectOk != nil))
+
+			cstr := fill(ts.input)
+			t.Logf("testing config:\n%s", cstr)
+			c, err := config.ParseConfigBytes([]byte(cstr))
+			require.NoError(t, err)
+			jobs, err := JobsFromConfig(c, config.ParseFlagsNone)
+			if ts.expectOk != nil {
+				require.NoError(t, err)
+				require.NotNil(t, c)
+				require.NoError(t, err)
+				require.Len(t, jobs, 1)
+				a := jobs[0].(*ActiveSide)
+				m := a.mode.(*modePush)
+				ts.expectOk(t, a, m)
+			} else if ts.expectError {
+				require.Error(t, err)
+			} else {
+				t.Fatalf("test must define expectOk or expectError")
+			}
+
+		})
+	}
+
+}

daemon/job/job.go

@@ -7,6 +7,7 @@ import (
 
 	"github.com/prometheus/client_golang/prometheus"
 
+	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/endpoint"
 	"github.com/zrepl/zrepl/logger"
 	"github.com/zrepl/zrepl/zfs"
@@ -14,21 +15,8 @@ import (
 
 type Logger = logger.Logger
 
-type contextKey int
-
-const (
-	contextKeyLog contextKey = iota
-)
-
 func GetLogger(ctx context.Context) Logger {
-	if l, ok := ctx.Value(contextKeyLog).(Logger); ok {
+	return logging.GetLogger(ctx, logging.SubsysJob)
-		return l
-	}
-	return logger.NewNullLogger()
-}
-
-func WithLogger(ctx context.Context, l Logger) context.Context {
-	return context.WithValue(ctx, contextKeyLog, l)
 }
 
 type Job interface {

daemon/job/passive.go

@@ -7,8 +7,9 @@ import (
 	"github.com/pkg/errors"
 	"github.com/prometheus/client_golang/prometheus"
 
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+
 	"github.com/zrepl/zrepl/config"
-	"github.com/zrepl/zrepl/daemon/filters"
 	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/daemon/snapper"
 	"github.com/zrepl/zrepl/endpoint"
@@ -47,19 +48,9 @@ func (m *modeSink) SnapperReport() *snapper.Report { return nil }
 func modeSinkFromConfig(g *config.Global, in *config.SinkJob, jobID endpoint.JobID) (m *modeSink, err error) {
 	m = &modeSink{}
 
-	rootDataset, err := zfs.NewDatasetPath(in.RootFS)
+	m.receiverConfig, err = buildReceiverConfig(in, jobID)
 	if err != nil {
-		return nil, errors.New("root dataset is not a valid zfs filesystem path")
+		return nil, err
-	}
-
-	m.receiverConfig = endpoint.ReceiverConfig{
-		JobID:                      jobID,
-		RootWithoutClientComponent: rootDataset,
-		AppendClientIdentity:       true, // !
-		UpdateLastReceivedHold:     true,
-	}
-	if err := m.receiverConfig.Validate(); err != nil {
-		return nil, errors.Wrap(err, "cannot build receiver config")
 	}
 
 	return m, nil
@@ -67,23 +58,19 @@ func modeSinkFromConfig(g *config.Global, in *config.SinkJob, jobID endpoint.Job
 
 type modeSource struct {
 	senderConfig *endpoint.SenderConfig
-	snapper      *snapper.PeriodicOrManual
+	snapper      snapper.Snapper
 }
 
 func modeSourceFromConfig(g *config.Global, in *config.SourceJob, jobID endpoint.JobID) (m *modeSource, err error) {
 	// FIXME exact dedup of modePush
 	m = &modeSource{}
-	fsf, err := filters.DatasetMapFilterFromConfig(in.Filesystems)
+
+	m.senderConfig, err = buildSenderConfig(in, jobID)
 	if err != nil {
-		return nil, errors.Wrap(err, "cannot build filesystem filter")
+		return nil, errors.Wrap(err, "send options")
-	}
-	m.senderConfig = &endpoint.SenderConfig{
-		FSF:     fsf,
-		Encrypt: &zfs.NilBool{B: in.Send.Encrypted},
-		JobID:   jobID,
 	}
 
-	if m.snapper, err = snapper.FromConfig(g, fsf, in.Snapshotting); err != nil {
+	if m.snapper, err = snapper.FromConfig(g, m.senderConfig.FSF, in.Snapshotting); err != nil {
 		return nil, errors.Wrap(err, "cannot build snapper")
 	}
 
@@ -101,10 +88,11 @@ func (m *modeSource) RunPeriodic(ctx context.Context) {
 }
 
 func (m *modeSource) SnapperReport() *snapper.Report {
-	return m.snapper.Report()
+	r := m.snapper.Report()
+	return &r
 }
 
-func passiveSideFromConfig(g *config.Global, in *config.PassiveJob, configJob interface{}) (s *PassiveSide, err error) {
+func passiveSideFromConfig(g *config.Global, in *config.PassiveJob, configJob interface{}, parseFlags config.ParseFlags) (s *PassiveSide, err error) {
 
 	s = &PassiveSide{}
 
@@ -123,7 +111,7 @@ func passiveSideFromConfig(g *config.Global, in *config.PassiveJob, configJob in
 		return nil, err // no wrapping necessary
 	}
 
-	if s.listen, err = fromconfig.ListenerFactoryFromConfig(g, in.Serve); err != nil {
+	if s.listen, err = fromconfig.ListenerFactoryFromConfig(g, in.Serve, parseFlags); err != nil {
 		return nil, errors.Wrap(err, "cannot build listener factory")
 	}
 
@@ -164,12 +152,14 @@ func (j *PassiveSide) SenderConfig() *endpoint.SenderConfig {
 func (*PassiveSide) RegisterMetrics(registerer prometheus.Registerer) {}
 
 func (j *PassiveSide) Run(ctx context.Context) {
-
+	ctx, endTask := trace.WithTaskAndSpan(ctx, "passive-side-job", j.Name())
+	defer endTask()
 	log := GetLogger(ctx)
 	defer log.Info("job exiting")
-	ctx = logging.WithSubsystemLoggers(ctx, log)
 	{
-		ctx, cancel := context.WithCancel(ctx) // shadowing
+		ctx, endTask := trace.WithTask(ctx, "periodic") // shadowing
+		defer endTask()
+		ctx, cancel := context.WithCancel(ctx)
 		defer cancel()
 		go j.mode.RunPeriodic(ctx)
 	}
@@ -179,8 +169,14 @@ func (j *PassiveSide) Run(ctx context.Context) {
 		panic(fmt.Sprintf("implementation error: j.mode.Handler() returned nil: %#v", j))
 	}
 
-	ctxInterceptor := func(handlerCtx context.Context) context.Context {
+	ctxInterceptor := func(handlerCtx context.Context, info rpc.HandlerContextInterceptorData, handler func(ctx context.Context)) {
-		return logging.WithSubsystemLoggers(handlerCtx, log)
+		// the handlerCtx is clean => need to inherit logging and tracing config from job context
+		handlerCtx = logging.WithInherit(handlerCtx, ctx)
+		handlerCtx = trace.WithInherit(handlerCtx, ctx)
+
+		handlerCtx, endTask := trace.WithTaskAndSpan(handlerCtx, "handler", fmt.Sprintf("job=%q client=%q method=%q", j.Name(), info.ClientIdentity(), info.FullMethod()))
+		defer endTask()
+		handler(handlerCtx)
 	}
 
 	rpcLoggers := rpc.GetLoggersOrPanic(ctx) // WithSubsystemLoggers above

daemon/job/snapjob.go

@@ -2,15 +2,20 @@ package job
 
 import (
 	"context"
+	"fmt"
 	"sort"
+	"sync"
 
 	"github.com/pkg/errors"
 	"github.com/prometheus/client_golang/prometheus"
 
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+	"github.com/zrepl/zrepl/util/bandwidthlimit"
+	"github.com/zrepl/zrepl/util/nodefault"
+
 	"github.com/zrepl/zrepl/config"
 	"github.com/zrepl/zrepl/daemon/filters"
 	"github.com/zrepl/zrepl/daemon/job/wakeup"
-	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/daemon/pruner"
 	"github.com/zrepl/zrepl/daemon/snapper"
 	"github.com/zrepl/zrepl/endpoint"
@@ -21,13 +26,14 @@ import (
 type SnapJob struct {
 	name     endpoint.JobID
 	fsfilter zfs.DatasetFilter
-	snapper  *snapper.PeriodicOrManual
+	snapper  snapper.Snapper
 
 	prunerFactory *pruner.LocalPrunerFactory
 
 	promPruneSecs *prometheus.HistogramVec // labels: prune_side
 
-	pruner *pruner.Pruner
+	prunerMtx sync.Mutex
+	pruner    *pruner.Pruner
 }
 
 func (j *SnapJob) Name() string { return j.name.String() }
@@ -75,10 +81,13 @@ type SnapJobStatus struct {
 func (j *SnapJob) Status() *Status {
 	s := &SnapJobStatus{}
 	t := j.Type()
+	j.prunerMtx.Lock()
 	if j.pruner != nil {
 		s.Pruning = j.pruner.Report()
 	}
-	s.Snapshotting = j.snapper.Report()
+	j.prunerMtx.Unlock()
+	r := j.snapper.Report()
+	s.Snapshotting = &r
 	return &Status{Type: t, JobSpecific: s}
 }
 
@@ -89,15 +98,18 @@ func (j *SnapJob) OwnedDatasetSubtreeRoot() (rfs *zfs.DatasetPath, ok bool) {
 func (j *SnapJob) SenderConfig() *endpoint.SenderConfig { return nil }
 
 func (j *SnapJob) Run(ctx context.Context) {
+	ctx, endTask := trace.WithTaskAndSpan(ctx, "snap-job", j.Name())
+	defer endTask()
 	log := GetLogger(ctx)
-	ctx = logging.WithSubsystemLoggers(ctx, log)
 
 	defer log.Info("job exiting")
 
 	periodicDone := make(chan struct{})
 	ctx, cancel := context.WithCancel(ctx)
 	defer cancel()
-	go j.snapper.Run(ctx, periodicDone)
+	periodicCtx, endTask := trace.WithTask(ctx, "snapshotting")
+	defer endTask()
+	go j.snapper.Run(periodicCtx, periodicDone)
 
 	invocationCount := 0
 outer:
@@ -112,8 +124,10 @@ outer:
 		case <-periodicDone:
 		}
 		invocationCount++
-		invLog := log.WithField("invocation", invocationCount)
+
-		j.doPrune(WithLogger(ctx, invLog))
+		invocationCtx, endSpan := trace.WithSpan(ctx, fmt.Sprintf("invocation-%d", invocationCount))
+		j.doPrune(invocationCtx)
+		endSpan()
 	}
 }
 
@@ -124,7 +138,7 @@ outer:
 // TODO:
 // This is a work-around for the current package daemon/pruner
 // and package pruning.Snapshot limitation: they require the
-//  `Replicated` getter method be present, but obviously,
+// `Replicated` getter method be present, but obviously,
 // a local job like SnapJob can't deliver on that.
 // But the pruner.Pruner gives up on an FS if no replication
 // cursor is present, which is why this pruner returns the
@@ -134,7 +148,7 @@ type alwaysUpToDateReplicationCursorHistory struct {
 	target pruner.Target
 }
 
-var _ pruner.History = (*alwaysUpToDateReplicationCursorHistory)(nil)
+var _ pruner.Sender = (*alwaysUpToDateReplicationCursorHistory)(nil)
 
 func (h alwaysUpToDateReplicationCursorHistory) ReplicationCursor(ctx context.Context, req *pdu.ReplicationCursorReq) (*pdu.ReplicationCursorRes, error) {
 	fsvReq := &pdu.ListFilesystemVersionsReq{
@@ -161,15 +175,21 @@ func (h alwaysUpToDateReplicationCursorHistory) ListFilesystems(ctx context.Cont
 }
 
 func (j *SnapJob) doPrune(ctx context.Context) {
+	ctx, endSpan := trace.WithSpan(ctx, "snap-job-do-prune")
+	defer endSpan()
 	log := GetLogger(ctx)
-	ctx = logging.WithSubsystemLoggers(ctx, log)
 	sender := endpoint.NewSender(endpoint.SenderConfig{
 		JobID: j.name,
 		FSF:   j.fsfilter,
-		// FIXME encryption setting is irrelevant for SnapJob because the endpoint is only used as pruner.Target
+		// FIXME the following config fields are irrelevant for SnapJob
-		Encrypt: &zfs.NilBool{B: true},
+		// because the endpoint is only used as pruner.Target.
+		// However, the implementation requires them to be set.
+		Encrypt:        &nodefault.Bool{B: true},
+		BandwidthLimit: bandwidthlimit.NoLimitConfig(),
 	})
+	j.prunerMtx.Lock()
 	j.pruner = j.prunerFactory.BuildLocalPruner(ctx, sender, alwaysUpToDateReplicationCursorHistory{sender})
+	j.prunerMtx.Unlock()
 	log.Info("start pruning")
 	j.pruner.Prune()
 	log.Info("finished pruning")

daemon/logging/build_logging.go

@@ -11,18 +11,9 @@ import (
 	"github.com/pkg/errors"
 
 	"github.com/zrepl/zrepl/config"
-	"github.com/zrepl/zrepl/daemon/hooks"
+	"github.com/zrepl/zrepl/daemon/logging/trace"
-	"github.com/zrepl/zrepl/daemon/pruner"
-	"github.com/zrepl/zrepl/daemon/snapper"
-	"github.com/zrepl/zrepl/endpoint"
 	"github.com/zrepl/zrepl/logger"
-	"github.com/zrepl/zrepl/replication/driver"
-	"github.com/zrepl/zrepl/replication/logic"
-	"github.com/zrepl/zrepl/rpc"
-	"github.com/zrepl/zrepl/rpc/transportmux"
 	"github.com/zrepl/zrepl/tlsconf"
-	"github.com/zrepl/zrepl/transport"
-	"github.com/zrepl/zrepl/zfs/zfscmd"
 )
 
 func OutletsFromConfig(in config.LoggingOutletEnumList) (*logger.Outlets, error) {
@@ -70,6 +61,8 @@ func OutletsFromConfig(in config.LoggingOutletEnumList) (*logger.Outlets, error)
 type Subsystem string
 
 const (
+	SubsysMeta         Subsystem = "meta"
+	SubsysJob          Subsystem = "job"
 	SubsysReplication  Subsystem = "repl"
 	SubsysEndpoint     Subsystem = "endpoint"
 	SubsysPruning      Subsystem = "pruning"
@@ -81,30 +74,103 @@ const (
 	SubsysRPCControl   Subsystem = "rpc.ctrl"
 	SubsysRPCData      Subsystem = "rpc.data"
 	SubsysZFSCmd       Subsystem = "zfs.cmd"
+	SubsysTraceData    Subsystem = "trace.data"
+	SubsysPlatformtest Subsystem = "platformtest"
 )
 
-func WithSubsystemLoggers(ctx context.Context, log logger.Logger) context.Context {
+var AllSubsystems = []Subsystem{
-	ctx = logic.WithLogger(ctx, log.WithField(SubsysField, SubsysReplication))
+	SubsysMeta,
-	ctx = driver.WithLogger(ctx, log.WithField(SubsysField, SubsysReplication))
+	SubsysJob,
-	ctx = endpoint.WithLogger(ctx, log.WithField(SubsysField, SubsysEndpoint))
+	SubsysReplication,
-	ctx = pruner.WithLogger(ctx, log.WithField(SubsysField, SubsysPruning))
+	SubsysEndpoint,
-	ctx = snapper.WithLogger(ctx, log.WithField(SubsysField, SubsysSnapshot))
+	SubsysPruning,
-	ctx = hooks.WithLogger(ctx, log.WithField(SubsysField, SubsysHooks))
+	SubsysSnapshot,
-	ctx = transport.WithLogger(ctx, log.WithField(SubsysField, SubsysTransport))
+	SubsysHooks,
-	ctx = transportmux.WithLogger(ctx, log.WithField(SubsysField, SubsysTransportMux))
+	SubsysTransport,
-	ctx = zfscmd.WithLogger(ctx, log.WithField(SubsysField, SubsysZFSCmd))
+	SubsysTransportMux,
-	ctx = rpc.WithLoggers(ctx,
+	SubsysRPC,
-		rpc.Loggers{
+	SubsysRPCControl,
-			General: log.WithField(SubsysField, SubsysRPC),
+	SubsysRPCData,
-			Control: log.WithField(SubsysField, SubsysRPCControl),
+	SubsysZFSCmd,
-			Data:    log.WithField(SubsysField, SubsysRPCData),
+	SubsysTraceData,
-		},
+	SubsysPlatformtest,
-	)
-	return ctx
 }
 
-func LogSubsystem(log logger.Logger, subsys Subsystem) logger.Logger {
+type injectedField struct {
-	return log.ReplaceField(SubsysField, subsys)
+	field  string
+	value  interface{}
+	parent *injectedField
+}
+
+func WithInjectedField(ctx context.Context, field string, value interface{}) context.Context {
+	var parent *injectedField
+	parentI := ctx.Value(contextKeyInjectedField)
+	if parentI != nil {
+		parent = parentI.(*injectedField)
+	}
+	// TODO sanity-check `field` now
+	this := &injectedField{field, value, parent}
+	return context.WithValue(ctx, contextKeyInjectedField, this)
+}
+
+func iterInjectedFields(ctx context.Context, cb func(field string, value interface{})) {
+	injI := ctx.Value(contextKeyInjectedField)
+	if injI == nil {
+		return
+	}
+	inj := injI.(*injectedField)
+	for ; inj != nil; inj = inj.parent {
+		cb(inj.field, inj.value)
+	}
+}
+
+type SubsystemLoggers map[Subsystem]logger.Logger
+
+func SubsystemLoggersWithUniversalLogger(l logger.Logger) SubsystemLoggers {
+	loggers := make(SubsystemLoggers)
+	for _, s := range AllSubsystems {
+		loggers[s] = l
+	}
+	return loggers
+}
+
+func WithLoggers(ctx context.Context, loggers SubsystemLoggers) context.Context {
+	return context.WithValue(ctx, contextKeyLoggers, loggers)
+}
+
+func GetLoggers(ctx context.Context) SubsystemLoggers {
+	loggers, ok := ctx.Value(contextKeyLoggers).(SubsystemLoggers)
+	if !ok {
+		return nil
+	}
+	return loggers
+}
+
+func GetLogger(ctx context.Context, subsys Subsystem) logger.Logger {
+	return getLoggerImpl(ctx, subsys, true)
+}
+
+func getLoggerImpl(ctx context.Context, subsys Subsystem, panicIfEnded bool) logger.Logger {
+	loggers, ok := ctx.Value(contextKeyLoggers).(SubsystemLoggers)
+	if !ok || loggers == nil {
+		return logger.NewNullLogger()
+	}
+	l, ok := loggers[subsys]
+	if !ok {
+		return logger.NewNullLogger()
+	}
+
+	l = l.WithField(SubsysField, subsys)
+
+	l = l.WithField(SpanField, trace.GetSpanStackOrDefault(ctx, *trace.StackKindId, "NOSPAN"))
+
+	fields := make(logger.Fields)
+	iterInjectedFields(ctx, func(field string, value interface{}) {
+		fields[field] = value
+	})
+	l = l.WithFields(fields)
+
+	return l
 }
 
 func parseLogFormat(i interface{}) (f EntryFormatter, err error) {

daemon/logging/logging_context.go

@@ -0,0 +1,24 @@
+package logging
+
+import "context"
+
+type contextKey int
+
+const (
+	contextKeyLoggers contextKey = 1 + iota
+	contextKeyInjectedField
+)
+
+var contextKeys = []contextKey{
+	contextKeyLoggers,
+	contextKeyInjectedField,
+}
+
+func WithInherit(ctx, inheritFrom context.Context) context.Context {
+	for _, k := range contextKeys {
+		if v := inheritFrom.Value(k); v != nil {
+			ctx = context.WithValue(ctx, k, v) // no shadow
+		}
+	}
+	return ctx
+}

daemon/logging/logging_formatters.go

@@ -22,6 +22,7 @@ const (
 const (
 	JobField    string = "job"
 	SubsysField string = "subsystem"
+	SpanField   string = "span"
 )
 
 type MetadataFlags int64
@@ -85,7 +86,7 @@ func (f *HumanFormatter) Format(e *logger.Entry) (out []byte, err error) {
 		fmt.Fprintf(&line, "[%s]", col.Sprint(e.Level.Short()))
 	}
 
-	prefixFields := []string{JobField, SubsysField}
+	prefixFields := []string{JobField, SubsysField, SpanField}
 	prefixed := make(map[string]bool, len(prefixFields)+2)
 	for _, field := range prefixFields {
 		val, ok := e.Fields[field]
@@ -174,8 +175,8 @@ func (f *LogfmtFormatter) Format(e *logger.Entry) ([]byte, error) {
 	}
 
 	// at least try and put job and task in front
-	prefixed := make(map[string]bool, 2)
+	prefixed := make(map[string]bool, 3)
-	prefix := []string{JobField, SubsysField}
+	prefix := []string{JobField, SubsysField, SpanField}
 	for _, pf := range prefix {
 		v, ok := e.Fields[pf]
 		if !ok {

daemon/logging/trace/trace.go

@@ -0,0 +1,422 @@
+// package trace provides activity tracing via ctx through Tasks and Spans
+//
+// # Basic Concepts
+//
+// Tracing can be used to identify where a piece of code spends its time.
+//
+// The Go standard library provides package runtime/trace which is useful to identify CPU bottlenecks or
+// to understand what happens inside the Go runtime.
+// However, it is not ideal for application level tracing, in particular if those traces should be understandable
+// to tech-savvy users (albeit not developers).
+//
+// This package provides the concept of Tasks and Spans to express what activity is happening within an application:
+// - Neither task nor span is really tangible but instead contained within the context.Context tree
+// - Tasks represent concurrent activity (i.e. goroutines).
+// - Spans represent a semantic stack trace within a task.
+// As a consequence, whenever a context is propagated across goroutine boundary, you need to create a child task:
+//
+//	go func(ctx context.Context) {
+//	  ctx, endTask = WithTask(ctx, "what-happens-inside-the-child-task")
+//	  defer endTask()
+//	  // ...
+//	}(ctx)
+//
+// Within the task, you can open up a hierarchy of spans.
+// In contrast to tasks, which have can multiple concurrently running child tasks,
+// spans must nest and not cross the goroutine boundary.
+//
+//	ctx, endSpan = WithSpan(ctx, "copy-dir")
+//	defer endSpan()
+//	for _, f := range dir.Files() {
+//	  func() {
+//	    ctx, endSpan := WithSpan(ctx, fmt.Sprintf("copy-file %q", f))
+//	    defer endspan()
+//	    b, _ := ioutil.ReadFile(f)
+//	    _ = ioutil.WriteFile(f + ".copy", b, 0600)
+//	  }()
+//	}
+//
+// In combination:
+//
+//	ctx, endTask = WithTask(ctx, "copy-dirs")
+//	defer endTask()
+//	for i := range dirs {
+//	  go func(dir string) {
+//	    ctx, endTask := WithTask(ctx, "copy-dir")
+//	    defer endTask()
+//	    for _, f := range filesIn(dir) {
+//	      func() {
+//	        ctx, endSpan := WithSpan(ctx, fmt.Sprintf("copy-file %q", f))
+//	        defer endspan()
+//	        b, _ := ioutil.ReadFile(f)
+//	        _ = ioutil.WriteFile(f + ".copy", b, 0600)
+//	      }()
+//	    }
+//	  }()
+//	}
+//
+// Note that a span ends at the time you call endSpan - not before and not after that.
+// If you violate the stack-like nesting of spans by forgetting an endSpan() invocation,
+// the out-of-order endSpan() will panic.
+//
+// A similar rule applies to the endTask closure returned by WithTask:
+// If a task has live child tasks at the time you call endTask(), the call will panic.
+//
+// Recovering from endSpan() or endTask() panics will corrupt the trace stack and lead to corrupt tracefile output.
+//
+// # Best Practices For Naming Tasks And Spans
+//
+// Tasks should always have string constants as names, and must not contain the `#` character. WHy?
+// First, the visualization by chrome://tracing draws a horizontal bar for each task in the trace.
+// Also, the package appends `#NUM` for each concurrently running instance of a task name.
+// Note that the `#NUM` suffix will be reused if a task has ended, in order to avoid an
+// infinite number of horizontal bars in the visualization.
+//
+// # Chrome-compatible Tracefile Support
+//
+// The activity trace generated by usage of WithTask and WithSpan can be rendered to a JSON output file
+// that can be loaded into chrome://tracing .
+// Apart from function GetSpanStackOrDefault, this is the main benefit of this package.
+//
+// First, there is a convenience environment variable 'ZREPL_ACTIVITY_TRACE' that can be set to an output path.
+// From process start onward, a trace is written to that path.
+//
+// More consumers can attach to the activity trace through the ChrometraceClientWebsocketHandler websocket handler.
+//
+// If a write error is encountered with any consumer (including the env-var based one), the consumer is closed and
+// will not receive further trace output.
+package trace
+
+import (
+	"context"
+	"fmt"
+	"os"
+	runtimedebug "runtime/debug"
+	"strings"
+	"time"
+
+	"github.com/kr/pretty"
+	"github.com/pkg/errors"
+	"github.com/prometheus/client_golang/prometheus"
+
+	"github.com/zrepl/zrepl/util/chainlock"
+)
+
+var metrics struct {
+	activeTasks prometheus.Gauge
+}
+var taskNamer *uniqueConcurrentTaskNamer = newUniqueTaskNamer()
+
+func init() {
+	metrics.activeTasks = prometheus.NewGauge(prometheus.GaugeOpts{
+		Namespace: "zrepl",
+		Subsystem: "trace",
+		Name:      "active_tasks",
+		Help:      "number of active (tracing-level) tasks in the daemon",
+	})
+}
+
+func RegisterMetrics(r prometheus.Registerer) {
+	r.MustRegister(metrics.activeTasks)
+}
+
+type traceNode struct {
+	// NOTE: members with prefix debug are only valid if the package variable debugEnabled is true.
+
+	id                 string
+	annotation         string
+	parentTask         *traceNode
+	debugCreationStack string // debug: stack trace of when the traceNode was created
+
+	mtx chainlock.L
+
+	activeChildTasks      int32               // only for task nodes, insignificant for span nodes
+	debugActiveChildTasks map[*traceNode]bool // debug: set of active child tasks
+	parentSpan            *traceNode
+	activeChildSpan       *traceNode // nil if task or span doesn't have an active child span
+
+	startedAt time.Time
+	endedAt   time.Time
+}
+
+func (s *traceNode) StartedAt() time.Time { return s.startedAt }
+func (s *traceNode) EndedAt() time.Time   { return s.endedAt }
+
+// caller must hold mtx
+func (s *traceNode) debugString() string {
+	if !debugEnabled {
+		return "<debugEnabled=false>"
+	}
+	return pretty.Sprint(s)
+}
+
+// Returned from WithTask or WithSpan.
+// Must be called once the task or span ends.
+// See package-level docs for nesting rules.
+// Wrong call order / forgetting to call it will result in panics.
+type DoneFunc func()
+
+var ErrTaskStillHasActiveChildTasks = fmt.Errorf("end task: task still has active child tasks")
+
+// Start a new root task or create a child task of an existing task.
+//
+// This is required when starting a new goroutine and
+// passing an existing task context to it.
+//
+// taskName should be a constantand must not contain '#'
+//
+// The implementation ensures that,
+// if multiple tasks with the same name exist simultaneously,
+// a unique suffix is appended to uniquely identify the task opened with this function.
+func WithTask(ctx context.Context, taskName string) (context.Context, DoneFunc) {
+
+	var parentTask *traceNode
+	nodeI := ctx.Value(contextKeyTraceNode)
+	if nodeI != nil {
+		node := nodeI.(*traceNode)
+		if node.parentSpan != nil {
+			parentTask = node.parentTask
+		} else {
+			parentTask = node
+		}
+	}
+	// find the first ancestor that hasn't ended yet (nil if need be)
+	if parentTask != nil {
+		parentTask.mtx.Lock()
+	}
+	for parentTask != nil && !parentTask.endedAt.IsZero() {
+		thisParent := parentTask
+		parentTask = parentTask.parentTask
+		// lock parent first such that it isn't modified by other callers to thisParent
+		if parentTask != nil {
+			parentTask.mtx.Lock()
+		}
+		thisParent.mtx.Unlock()
+	}
+	// invariant: either parentTask != nil and we hold the lock on parentTask, or parentTask is nil
+
+	taskName, taskNameDone := taskNamer.UniqueConcurrentTaskName(taskName)
+
+	this := &traceNode{
+		id:               genID(),
+		annotation:       taskName,
+		parentTask:       parentTask,
+		activeChildTasks: 0,
+		parentSpan:       nil,
+		activeChildSpan:  nil,
+
+		startedAt: time.Now(),
+		endedAt:   time.Time{},
+	}
+
+	if debugEnabled {
+		this.debugCreationStack = string(runtimedebug.Stack())
+		this.debugActiveChildTasks = map[*traceNode]bool{}
+	}
+
+	if parentTask != nil {
+		this.parentTask.activeChildTasks++
+		if debugEnabled {
+			this.parentTask.debugActiveChildTasks[this] = true
+		}
+		parentTask.mtx.Unlock()
+	}
+
+	ctx = context.WithValue(ctx, contextKeyTraceNode, this)
+
+	chrometraceBeginTask(this)
+
+	metrics.activeTasks.Inc()
+
+	endTaskFunc := func() {
+
+		// only hold locks while manipulating the tree
+		// (trace writer might block too long and unlike spans, tasks are updated concurrently)
+		alreadyEnded := func() (alreadyEnded bool) {
+			if this.parentTask != nil {
+				defer this.parentTask.mtx.Lock().Unlock()
+			}
+			defer this.mtx.Lock().Unlock()
+
+			if this.activeChildTasks != 0 {
+				if debugEnabled {
+					// the debugString can be quite long and panic won't print it completely
+					fmt.Fprintf(os.Stderr, "going to panic due to activeChildTasks:\n%s\n", this.debugString())
+				}
+				panic(errors.WithMessagef(ErrTaskStillHasActiveChildTasks, "end task: %v active child tasks (run daemon with env var %s=1 for more details)\n", this.activeChildTasks, debugEnabledEnvVar))
+			}
+
+			// support idempotent task ends
+			if !this.endedAt.IsZero() {
+				return true
+			}
+			this.endedAt = time.Now()
+
+			if this.parentTask != nil {
+				this.parentTask.activeChildTasks--
+				if debugEnabled {
+					delete(this.parentTask.debugActiveChildTasks, this)
+				}
+				if this.parentTask.activeChildTasks < 0 {
+					if debugEnabled {
+						// the debugString can be quite long and panic won't print it completely
+						fmt.Fprintf(os.Stderr, "going to panic due to activeChildTasks < 0:\n%s\n", this.parentTask.debugString())
+					}
+					panic(fmt.Sprintf("impl error: parent task with negative activeChildTasks count: %v", this.parentTask.activeChildTasks))
+				}
+			}
+			return false
+		}()
+		if alreadyEnded {
+			return
+		}
+
+		chrometraceEndTask(this)
+
+		metrics.activeTasks.Dec()
+
+		taskNameDone()
+	}
+
+	return ctx, endTaskFunc
+}
+
+var ErrAlreadyActiveChildSpan = fmt.Errorf("create child span: span already has an active child span")
+var ErrSpanStillHasActiveChildSpan = fmt.Errorf("end span: span still has active child spans")
+
+// Start a new span.
+// Important: ctx must have an active task (see WithTask)
+func WithSpan(ctx context.Context, annotation string) (context.Context, DoneFunc) {
+	var parentSpan, parentTask *traceNode
+	nodeI := ctx.Value(contextKeyTraceNode)
+	if nodeI != nil {
+		parentSpan = nodeI.(*traceNode)
+		if parentSpan.parentSpan == nil {
+			parentTask = parentSpan
+		} else {
+			parentTask = parentSpan.parentTask
+		}
+	} else {
+		panic("must be called from within a task")
+	}
+
+	this := &traceNode{
+		id:              genID(),
+		annotation:      annotation,
+		parentTask:      parentTask,
+		parentSpan:      parentSpan,
+		activeChildSpan: nil,
+
+		startedAt: time.Now(),
+		endedAt:   time.Time{},
+	}
+
+	if debugEnabled {
+		this.debugCreationStack = string(runtimedebug.Stack())
+	}
+
+	parentSpan.mtx.HoldWhile(func() {
+		if parentSpan.activeChildSpan != nil {
+			panic(ErrAlreadyActiveChildSpan)
+		}
+		parentSpan.activeChildSpan = this
+	})
+
+	ctx = context.WithValue(ctx, contextKeyTraceNode, this)
+	chrometraceBeginSpan(this)
+	callbackEndSpan := callbackBeginSpan(ctx)
+
+	endTaskFunc := func() {
+
+		defer parentSpan.mtx.Lock().Unlock()
+		if parentSpan.activeChildSpan != this && this.endedAt.IsZero() {
+			panic("impl error: activeChildSpan should not change while != nil because there can only be one")
+		}
+
+		defer this.mtx.Lock().Unlock()
+		if this.activeChildSpan != nil {
+			panic(ErrSpanStillHasActiveChildSpan)
+		}
+
+		if !this.endedAt.IsZero() {
+			return // support idempotent span ends
+		}
+
+		parentSpan.activeChildSpan = nil
+		this.endedAt = time.Now()
+
+		chrometraceEndSpan(this)
+		callbackEndSpan(this)
+	}
+
+	return ctx, endTaskFunc
+}
+
+type StackKind struct {
+	symbolizeTask func(t *traceNode) string
+	symbolizeSpan func(s *traceNode) string
+}
+
+var (
+	StackKindId = &StackKind{
+		symbolizeTask: func(t *traceNode) string { return t.id },
+		symbolizeSpan: func(s *traceNode) string { return s.id },
+	}
+	SpanStackKindCombined = &StackKind{
+		symbolizeTask: func(t *traceNode) string { return fmt.Sprintf("(%s %q)", t.id, t.annotation) },
+		symbolizeSpan: func(s *traceNode) string { return fmt.Sprintf("(%s %q)", s.id, s.annotation) },
+	}
+	SpanStackKindAnnotation = &StackKind{
+		symbolizeTask: func(t *traceNode) string { return t.annotation },
+		symbolizeSpan: func(s *traceNode) string { return s.annotation },
+	}
+)
+
+func (n *traceNode) task() *traceNode {
+	task := n.parentTask
+	if n.parentSpan == nil {
+		task = n
+	}
+	return task
+}
+
+func (n *traceNode) TaskName() string {
+	task := n.task()
+	return task.annotation
+}
+
+func (this *traceNode) TaskAndSpanStack(kind *StackKind) (spanIdStack string) {
+	task := this.task()
+
+	var spansInTask []*traceNode
+	for s := this; s != nil; s = s.parentSpan {
+		spansInTask = append(spansInTask, s)
+	}
+
+	var tasks []*traceNode
+	for t := task; t != nil; t = t.parentTask {
+		tasks = append(tasks, t)
+	}
+
+	var taskIdsRev []string
+	for i := len(tasks) - 1; i >= 0; i-- {
+		taskIdsRev = append(taskIdsRev, kind.symbolizeTask(tasks[i]))
+	}
+
+	var spanIdsRev []string
+	for i := len(spansInTask) - 1; i >= 0; i-- {
+		spanIdsRev = append(spanIdsRev, kind.symbolizeSpan(spansInTask[i]))
+	}
+
+	taskStack := strings.Join(taskIdsRev, "$")
+	return fmt.Sprintf("%s$%s", taskStack, strings.Join(spanIdsRev, "."))
+}
+
+func GetSpanStackOrDefault(ctx context.Context, kind StackKind, def string) string {
+	if nI := ctx.Value(contextKeyTraceNode); nI != nil {
+		n := nI.(*traceNode)
+		return n.TaskAndSpanStack(StackKindId)
+	} else {
+		return def
+	}
+}

daemon/logging/trace/trace_callback.go

@@ -0,0 +1,54 @@
+package trace
+
+import (
+	"context"
+	"time"
+
+	"github.com/zrepl/zrepl/util/chainlock"
+)
+
+type SpanInfo interface {
+	StartedAt() time.Time
+	EndedAt() time.Time
+	TaskAndSpanStack(kind *StackKind) string
+}
+
+type Callback struct {
+	OnBegin func(ctx context.Context)
+	OnEnd   func(ctx context.Context, spanInfo SpanInfo)
+}
+
+var callbacks struct {
+	mtx chainlock.L
+	cs  []Callback
+}
+
+func RegisterCallback(c Callback) {
+	callbacks.mtx.HoldWhile(func() {
+		callbacks.cs = append(callbacks.cs, c)
+	})
+}
+
+func callbackBeginSpan(ctx context.Context) func(SpanInfo) {
+	// capture the current state of callbacks into a local variable
+	// this is safe because the slice is append-only and immutable
+
+	// (it is important that a callback registered _after_ callbackBeginSpin is called does not get called on OnEnd)
+
+	var cbs []Callback
+	callbacks.mtx.HoldWhile(func() {
+		cbs = callbacks.cs
+	})
+	for _, cb := range cbs {
+		if cb.OnBegin != nil {
+			cb.OnBegin(ctx)
+		}
+	}
+	return func(spanInfo SpanInfo) {
+		for _, cb := range cbs {
+			if cb.OnEnd != nil {
+				cb.OnEnd(ctx, spanInfo)
+			}
+		}
+	}
+}

daemon/logging/trace/trace_chrometrace.go

@@ -0,0 +1,230 @@
+package trace
+
+// The functions in this file are concerned with the generation
+// of trace files based on the information from WithTask and WithSpan.
+//
+// The emitted trace files are open-ended array of JSON objects
+// that follow the Chrome trace file format:
+//   https://docs.google.com/document/d/1CvAClvFfyA5R-PhYUmn5OOQtYMH4h6I0nSsKchNAySU/preview
+//
+// The emitted JSON can be loaded into Chrome's chrome://tracing view.
+//
+// The trace file can be written to a file whose path is specified in an env file,
+// and be written to web sockets established on ChrometraceHttpHandler
+
+import (
+	"bufio"
+	"bytes"
+	"encoding/json"
+	"fmt"
+	"io"
+	"os"
+	"sync"
+	"sync/atomic"
+
+	"golang.org/x/net/websocket"
+
+	"github.com/zrepl/zrepl/util/envconst"
+)
+
+var chrometracePID string
+
+func init() {
+	var err error
+	chrometracePID, err = os.Hostname()
+	if err != nil {
+		panic(err)
+	}
+	chrometracePID = fmt.Sprintf("%q", chrometracePID)
+}
+
+type chrometraceEvent struct {
+	Cat                       string   `json:"cat,omitempty"`
+	Name                      string   `json:"name"`
+	Stack                     []string `json:"stack,omitempty"`
+	Phase                     string   `json:"ph"`
+	TimestampUnixMicroseconds int64    `json:"ts"`
+	DurationMicroseconds      int64    `json:"dur,omitempty"`
+	Pid                       string   `json:"pid"`
+	Tid                       string   `json:"tid"`
+	Id                        string   `json:"id,omitempty"`
+}
+
+func chrometraceBeginSpan(s *traceNode) {
+	taskName := s.TaskName()
+	chrometraceWrite(chrometraceEvent{
+		Name:                      s.annotation,
+		Phase:                     "B",
+		TimestampUnixMicroseconds: s.startedAt.UnixNano() / 1000,
+		Pid:                       chrometracePID,
+		Tid:                       taskName,
+	})
+}
+
+func chrometraceEndSpan(s *traceNode) {
+	taskName := s.TaskName()
+	chrometraceWrite(chrometraceEvent{
+		Name:                      s.annotation,
+		Phase:                     "E",
+		TimestampUnixMicroseconds: s.endedAt.UnixNano() / 1000,
+		Pid:                       chrometracePID,
+		Tid:                       taskName,
+	})
+}
+
+var chrometraceFlowId uint64
+
+func chrometraceBeginTask(s *traceNode) {
+	chrometraceBeginSpan(s)
+
+	if s.parentTask == nil {
+		return
+	}
+	// beginning of a task that has a parent
+	// => use flow events to link parent and child
+
+	flowId := atomic.AddUint64(&chrometraceFlowId, 1)
+	flowIdStr := fmt.Sprintf("%x", flowId)
+
+	parentTask := s.parentTask.TaskName()
+	chrometraceWrite(chrometraceEvent{
+		Cat:                       "task", // seems to be necessary, otherwise the GUI shows some `indexOf` JS error
+		Name:                      "child-task",
+		Phase:                     "s",
+		TimestampUnixMicroseconds: s.startedAt.UnixNano() / 1000, // yes, the child's timestamp (=> from-point of the flow line is at right x-position of parent's bar)
+		Pid:                       chrometracePID,
+		Tid:                       parentTask,
+		Id:                        flowIdStr,
+	})
+
+	childTask := s.TaskName()
+	if parentTask == childTask {
+		panic(parentTask)
+	}
+	chrometraceWrite(chrometraceEvent{
+		Cat:                       "task", // seems to be necessary, otherwise the GUI shows some `indexOf` JS error
+		Name:                      "child-task",
+		Phase:                     "f",
+		TimestampUnixMicroseconds: s.startedAt.UnixNano() / 1000,
+		Pid:                       chrometracePID,
+		Tid:                       childTask,
+		Id:                        flowIdStr,
+	})
+}
+
+func chrometraceEndTask(s *traceNode) {
+	chrometraceEndSpan(s)
+}
+
+type chrometraceConsumerRegistration struct {
+	w io.Writer
+	// errored must have capacity 1, the writer thread will send to it non-blocking, then close it
+	errored chan error
+}
+
+var chrometraceConsumers struct {
+	register  chan chrometraceConsumerRegistration
+	consumers map[chrometraceConsumerRegistration]bool
+	write     chan []byte
+}
+
+func init() {
+	chrometraceConsumers.register = make(chan chrometraceConsumerRegistration)
+	chrometraceConsumers.consumers = make(map[chrometraceConsumerRegistration]bool)
+	chrometraceConsumers.write = make(chan []byte)
+	go func() {
+		kickConsumer := func(c chrometraceConsumerRegistration, err error) {
+			debug("chrometrace kicking consumer %#v after error %v", c, err)
+			select {
+			case c.errored <- err:
+			default:
+			}
+			close(c.errored)
+			delete(chrometraceConsumers.consumers, c)
+		}
+		for {
+			select {
+			case reg := <-chrometraceConsumers.register:
+				debug("registered chrometrace consumer %#v", reg)
+				chrometraceConsumers.consumers[reg] = true
+				n, err := reg.w.Write([]byte("[\n"))
+				if err != nil {
+					kickConsumer(reg, err)
+				} else if n != 2 {
+					kickConsumer(reg, fmt.Errorf("short write: %v", n))
+				}
+				// successfully registered
+
+			case buf := <-chrometraceConsumers.write:
+				debug("chrometrace write request: %s", string(buf))
+				var r bytes.Reader
+				for c := range chrometraceConsumers.consumers {
+					r.Reset(buf)
+					n, err := io.Copy(c.w, &r)
+					debug("chrometrace wrote n=%v bytes to consumer %#v", n, c)
+					if err != nil {
+						kickConsumer(c, err)
+					}
+				}
+			}
+		}
+	}()
+}
+
+func chrometraceWrite(i interface{}) {
+	var buf bytes.Buffer
+	err := json.NewEncoder(&buf).Encode(i)
+	if err != nil {
+		panic(err)
+	}
+	buf.WriteString(",")
+	chrometraceConsumers.write <- buf.Bytes()
+}
+
+func ChrometraceClientWebsocketHandler(conn *websocket.Conn) {
+	defer conn.Close()
+
+	var wg sync.WaitGroup
+	defer wg.Wait()
+	wg.Add(1)
+	go func() {
+		defer wg.Done()
+		r := bufio.NewReader(conn)
+		_, _, _ = r.ReadLine() // ignore errors
+		conn.Close()
+	}()
+
+	errored := make(chan error, 1)
+	chrometraceConsumers.register <- chrometraceConsumerRegistration{
+		w:       conn,
+		errored: errored,
+	}
+
+	wg.Add(1)
+	go func() {
+		defer wg.Done()
+		<-errored
+		conn.Close()
+	}()
+}
+
+var chrometraceFileConsumerPath = envconst.String("ZREPL_ACTIVITY_TRACE", "")
+
+func init() {
+	if chrometraceFileConsumerPath != "" {
+		var err error
+		f, err := os.Create(chrometraceFileConsumerPath)
+		if err != nil {
+			panic(err)
+		}
+		errored := make(chan error, 1)
+		chrometraceConsumers.register <- chrometraceConsumerRegistration{
+			w:       f,
+			errored: errored,
+		}
+		go func() {
+			<-errored
+			f.Close()
+		}()
+	}
+}

daemon/logging/trace/trace_context.go

@@ -0,0 +1,27 @@
+package trace
+
+import "context"
+
+type contextKey int
+
+const (
+	contextKeyTraceNode contextKey = 1 + iota
+)
+
+var contextKeys = []contextKey{
+	contextKeyTraceNode,
+}
+
+// WithInherit inherits the task hierarchy from inheritFrom into ctx.
+// The returned context is a child of ctx, but its task and span are those of inheritFrom.
+//
+// Note that in most use cases, callers most likely want to call WithTask since it will most likely
+// be in some sort of connection handler context.
+func WithInherit(ctx, inheritFrom context.Context) context.Context {
+	for _, k := range contextKeys {
+		if v := inheritFrom.Value(k); v != nil {
+			ctx = context.WithValue(ctx, k, v) // no shadow
+		}
+	}
+	return ctx
+}

daemon/logging/trace/trace_convenience.go

@@ -0,0 +1,76 @@
+package trace
+
+import (
+	"context"
+	"fmt"
+	"runtime"
+	"strings"
+	"sync"
+)
+
+// use like this:
+//
+//	defer WithSpanFromStackUpdateCtx(&existingCtx)()
+func WithSpanFromStackUpdateCtx(ctx *context.Context) DoneFunc {
+	childSpanCtx, end := WithSpan(*ctx, getMyCallerOrPanic())
+	*ctx = childSpanCtx
+	return end
+}
+
+// derive task name from call stack (caller's name)
+func WithTaskFromStack(ctx context.Context) (context.Context, DoneFunc) {
+	return WithTask(ctx, getMyCallerOrPanic())
+}
+
+// derive task name from call stack (caller's name) and update *ctx
+// to point to be the child task ctx
+func WithTaskFromStackUpdateCtx(ctx *context.Context) DoneFunc {
+	child, end := WithTask(*ctx, getMyCallerOrPanic())
+	*ctx = child
+	return end
+}
+
+// create a task and a span within it in one call
+func WithTaskAndSpan(ctx context.Context, task string, span string) (context.Context, DoneFunc) {
+	ctx, endTask := WithTask(ctx, task)
+	ctx, endSpan := WithSpan(ctx, fmt.Sprintf("%s %s", task, span))
+	return ctx, func() {
+		endSpan()
+		endTask()
+	}
+}
+
+// create a span during which several child tasks are spawned using the `add` function
+//
+// IMPORTANT FOR USERS: Caller must ensure that the capturing behavior is correct, the Go linter doesn't catch this.
+func WithTaskGroup(ctx context.Context, taskGroup string) (_ context.Context, add func(f func(context.Context)), waitEnd DoneFunc) {
+	var wg sync.WaitGroup
+	ctx, endSpan := WithSpan(ctx, taskGroup)
+	add = func(f func(context.Context)) {
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			ctx, endTask := WithTask(ctx, taskGroup)
+			defer endTask()
+			f(ctx)
+		}()
+	}
+	waitEnd = func() {
+		wg.Wait()
+		endSpan()
+	}
+	return ctx, add, waitEnd
+}
+
+func getMyCallerOrPanic() string {
+	pc, _, _, ok := runtime.Caller(2)
+	if !ok {
+		panic("cannot get caller")
+	}
+	details := runtime.FuncForPC(pc)
+	if ok && details != nil {
+		const prefix = "github.com/zrepl/zrepl"
+		return strings.TrimPrefix(strings.TrimPrefix(details.Name(), prefix), "/")
+	}
+	return ""
+}

daemon/logging/trace/trace_convenience_test.go

@@ -0,0 +1,62 @@
+package trace
+
+import (
+	"context"
+	"sync/atomic"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestGetCallerOrPanic(t *testing.T) {
+	withStackFromCtxMock := func() string {
+		return getMyCallerOrPanic()
+	}
+	ret := withStackFromCtxMock()
+	// zrepl prefix is stripped
+	assert.Equal(t, "daemon/logging/trace.TestGetCallerOrPanic", ret)
+}
+
+func TestWithTaskGroupRunTasksConcurrently(t *testing.T) {
+
+	// spawn a task group where each task waits for the other to start
+	// => without concurrency, they would hang
+
+	rootCtx, endRoot := WithTaskFromStack(context.Background())
+	defer endRoot()
+
+	_, add, waitEnd := WithTaskGroup(rootCtx, "test-task-group")
+
+	schedulerTimeout := 2 * time.Second
+	timeout := time.After(schedulerTimeout)
+	var hadTimeout uint32
+	started0, started1 := make(chan struct{}), make(chan struct{})
+	for i := 0; i < 2; i++ {
+		i := i // capture by copy
+		add(func(ctx context.Context) {
+			switch i {
+			case 0:
+				close(started0)
+				select {
+				case <-started1:
+				case <-timeout:
+					atomic.AddUint32(&hadTimeout, 1)
+				}
+			case 1:
+				close(started1)
+				select {
+				case <-started0:
+				case <-timeout:
+					atomic.AddUint32(&hadTimeout, 1)
+				}
+			default:
+				panic("unreachable")
+			}
+		})
+	}
+
+	waitEnd()
+	assert.Zero(t, hadTimeout, "either bad impl or scheduler timeout (which is %v)", schedulerTimeout)
+
+}

daemon/logging/trace/trace_debug.go

@@ -0,0 +1,19 @@
+package trace
+
+import (
+	"fmt"
+	"os"
+
+	"github.com/zrepl/zrepl/util/envconst"
+)
+
+const debugEnabledEnvVar = "ZREPL_TRACE_DEBUG_ENABLED"
+
+var debugEnabled = envconst.Bool(debugEnabledEnvVar, false)
+
+func debug(format string, args ...interface{}) {
+	if !debugEnabled {
+		return
+	}
+	fmt.Fprintf(os.Stderr, format+"\n", args...)
+}

daemon/logging/trace/trace_genID.go

@@ -0,0 +1,38 @@
+package trace
+
+import (
+	"encoding/base64"
+	"math/rand"
+	"strings"
+
+	"github.com/zrepl/zrepl/util/envconst"
+)
+
+var genIdNumBytes = envconst.Int("ZREPL_TRACE_ID_NUM_BYTES", 3)
+
+func init() {
+	if genIdNumBytes < 1 {
+		panic("trace node id byte length must be at least 1")
+	}
+}
+
+func genID() string {
+	var out strings.Builder
+	enc := base64.NewEncoder(base64.RawStdEncoding, &out)
+	buf := make([]byte, genIdNumBytes)
+	for i := 0; i < len(buf); {
+		n, err := rand.Read(buf[i:])
+		if err != nil {
+			panic(err)
+		}
+		i += n
+	}
+	n, err := enc.Write(buf[:])
+	if err != nil || n != len(buf) {
+		panic(err)
+	}
+	if err := enc.Close(); err != nil {
+		panic(err)
+	}
+	return out.String()
+}

daemon/logging/trace/trace_test.go

@@ -0,0 +1,205 @@
+package trace
+
+import (
+	"context"
+	"fmt"
+	"testing"
+
+	"github.com/gitchander/permutation"
+	"github.com/pkg/errors"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+func TestRegularSpanUsage(t *testing.T) {
+	root, endRoot := WithTask(context.Background(), "root")
+	defer endRoot()
+
+	s1, endS1 := WithSpan(root, "parent")
+	s2, endS2 := WithSpan(s1, "child")
+	_, endS3 := WithSpan(s2, "grand-child")
+	require.NotPanics(t, func() { endS3() })
+	require.NotPanics(t, func() { endS2() })
+
+	// reuse
+	_, endS4 := WithSpan(s1, "child-2")
+	require.NotPanics(t, func() { endS4() })
+
+	// close parent
+	require.NotPanics(t, func() { endS1() })
+}
+
+func TestMultipleActiveChildSpansNotAllowed(t *testing.T) {
+	root, endRoot := WithTask(context.Background(), "root")
+	defer endRoot()
+
+	s1, _ := WithSpan(root, "s1")
+	_, endS2 := WithSpan(s1, "s1-child1")
+
+	require.PanicsWithValue(t, ErrAlreadyActiveChildSpan, func() {
+		_, _ = WithSpan(s1, "s1-child2")
+	})
+
+	endS2()
+
+	require.NotPanics(t, func() {
+		_, _ = WithSpan(s1, "s1-child2")
+	})
+}
+
+func TestForkingChildSpansNotAllowed(t *testing.T) {
+	root, endRoot := WithTask(context.Background(), "root")
+	defer endRoot()
+
+	s1, _ := WithSpan(root, "s1")
+	sc, endSC := WithSpan(s1, "s1-child")
+	_, _ = WithSpan(sc, "s1-child-child")
+
+	require.PanicsWithValue(t, ErrSpanStillHasActiveChildSpan, func() {
+		endSC()
+	})
+}
+
+func TestRegularTaskUsage(t *testing.T) {
+	// assert concurrent activities on different tasks can end in any order
+	closeOrder := []int{0, 1, 2}
+	closeOrders := permutation.New(permutation.IntSlice(closeOrder))
+	for closeOrders.Next() {
+		t.Run(fmt.Sprintf("%v", closeOrder), func(t *testing.T) {
+			root, endRoot := WithTask(context.Background(), "root")
+			defer endRoot()
+
+			c1, endC1 := WithTask(root, "c1")
+			defer endC1()
+			c2, endC2 := WithTask(root, "c2")
+			defer endC2()
+
+			// begin 3 concurrent activities
+			_, endAR := WithSpan(root, "aR")
+			_, endAC1 := WithSpan(c1, "aC1")
+			_, endAC2 := WithSpan(c2, "aC2")
+
+			endFuncs := []DoneFunc{endAR, endAC1, endAC2}
+			for _, i := range closeOrder {
+				require.NotPanics(t, func() {
+					endFuncs[i]()
+				}, "%v", i)
+			}
+		})
+	}
+}
+
+func TestTaskEndWithActiveChildTaskNotAllowed(t *testing.T) {
+	root, _ := WithTask(context.Background(), "root")
+	c, endC := WithTask(root, "child")
+	_, _ = WithTask(c, "grand-child")
+	func() {
+		defer func() {
+			r := recover()
+			require.NotNil(t, r)
+			err, ok := r.(error)
+			require.True(t, ok)
+			require.Equal(t, ErrTaskStillHasActiveChildTasks, errors.Cause(err))
+		}()
+		endC()
+	}()
+
+}
+
+func TestIdempotentEndTask(t *testing.T) {
+	_, end := WithTask(context.Background(), "root")
+	end()
+	require.NotPanics(t, func() { end() })
+}
+
+func TestSpansPanicIfNoParentTask(t *testing.T) {
+	require.Panics(t, func() { WithSpan(context.Background(), "taskless-span") })
+}
+
+func TestIdempotentEndSpan(t *testing.T) {
+	root, _ := WithTask(context.Background(), "root")
+	_, end := WithSpan(root, "span")
+	end()
+	require.NotPanics(t, func() { end() })
+}
+
+func logAndGetTraceNode(t *testing.T, descr string, ctx context.Context) *traceNode {
+	n, ok := ctx.Value(contextKeyTraceNode).(*traceNode)
+	require.True(t, ok)
+	t.Logf("% 20s %p %#v", descr, n, n)
+	return n
+}
+
+func TestWhiteboxHierachy(t *testing.T) {
+	root, e1 := WithTask(context.Background(), "root")
+	rootN := logAndGetTraceNode(t, "root", root)
+	assert.Nil(t, rootN.parentTask)
+	assert.Nil(t, rootN.parentSpan)
+
+	child, e2 := WithSpan(root, "child")
+	childN := logAndGetTraceNode(t, "child", child)
+	assert.Equal(t, rootN, childN.parentTask)
+	assert.Equal(t, rootN, childN.parentSpan)
+
+	grandchild, e3 := WithSpan(child, "grandchild")
+	grandchildN := logAndGetTraceNode(t, "grandchild", grandchild)
+	assert.Equal(t, rootN, grandchildN.parentTask)
+	assert.Equal(t, childN, grandchildN.parentSpan)
+
+	gcTask, e4 := WithTask(grandchild, "grandchild-task")
+	gcTaskN := logAndGetTraceNode(t, "grandchild-task", gcTask)
+	assert.Equal(t, rootN, gcTaskN.parentTask)
+	assert.Nil(t, gcTaskN.parentSpan)
+
+	// it is allowed that a child task outlives the _span_ in which it was created
+	// (albeit not its parent task)
+	e3()
+	e2()
+	gcTaskSpan, e5 := WithSpan(gcTask, "granschild-task-span")
+	gcTaskSpanN := logAndGetTraceNode(t, "granschild-task-span", gcTaskSpan)
+	assert.Equal(t, gcTaskN, gcTaskSpanN.parentTask)
+	assert.Equal(t, gcTaskN, gcTaskSpanN.parentSpan)
+	e5()
+
+	e4()
+	e1()
+}
+
+func TestOrphanTasksBecomeNewRootWhenAllAncestorsAreDead(t *testing.T) {
+	parent, e1 := WithTask(context.Background(), "parent")
+	_ = logAndGetTraceNode(t, "parent-task", parent)
+	e1()
+
+	var child context.Context
+	var e2 DoneFunc
+	require.NotPanics(t, func() {
+		child, e2 = WithTask(parent, "child")
+	})
+	childN := logAndGetTraceNode(t, "child-task", child)
+	assert.Nil(t, childN.parentTask)
+
+	grandchild, _ := WithTask(child, "grandchild")
+	grandchildN := logAndGetTraceNode(t, "grandchild-task", grandchild)
+	assert.Equal(t, childN, grandchildN.parentTask)
+
+	require.Panics(t, func() { e2() }, "if the parent was alive at creation of child task, wrong termination order remains a panicable offense though")
+}
+
+func TestOrphanTaskBecomesChildToNearestLiveAncestor(t *testing.T) {
+	parent, e1 := WithTask(context.Background(), "parent")
+	parentN := logAndGetTraceNode(t, "parent-task", parent)
+
+	child, e2 := WithTask(parent, "child")
+	childN := logAndGetTraceNode(t, "child-task", child)
+	assert.Equal(t, parentN, childN.parentTask)
+	e2()
+
+	grandchild, e3 := WithTask(child, "grandchild")
+	grandchildN := logAndGetTraceNode(t, "grandchild-task", grandchild)
+	assert.Equal(t, parentN, grandchildN.parentTask) // child is already dead
+
+	require.NotPanics(t, func() {
+		e3()
+		e1()
+	})
+}

daemon/logging/trace/trace_unique_concurrent_task_namer.go

@@ -0,0 +1,53 @@
+package trace
+
+import (
+	"fmt"
+	"strings"
+	"sync"
+
+	"github.com/willf/bitset"
+)
+
+type uniqueConcurrentTaskNamer struct {
+	mtx    sync.Mutex
+	active map[string]*bitset.BitSet
+}
+
+// bitvecLengthGauge may be nil
+func newUniqueTaskNamer() *uniqueConcurrentTaskNamer {
+	return &uniqueConcurrentTaskNamer{
+		active: make(map[string]*bitset.BitSet),
+	}
+}
+
+// appends `#%d` to `name` such that until `done` is called,
+// it is guaranteed that `#%d` is not returned a second time for the same `name`
+func (namer *uniqueConcurrentTaskNamer) UniqueConcurrentTaskName(name string) (uniqueName string, done func()) {
+	if strings.Contains(name, "#") {
+		panic(name)
+	}
+	namer.mtx.Lock()
+	act, ok := namer.active[name]
+	if !ok {
+		act = bitset.New(64) // FIXME magic const
+		namer.active[name] = act
+	}
+	id, ok := act.NextClear(0)
+	if !ok {
+		// if !ok, all bits are 1 and act.Len() returns the next bit
+		id = act.Len()
+		// FIXME unbounded growth without reclamation
+	}
+	act.Set(id)
+	namer.mtx.Unlock()
+
+	return fmt.Sprintf("%s#%d", name, id), func() {
+		namer.mtx.Lock()
+		defer namer.mtx.Unlock()
+		act, ok := namer.active[name]
+		if !ok {
+			panic("must be initialized upon entry")
+		}
+		act.Clear(id)
+	}
+}

daemon/logging/trace/trace_unique_concurrent_task_namer_test.go

@@ -0,0 +1,58 @@
+package trace
+
+import (
+	"fmt"
+	"sync"
+	"sync/atomic"
+	"testing"
+
+	"github.com/stretchr/testify/require"
+	"github.com/willf/bitset"
+)
+
+func TestBitsetFeaturesForUniqueConcurrentTaskNamer(t *testing.T) {
+	var b bitset.BitSet
+	require.Equal(t, uint(0), b.Len())
+	require.Equal(t, uint(0), b.Count())
+
+	b.Set(0)
+	require.Equal(t, uint(1), b.Len())
+	require.Equal(t, uint(1), b.Count())
+
+	b.Set(8)
+	require.Equal(t, uint(9), b.Len())
+	require.Equal(t, uint(2), b.Count())
+
+	b.Set(1)
+	require.Equal(t, uint(9), b.Len())
+	require.Equal(t, uint(3), b.Count())
+}
+
+func TestUniqueConcurrentTaskNamer(t *testing.T) {
+	namer := newUniqueTaskNamer()
+
+	var wg sync.WaitGroup
+
+	const N = 8128
+	const Q = 23
+
+	var fails uint32
+	var m sync.Map
+	wg.Add(N)
+	for i := 0; i < N; i++ {
+		go func(i int) {
+			defer wg.Done()
+			name := fmt.Sprintf("%d", i/Q)
+			uniqueName, done := namer.UniqueConcurrentTaskName(name)
+			act, _ := m.LoadOrStore(uniqueName, i)
+			if act.(int) != i {
+				atomic.AddUint32(&fails, 1)
+			}
+			m.Delete(uniqueName)
+			done()
+		}(i)
+	}
+	wg.Wait()
+
+	require.Equal(t, uint32(0), fails)
+}

daemon/main.go

@@ -1,6 +1,8 @@
 package daemon
 
 import (
+	"context"
+
 	"github.com/zrepl/zrepl/cli"
 	"github.com/zrepl/zrepl/logger"
 )
@@ -10,7 +12,7 @@ type Logger = logger.Logger
 var DaemonCmd = &cli.Subcommand{
 	Use:   "daemon",
 	Short: "run the zrepl daemon",
-	Run: func(subcommand *cli.Subcommand, args []string) error {
+	Run: func(ctx context.Context, subcommand *cli.Subcommand, args []string) error {
-		return Run(subcommand.Config())
+		return Run(ctx, subcommand.Config())
 	},
 }

daemon/pprof.go

@@ -8,7 +8,11 @@ import (
 	"net"
 	"net/http/pprof"
 
+	"github.com/prometheus/client_golang/prometheus/promhttp"
+	"golang.org/x/net/websocket"
+
 	"github.com/zrepl/zrepl/daemon/job"
+	"github.com/zrepl/zrepl/daemon/logging/trace"
 )
 
 type pprofServer struct {
@@ -64,6 +68,8 @@ 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())
+			mux.Handle("/debug/zrepl/activity-trace", websocket.Handler(trace.ChrometraceClientWebsocketHandler))
 			go func() {
 				err := http.Serve(s.listener, mux)
 				if ctx.Err() != nil {

daemon/prometheus.go

@@ -10,6 +10,7 @@ import (
 
 	"github.com/zrepl/zrepl/config"
 	"github.com/zrepl/zrepl/daemon/job"
+	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/endpoint"
 	"github.com/zrepl/zrepl/logger"
 	"github.com/zrepl/zrepl/rpc/dataconn/frameconn"
@@ -86,16 +87,19 @@ func (j *prometheusJob) Run(ctx context.Context) {
 }
 
 type prometheusJobOutlet struct {
-	jobName string
 }
 
 var _ logger.Outlet = prometheusJobOutlet{}
 
-func newPrometheusLogOutlet(jobName string) prometheusJobOutlet {
+func newPrometheusLogOutlet() prometheusJobOutlet {
-	return prometheusJobOutlet{jobName}
+	return prometheusJobOutlet{}
 }
 
 func (o prometheusJobOutlet) WriteEntry(entry logger.Entry) error {
-	prom.taskLogEntries.WithLabelValues(o.jobName, entry.Level.String()).Inc()
+	jobFieldVal, ok := entry.Fields[logging.JobField].(string)
+	if !ok {
+		jobFieldVal = "_nojobid"
+	}
+	prom.taskLogEntries.WithLabelValues(jobFieldVal, entry.Level.String()).Inc()
 	return nil
 }

daemon/pruner/pruner.go

@@ -12,18 +12,27 @@ import (
 	"github.com/prometheus/client_golang/prometheus"
 
 	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/daemon/logging"
 	"github.com/zrepl/zrepl/logger"
 	"github.com/zrepl/zrepl/pruning"
 	"github.com/zrepl/zrepl/replication/logic/pdu"
 	"github.com/zrepl/zrepl/util/envconst"
 )
 
+// The sender in the replication setup.
+// The pruner uses the Sender to determine which of the Target's filesystems need to be pruned.
+// Also, it asks the Sender about the replication cursor of each filesystem
+// to enable the 'not_replicated' pruning rule.
+//
 // Try to keep it compatible with github.com/zrepl/zrepl/endpoint.Endpoint
-type History interface {
+type Sender interface {
 	ReplicationCursor(ctx context.Context, req *pdu.ReplicationCursorReq) (*pdu.ReplicationCursorRes, error)
 	ListFilesystems(ctx context.Context, req *pdu.ListFilesystemReq) (*pdu.ListFilesystemRes, error)
 }
 
+// The pruning target, i.e., on which snapshots are destroyed.
+// This can be a replication sender or receiver.
+//
 // Try to keep it compatible with github.com/zrepl/zrepl/endpoint.Endpoint
 type Target interface {
 	ListFilesystems(ctx context.Context, req *pdu.ListFilesystemReq) (*pdu.ListFilesystemRes, error)
@@ -35,23 +44,19 @@ type Logger = logger.Logger
 
 type contextKey int
 
-const contextKeyLogger contextKey = 0
+const (
-
+	contextKeyPruneSide contextKey = 1 + iota
-func WithLogger(ctx context.Context, log Logger) context.Context {
+)
-	return context.WithValue(ctx, contextKeyLogger, log)
-}
 
 func GetLogger(ctx context.Context) Logger {
-	if l, ok := ctx.Value(contextKeyLogger).(Logger); ok {
+	pruneSide := ctx.Value(contextKeyPruneSide).(string)
-		return l
+	return logging.GetLogger(ctx, logging.SubsysPruning).WithField("prune_side", pruneSide)
-	}
-	return logger.NewNullLogger()
 }
 
 type args struct {
 	ctx                            context.Context
 	target                         Target
-	receiver                       History
+	sender                         Sender
 	rules                          []pruning.KeepRule
 	retryWait                      time.Duration
 	considerSnapAtCursorReplicated bool
@@ -135,12 +140,12 @@ func NewPrunerFactory(in config.PruningSenderReceiver, promPruneSecs *prometheus
 	return f, nil
 }
 
-func (f *PrunerFactory) BuildSenderPruner(ctx context.Context, target Target, receiver History) *Pruner {
+func (f *PrunerFactory) BuildSenderPruner(ctx context.Context, target Target, sender Sender) *Pruner {
 	p := &Pruner{
 		args: args{
-			WithLogger(ctx, GetLogger(ctx).WithField("prune_side", "sender")),
+			context.WithValue(ctx, contextKeyPruneSide, "sender"),
 			target,
-			receiver,
+			sender,
 			f.senderRules,
 			f.retryWait,
 			f.considerSnapAtCursorReplicated,
@@ -151,12 +156,12 @@ func (f *PrunerFactory) BuildSenderPruner(ctx context.Context, target Target, re
 	return p
 }
 
-func (f *PrunerFactory) BuildReceiverPruner(ctx context.Context, target Target, receiver History) *Pruner {
+func (f *PrunerFactory) BuildReceiverPruner(ctx context.Context, target Target, sender Sender) *Pruner {
 	p := &Pruner{
 		args: args{
-			WithLogger(ctx, GetLogger(ctx).WithField("prune_side", "receiver")),
+			context.WithValue(ctx, contextKeyPruneSide, "receiver"),
 			target,
-			receiver,
+			sender,
 			f.receiverRules,
 			f.retryWait,
 			false, // senseless here anyways
@@ -167,12 +172,12 @@ func (f *PrunerFactory) BuildReceiverPruner(ctx context.Context, target Target,
 	return p
 }
 
-func (f *LocalPrunerFactory) BuildLocalPruner(ctx context.Context, target Target, receiver History) *Pruner {
+func (f *LocalPrunerFactory) BuildLocalPruner(ctx context.Context, target Target, history Sender) *Pruner {
 	p := &Pruner{
 		args: args{
-			ctx,
+			context.WithValue(ctx, contextKeyPruneSide, "local"),
 			target,
-			receiver,
+			history,
 			f.keepRules,
 			f.retryWait,
 			false, // considerSnapAtCursorReplicated is not relevant for local pruning
@@ -194,6 +199,16 @@ const (
 	Done
 )
 
+// Returns true in case the State is a terminal state(PlanErr, ExecErr, Done)
+func (s State) IsTerminal() bool {
+	switch s {
+	case PlanErr, ExecErr, Done:
+		return true
+	default:
+		return false
+	}
+}
+
 type updater func(func(*Pruner))
 
 func (p *Pruner) Prune() {
@@ -346,9 +361,9 @@ func (s snapshot) Date() time.Time { return s.date }
 
 func doOneAttempt(a *args, u updater) {
 
-	ctx, target, receiver := a.ctx, a.target, a.receiver
+	ctx, target, sender := a.ctx, a.target, a.sender
 
-	sfssres, err := receiver.ListFilesystems(ctx, &pdu.ListFilesystemReq{})
+	sfssres, err := sender.ListFilesystems(ctx, &pdu.ListFilesystemReq{})
 	if err != nil {
 		u(func(p *Pruner) {
 			p.state = PlanErr
@@ -413,7 +428,7 @@ tfss_loop:
 		rcReq := &pdu.ReplicationCursorReq{
 			Filesystem: tfs.Path,
 		}
-		rc, err := receiver.ReplicationCursor(ctx, rcReq)
+		rc, err := sender.ReplicationCursor(ctx, rcReq)
 		if err != nil {
 			pfsPlanErrAndLog(err, "cannot get replication cursor bookmark")
 			continue tfss_loop
@@ -459,7 +474,7 @@ tfss_loop:
 			})
 		}
 		if preCursor {
-			pfsPlanErrAndLog(fmt.Errorf("replication cursor not found in prune target filesystem versions"), "")
+			pfsPlanErrAndLog(fmt.Errorf("prune target has no snapshot that corresponds to sender replication cursor bookmark"), "")
 			continue tfss_loop
 		}
 

daemon/snapper/cron.go

@@ -0,0 +1,152 @@
+package snapper
+
+import (
+	"context"
+	"fmt"
+	"sync"
+	"time"
+
+	"github.com/pkg/errors"
+
+	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/daemon/hooks"
+	"github.com/zrepl/zrepl/util/suspendresumesafetimer"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+func cronFromConfig(fsf zfs.DatasetFilter, in config.SnapshottingCron) (*Cron, error) {
+
+	hooksList, err := hooks.ListFromConfig(&in.Hooks)
+	if err != nil {
+		return nil, errors.Wrap(err, "hook config error")
+	}
+	planArgs := planArgs{
+		prefix:          in.Prefix,
+		timestampFormat: in.TimestampFormat,
+		hooks:           hooksList,
+	}
+	return &Cron{config: in, fsf: fsf, planArgs: planArgs}, nil
+}
+
+type Cron struct {
+	config   config.SnapshottingCron
+	fsf      zfs.DatasetFilter
+	planArgs planArgs
+
+	mtx sync.RWMutex
+
+	running                 bool
+	wakeupTime              time.Time // zero value means uninit
+	lastError               error
+	lastPlan                *plan
+	wakeupWhileRunningCount int
+}
+
+func (s *Cron) Run(ctx context.Context, snapshotsTaken chan<- struct{}) {
+
+	for {
+		now := time.Now()
+		s.mtx.Lock()
+		s.wakeupTime = s.config.Cron.Schedule.Next(now)
+		s.mtx.Unlock()
+
+		ctxDone := suspendresumesafetimer.SleepUntil(ctx, s.wakeupTime)
+		if ctxDone != nil {
+			return
+		}
+
+		getLogger(ctx).Debug("cron timer fired")
+		s.mtx.Lock()
+		if s.running {
+			getLogger(ctx).Warn("snapshotting triggered according to cron rules but previous snapshotting is not done; not taking a snapshot this time")
+			s.wakeupWhileRunningCount++
+			s.mtx.Unlock()
+			continue
+		}
+		s.lastError = nil
+		s.lastPlan = nil
+		s.wakeupWhileRunningCount = 0
+		s.running = true
+		s.mtx.Unlock()
+		go func() {
+			err := s.do(ctx)
+			s.mtx.Lock()
+			s.lastError = err
+			s.running = false
+			s.mtx.Unlock()
+
+			select {
+			case snapshotsTaken <- struct{}{}:
+			default:
+				if snapshotsTaken != nil {
+					getLogger(ctx).Warn("callback channel is full, discarding snapshot update event")
+				}
+			}
+		}()
+	}
+
+}
+
+func (s *Cron) do(ctx context.Context) error {
+	fss, err := zfs.ZFSListMapping(ctx, s.fsf)
+	if err != nil {
+		return errors.Wrap(err, "cannot list filesystems")
+	}
+	p := makePlan(s.planArgs, fss)
+
+	s.mtx.Lock()
+	s.lastPlan = p
+	s.lastError = nil
+	s.mtx.Unlock()
+
+	ok := p.execute(ctx, false)
+	if !ok {
+		return errors.New("one or more snapshots could not be created, check logs for details")
+	} else {
+		return nil
+	}
+}
+
+type CronState string
+
+const (
+	CronStateRunning CronState = "running"
+	CronStateWaiting CronState = "waiting"
+)
+
+type CronReport struct {
+	State      CronState
+	WakeupTime time.Time
+	Errors     []string
+	Progress   []*ReportFilesystem
+}
+
+func (s *Cron) Report() Report {
+
+	s.mtx.Lock()
+	defer s.mtx.Unlock()
+
+	r := CronReport{}
+
+	r.WakeupTime = s.wakeupTime
+
+	if s.running {
+		r.State = CronStateRunning
+	} else {
+		r.State = CronStateWaiting
+	}
+
+	if s.lastError != nil {
+		r.Errors = append(r.Errors, s.lastError.Error())
+	}
+	if s.wakeupWhileRunningCount > 0 {
+		r.Errors = append(r.Errors, fmt.Sprintf("cron frequency is too high; snapshots were not taken %d times", s.wakeupWhileRunningCount))
+	}
+
+	r.Progress = nil
+	if s.lastPlan != nil {
+		r.Progress = s.lastPlan.report()
+	}
+
+	return Report{Type: TypeCron, Cron: &r}
+}

daemon/snapper/cron_test.go

@@ -0,0 +1,68 @@
+package snapper
+
+import (
+	"fmt"
+	"testing"
+	"time"
+
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+	"github.com/zrepl/yaml-config"
+
+	"github.com/zrepl/zrepl/config"
+)
+
+func TestCronLibraryWorks(t *testing.T) {
+
+	type testCase struct {
+		spec   string
+		in     time.Time
+		expect time.Time
+	}
+	dhm := func(day, hour, minutes int) time.Time {
+		return time.Date(2022, 7, day, hour, minutes, 0, 0, time.UTC)
+	}
+	hm := func(hour, minutes int) time.Time {
+		return dhm(23, hour, minutes)
+	}
+
+	tcs := []testCase{
+		{"0-10 * * * *", dhm(17, 1, 10), dhm(17, 2, 0)},
+		{"0-10 * * * *", dhm(17, 23, 10), dhm(18, 0, 0)},
+		{"0-10 * * * *", hm(1, 9), hm(1, 10)},
+		{"0-10 * * * *", hm(1, 9), hm(1, 10)},
+
+		{"1,3,5 * * * *", hm(1, 1), hm(1, 3)},
+		{"1,3,5 * * * *", hm(1, 2), hm(1, 3)},
+		{"1,3,5 * * * *", hm(1, 3), hm(1, 5)},
+		{"1,3,5 * * * *", hm(1, 5), hm(2, 1)},
+
+		{"* 0-5,8,12 * * *", hm(0, 0), hm(0, 1)},
+		{"* 0-5,8,12 * * *", hm(4, 59), hm(5, 0)},
+		{"* 0-5,8,12 * * *", hm(5, 0), hm(5, 1)},
+		{"* 0-5,8,12 * * *", hm(5, 59), hm(8, 0)},
+		{"* 0-5,8,12 * * *", hm(8, 59), hm(12, 0)},
+
+		// https://github.com/zrepl/zrepl/pull/614#issuecomment-1188358989
+		{"53 17,18,19 * * *", dhm(23, 17, 52), dhm(23, 17, 53)},
+		{"53 17,18,19 * * *", dhm(23, 17, 53), dhm(23, 18, 53)},
+		{"53 17,18,19 * * *", dhm(23, 18, 53), dhm(23, 19, 53)},
+		{"53 17,18,19 * * *", dhm(23, 19, 53), dhm(24 /* ! */, 17, 53)},
+	}
+
+	for i, tc := range tcs {
+		t.Run(fmt.Sprintf("%d", i), func(t *testing.T) {
+			var s struct {
+				Cron config.CronSpec `yaml:"cron"`
+			}
+			inp := fmt.Sprintf("cron: %q", tc.spec)
+			fmt.Println("spec is ", inp)
+			err := yaml.UnmarshalStrict([]byte(inp), &s)
+			require.NoError(t, err)
+
+			actual := s.Cron.Schedule.Next(tc.in)
+			assert.Equal(t, tc.expect, actual)
+		})
+	}
+
+}

daemon/snapper/impl.go

@@ -0,0 +1,268 @@
+package snapper
+
+import (
+	"context"
+	"fmt"
+	"sort"
+	"strconv"
+	"strings"
+	"time"
+
+	"github.com/zrepl/zrepl/daemon/hooks"
+	"github.com/zrepl/zrepl/daemon/logging"
+	"github.com/zrepl/zrepl/util/chainlock"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+type planArgs struct {
+	prefix          string
+	timestampFormat string
+	hooks           *hooks.List
+}
+
+type plan struct {
+	mtx   chainlock.L
+	args  planArgs
+	snaps map[*zfs.DatasetPath]*snapProgress
+}
+
+func makePlan(args planArgs, fss []*zfs.DatasetPath) *plan {
+	snaps := make(map[*zfs.DatasetPath]*snapProgress, len(fss))
+	for _, fs := range fss {
+		snaps[fs] = &snapProgress{state: SnapPending}
+	}
+	return &plan{snaps: snaps, args: args}
+}
+
+//go:generate stringer -type=SnapState
+type SnapState uint
+
+const (
+	SnapPending SnapState = 1 << iota
+	SnapStarted
+	SnapDone
+	SnapError
+)
+
+// All fields protected by Snapper.mtx
+type snapProgress struct {
+	state SnapState
+
+	// SnapStarted, SnapDone, SnapError
+	name     string
+	startAt  time.Time
+	hookPlan *hooks.Plan
+
+	// SnapDone
+	doneAt time.Time
+
+	// SnapErr TODO disambiguate state
+	runResults hooks.PlanReport
+}
+
+func (plan *plan) formatNow(format string) string {
+	now := time.Now().UTC()
+	switch strings.ToLower(format) {
+	case "dense":
+		format = "20060102_150405_000"
+	case "human":
+		format = "2006-01-02_15:04:05"
+	case "iso-8601":
+		format = "2006-01-02T15:04:05.000Z"
+	case "unix-seconds":
+		return strconv.FormatInt(now.Unix(), 10)
+	}
+	return now.Format(format)
+}
+
+func (plan *plan) execute(ctx context.Context, dryRun bool) (ok bool) {
+
+	hookMatchCount := make(map[hooks.Hook]int, len(*plan.args.hooks))
+	for _, h := range *plan.args.hooks {
+		hookMatchCount[h] = 0
+	}
+
+	anyFsHadErr := false
+	// TODO channel programs -> allow a little jitter?
+	for fs, progress := range plan.snaps {
+		suffix := plan.formatNow(plan.args.timestampFormat)
+		snapname := fmt.Sprintf("%s%s", plan.args.prefix, suffix)
+
+		ctx := logging.WithInjectedField(ctx, "fs", fs.ToString())
+		ctx = logging.WithInjectedField(ctx, "snap", snapname)
+
+		hookEnvExtra := hooks.Env{
+			hooks.EnvFS:       fs.ToString(),
+			hooks.EnvSnapshot: snapname,
+		}
+
+		jobCallback := hooks.NewCallbackHookForFilesystem("snapshot", fs, func(ctx context.Context) (err error) {
+			l := getLogger(ctx)
+			l.Debug("create snapshot")
+			err = zfs.ZFSSnapshot(ctx, fs, snapname, false) // TODO propagate context to ZFSSnapshot
+			if err != nil {
+				l.WithError(err).Error("cannot create snapshot")
+			}
+			return
+		})
+
+		fsHadErr := false
+		var hookPlanReport hooks.PlanReport
+		var hookPlan *hooks.Plan
+		{
+			filteredHooks, err := plan.args.hooks.CopyFilteredForFilesystem(fs)
+			if err != nil {
+				getLogger(ctx).WithError(err).Error("unexpected filter error")
+				fsHadErr = true
+				goto updateFSState
+			}
+			// account for running hooks
+			for _, h := range filteredHooks {
+				hookMatchCount[h] = hookMatchCount[h] + 1
+			}
+
+			var planErr error
+			hookPlan, planErr = hooks.NewPlan(&filteredHooks, hooks.PhaseSnapshot, jobCallback, hookEnvExtra)
+			if planErr != nil {
+				fsHadErr = true
+				getLogger(ctx).WithError(planErr).Error("cannot create job hook plan")
+				goto updateFSState
+			}
+		}
+
+		plan.mtx.HoldWhile(func() {
+			progress.name = snapname
+			progress.startAt = time.Now()
+			progress.hookPlan = hookPlan
+			progress.state = SnapStarted
+		})
+
+		{
+			getLogger(ctx).WithField("report", hookPlan.Report().String()).Debug("begin run job plan")
+			hookPlan.Run(ctx, dryRun)
+			hookPlanReport = hookPlan.Report()
+			fsHadErr = hookPlanReport.HadError() // not just fatal errors
+			if fsHadErr {
+				getLogger(ctx).WithField("report", hookPlanReport.String()).Error("end run job plan with error")
+			} else {
+				getLogger(ctx).WithField("report", hookPlanReport.String()).Info("end run job plan successful")
+			}
+		}
+
+	updateFSState:
+		anyFsHadErr = anyFsHadErr || fsHadErr
+		plan.mtx.HoldWhile(func() {
+			progress.doneAt = time.Now()
+			progress.state = SnapDone
+			if fsHadErr {
+				progress.state = SnapError
+			}
+			progress.runResults = hookPlanReport
+		})
+	}
+
+	for h, mc := range hookMatchCount {
+		if mc == 0 {
+			hookIdx := -1
+			for idx, ah := range *plan.args.hooks {
+				if ah == h {
+					hookIdx = idx
+					break
+				}
+			}
+			getLogger(ctx).WithField("hook", h.String()).WithField("hook_number", hookIdx+1).Warn("hook did not match any snapshotted filesystems")
+		}
+	}
+
+	return !anyFsHadErr
+}
+
+type ReportFilesystem struct {
+	Path  string
+	State SnapState
+
+	// Valid in SnapStarted and later
+	SnapName      string
+	StartAt       time.Time
+	Hooks         string
+	HooksHadError bool
+
+	// Valid in SnapDone | SnapError
+	DoneAt time.Time
+}
+
+func (plan *plan) report() []*ReportFilesystem {
+	plan.mtx.Lock()
+	defer plan.mtx.Unlock()
+
+	pReps := make([]*ReportFilesystem, 0, len(plan.snaps))
+	for fs, p := range plan.snaps {
+		var hooksStr string
+		var hooksHadError bool
+		if p.hookPlan != nil {
+			hooksStr, hooksHadError = p.report()
+		}
+		pReps = append(pReps, &ReportFilesystem{
+			Path:          fs.ToString(),
+			State:         p.state,
+			SnapName:      p.name,
+			StartAt:       p.startAt,
+			DoneAt:        p.doneAt,
+			Hooks:         hooksStr,
+			HooksHadError: hooksHadError,
+		})
+	}
+
+	sort.Slice(pReps, func(i, j int) bool {
+		return strings.Compare(pReps[i].Path, pReps[j].Path) == -1
+	})
+
+	return pReps
+}
+
+func (p *snapProgress) report() (hooksStr string, hooksHadError bool) {
+	hr := p.hookPlan.Report()
+	// FIXME: technically this belongs into client
+	// but we can't serialize hooks.Step ATM
+	rightPad := func(str string, length int, pad string) string {
+		if len(str) > length {
+			return str[:length]
+		}
+		return str + strings.Repeat(pad, length-len(str))
+	}
+	hooksHadError = hr.HadError()
+	rows := make([][]string, len(hr))
+	const numCols = 4
+	lens := make([]int, numCols)
+	for i, e := range hr {
+		rows[i] = make([]string, numCols)
+		rows[i][0] = fmt.Sprintf("%d", i+1)
+		rows[i][1] = e.Status.String()
+		runTime := "..."
+		if e.Status != hooks.StepPending {
+			runTime = e.End.Sub(e.Begin).Round(time.Millisecond).String()
+		}
+		rows[i][2] = runTime
+		rows[i][3] = ""
+		if e.Report != nil {
+			rows[i][3] = e.Report.String()
+		}
+		for j, col := range lens {
+			if len(rows[i][j]) > col {
+				lens[j] = len(rows[i][j])
+			}
+		}
+	}
+	rowsFlat := make([]string, len(hr))
+	for i, r := range rows {
+		colsPadded := make([]string, len(r))
+		for j, c := range r[:len(r)-1] {
+			colsPadded[j] = rightPad(c, lens[j], " ")
+		}
+		colsPadded[len(r)-1] = r[len(r)-1]
+		rowsFlat[i] = strings.Join(colsPadded, " ")
+	}
+	hooksStr = strings.Join(rowsFlat, "\n")
+
+	return hooksStr, hooksHadError
+}

daemon/snapper/manual.go

@@ -0,0 +1,15 @@
+package snapper
+
+import (
+	"context"
+)
+
+type manual struct{}
+
+func (s *manual) Run(ctx context.Context, wakeUpCommon chan<- struct{}) {
+	// nothing to do
+}
+
+func (s *manual) Report() Report {
+	return Report{Type: TypeManual, Manual: &struct{}{}}
+}

daemon/snapper/periodic.go

@@ -0,0 +1,394 @@
+package snapper
+
+import (
+	"context"
+	"fmt"
+	"sort"
+	"sync"
+	"time"
+
+	"github.com/pkg/errors"
+
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+
+	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/daemon/hooks"
+	"github.com/zrepl/zrepl/daemon/logging"
+	"github.com/zrepl/zrepl/util/envconst"
+	"github.com/zrepl/zrepl/util/suspendresumesafetimer"
+	"github.com/zrepl/zrepl/zfs"
+)
+
+func periodicFromConfig(g *config.Global, fsf zfs.DatasetFilter, in *config.SnapshottingPeriodic) (*Periodic, error) {
+	if in.Prefix == "" {
+		return nil, errors.New("prefix must not be empty")
+	}
+	if in.Interval.Duration() <= 0 {
+		return nil, errors.New("interval must be positive")
+	}
+
+	hookList, err := hooks.ListFromConfig(&in.Hooks)
+	if err != nil {
+		return nil, errors.Wrap(err, "hook config error")
+	}
+
+	args := periodicArgs{
+		interval: in.Interval.Duration(),
+		fsf:      fsf,
+		planArgs: planArgs{
+			prefix:          in.Prefix,
+			timestampFormat: in.TimestampFormat,
+			hooks:           hookList,
+		},
+		// ctx and log is set in Run()
+	}
+
+	return &Periodic{state: SyncUp, args: args}, nil
+}
+
+type periodicArgs struct {
+	ctx            context.Context
+	interval       time.Duration
+	fsf            zfs.DatasetFilter
+	planArgs       planArgs
+	snapshotsTaken chan<- struct{}
+	dryRun         bool
+}
+
+type Periodic struct {
+	args periodicArgs
+
+	mtx   sync.Mutex
+	state State
+
+	// set in state Plan, used in Waiting
+	lastInvocation time.Time
+
+	// valid for state Snapshotting
+	plan *plan
+
+	// valid for state SyncUp and Waiting
+	sleepUntil time.Time
+
+	// valid for state Err
+	err error
+}
+
+//go:generate stringer -type=State
+type State uint
+
+const (
+	SyncUp State = 1 << iota
+	SyncUpErrWait
+	Planning
+	Snapshotting
+	Waiting
+	ErrorWait
+	Stopped
+)
+
+func (s State) sf() state {
+	m := map[State]state{
+		SyncUp:        periodicStateSyncUp,
+		SyncUpErrWait: periodicStateWait,
+		Planning:      periodicStatePlan,
+		Snapshotting:  periodicStateSnapshot,
+		Waiting:       periodicStateWait,
+		ErrorWait:     periodicStateWait,
+		Stopped:       nil,
+	}
+	return m[s]
+}
+
+type updater func(u func(*Periodic)) State
+type state func(a periodicArgs, u updater) state
+
+func (s *Periodic) Run(ctx context.Context, snapshotsTaken chan<- struct{}) {
+	defer trace.WithSpanFromStackUpdateCtx(&ctx)()
+	getLogger(ctx).Debug("start")
+	defer getLogger(ctx).Debug("stop")
+
+	s.args.snapshotsTaken = snapshotsTaken
+	s.args.ctx = ctx
+	s.args.dryRun = false // for future expansion
+
+	u := func(u func(*Periodic)) State {
+		s.mtx.Lock()
+		defer s.mtx.Unlock()
+		if u != nil {
+			u(s)
+		}
+		return s.state
+	}
+
+	var st state = periodicStateSyncUp
+
+	for st != nil {
+		pre := u(nil)
+		st = st(s.args, u)
+		post := u(nil)
+		getLogger(ctx).
+			WithField("transition", fmt.Sprintf("%s=>%s", pre, post)).
+			Debug("state transition")
+
+	}
+
+}
+
+func onErr(err error, u updater) state {
+	return u(func(s *Periodic) {
+		s.err = err
+		preState := s.state
+		switch s.state {
+		case SyncUp:
+			s.state = SyncUpErrWait
+		case Planning:
+			fallthrough
+		case Snapshotting:
+			s.state = ErrorWait
+		}
+		getLogger(s.args.ctx).WithError(err).WithField("pre_state", preState).WithField("post_state", s.state).Error("snapshotting error")
+	}).sf()
+}
+
+func onMainCtxDone(ctx context.Context, u updater) state {
+	return u(func(s *Periodic) {
+		s.err = ctx.Err()
+		s.state = Stopped
+	}).sf()
+}
+
+func periodicStateSyncUp(a periodicArgs, u updater) state {
+	u(func(snapper *Periodic) {
+		snapper.lastInvocation = time.Now()
+	})
+	fss, err := listFSes(a.ctx, a.fsf)
+	if err != nil {
+		return onErr(err, u)
+	}
+	syncPoint, err := findSyncPoint(a.ctx, fss, a.planArgs.prefix, a.interval)
+	if err != nil {
+		return onErr(err, u)
+	}
+	u(func(s *Periodic) {
+		s.sleepUntil = syncPoint
+	})
+	ctxDone := suspendresumesafetimer.SleepUntil(a.ctx, syncPoint)
+	if ctxDone != nil {
+		return onMainCtxDone(a.ctx, u)
+	}
+	return u(func(s *Periodic) {
+		s.state = Planning
+	}).sf()
+}
+
+func periodicStatePlan(a periodicArgs, u updater) state {
+	u(func(snapper *Periodic) {
+		snapper.lastInvocation = time.Now()
+	})
+	fss, err := listFSes(a.ctx, a.fsf)
+	if err != nil {
+		return onErr(err, u)
+	}
+	p := makePlan(a.planArgs, fss)
+	return u(func(s *Periodic) {
+		s.state = Snapshotting
+		s.plan = p
+		s.err = nil
+	}).sf()
+}
+
+func periodicStateSnapshot(a periodicArgs, u updater) state {
+
+	var plan *plan
+	u(func(snapper *Periodic) {
+		plan = snapper.plan
+	})
+
+	ok := plan.execute(a.ctx, false)
+
+	select {
+	case a.snapshotsTaken <- struct{}{}:
+	default:
+		if a.snapshotsTaken != nil {
+			getLogger(a.ctx).Warn("callback channel is full, discarding snapshot update event")
+		}
+	}
+
+	return u(func(snapper *Periodic) {
+		if !ok {
+			snapper.state = ErrorWait
+			snapper.err = errors.New("one or more snapshots could not be created, check logs for details")
+		} else {
+			snapper.state = Waiting
+			snapper.err = nil
+		}
+	}).sf()
+}
+
+func periodicStateWait(a periodicArgs, u updater) state {
+	var sleepUntil time.Time
+	u(func(snapper *Periodic) {
+		lastTick := snapper.lastInvocation
+		snapper.sleepUntil = lastTick.Add(a.interval)
+		sleepUntil = snapper.sleepUntil
+		log := getLogger(a.ctx).WithField("sleep_until", sleepUntil).WithField("duration", a.interval)
+		logFunc := log.Debug
+		if snapper.state == ErrorWait || snapper.state == SyncUpErrWait {
+			logFunc = log.Error
+		}
+		logFunc("enter wait-state after error")
+	})
+
+	ctxDone := suspendresumesafetimer.SleepUntil(a.ctx, sleepUntil)
+	if ctxDone != nil {
+		return onMainCtxDone(a.ctx, u)
+	}
+	return u(func(snapper *Periodic) {
+		snapper.state = Planning
+	}).sf()
+}
+
+func listFSes(ctx context.Context, mf zfs.DatasetFilter) (fss []*zfs.DatasetPath, err error) {
+	return zfs.ZFSListMapping(ctx, mf)
+}
+
+var syncUpWarnNoSnapshotUntilSyncupMinDuration = envconst.Duration("ZREPL_SNAPPER_SYNCUP_WARN_MIN_DURATION", 1*time.Second)
+
+// see docs/snapshotting.rst
+func findSyncPoint(ctx context.Context, fss []*zfs.DatasetPath, prefix string, interval time.Duration) (syncPoint time.Time, err error) {
+
+	const (
+		prioHasVersions int = iota
+		prioNoVersions
+	)
+
+	type snapTime struct {
+		ds   *zfs.DatasetPath
+		prio int // lower is higher
+		time time.Time
+	}
+
+	if len(fss) == 0 {
+		return time.Now(), nil
+	}
+
+	snaptimes := make([]snapTime, 0, len(fss))
+	hardErrs := 0
+
+	now := time.Now()
+
+	getLogger(ctx).Debug("examine filesystem state to find sync point")
+	for _, d := range fss {
+		ctx := logging.WithInjectedField(ctx, "fs", d.ToString())
+		syncPoint, err := findSyncPointFSNextOptimalSnapshotTime(ctx, now, interval, prefix, d)
+		if err == findSyncPointFSNoFilesystemVersionsErr {
+			snaptimes = append(snaptimes, snapTime{
+				ds:   d,
+				prio: prioNoVersions,
+				time: now,
+			})
+		} else if err != nil {
+			hardErrs++
+			getLogger(ctx).WithError(err).Error("cannot determine optimal sync point for this filesystem")
+		} else {
+			getLogger(ctx).WithField("syncPoint", syncPoint).Debug("found optimal sync point for this filesystem")
+			snaptimes = append(snaptimes, snapTime{
+				ds:   d,
+				prio: prioHasVersions,
+				time: syncPoint,
+			})
+		}
+	}
+
+	if hardErrs == len(fss) {
+		return time.Time{}, fmt.Errorf("hard errors in determining sync point for every matching filesystem")
+	}
+
+	if len(snaptimes) == 0 {
+		panic("implementation error: loop must either inc hardErrs or add result to snaptimes")
+	}
+
+	// sort ascending by (prio,time)
+	// => those filesystems with versions win over those without any
+	sort.Slice(snaptimes, func(i, j int) bool {
+		if snaptimes[i].prio == snaptimes[j].prio {
+			return snaptimes[i].time.Before(snaptimes[j].time)
+		}
+		return snaptimes[i].prio < snaptimes[j].prio
+	})
+
+	winnerSyncPoint := snaptimes[0].time
+	l := getLogger(ctx).WithField("syncPoint", winnerSyncPoint.String())
+	l.Info("determined sync point")
+	if winnerSyncPoint.Sub(now) > syncUpWarnNoSnapshotUntilSyncupMinDuration {
+		for _, st := range snaptimes {
+			if st.prio == prioNoVersions {
+				l.WithField("fs", st.ds.ToString()).Warn("filesystem will not be snapshotted until sync point")
+			}
+		}
+	}
+
+	return snaptimes[0].time, nil
+
+}
+
+var findSyncPointFSNoFilesystemVersionsErr = fmt.Errorf("no filesystem versions")
+
+func findSyncPointFSNextOptimalSnapshotTime(ctx context.Context, now time.Time, interval time.Duration, prefix string, d *zfs.DatasetPath) (time.Time, error) {
+
+	fsvs, err := zfs.ZFSListFilesystemVersions(ctx, d, zfs.ListFilesystemVersionsOptions{
+		Types:           zfs.Snapshots,
+		ShortnamePrefix: prefix,
+	})
+	if err != nil {
+		return time.Time{}, errors.Wrap(err, "list filesystem versions")
+	}
+	if len(fsvs) <= 0 {
+		return time.Time{}, findSyncPointFSNoFilesystemVersionsErr
+	}
+
+	// Sort versions by creation
+	sort.SliceStable(fsvs, func(i, j int) bool {
+		return fsvs[i].CreateTXG < fsvs[j].CreateTXG
+	})
+
+	latest := fsvs[len(fsvs)-1]
+	getLogger(ctx).WithField("creation", latest.Creation).Debug("found latest snapshot")
+
+	since := now.Sub(latest.Creation)
+	if since < 0 {
+		return time.Time{}, fmt.Errorf("snapshot %q is from the future: creation=%q now=%q", latest.ToAbsPath(d), latest.Creation, now)
+	}
+
+	return latest.Creation.Add(interval), nil
+}
+
+type PeriodicReport struct {
+	State State
+	// valid in state SyncUp and Waiting
+	SleepUntil time.Time
+	// valid in state Err
+	Error string
+	// valid in state Snapshotting
+	Progress []*ReportFilesystem
+}
+
+func (s *Periodic) Report() Report {
+	s.mtx.Lock()
+	defer s.mtx.Unlock()
+
+	var progress []*ReportFilesystem = nil
+	if s.plan != nil {
+		progress = s.plan.report()
+	}
+
+	r := &PeriodicReport{
+		State:      s.state,
+		SleepUntil: s.sleepUntil,
+		Error:      errOrEmptyString(s.err),
+		Progress:   progress,
+	}
+
+	return Report{Type: TypePeriodic, Periodic: r}
+}

daemon/snapper/snapper.go

@@ -3,511 +3,40 @@ package snapper
 import (
 	"context"
 	"fmt"
-	"sort"
-	"sync"
-	"time"
-
-	"github.com/pkg/errors"
 
 	"github.com/zrepl/zrepl/config"
-	"github.com/zrepl/zrepl/daemon/filters"
-	"github.com/zrepl/zrepl/daemon/hooks"
-	"github.com/zrepl/zrepl/logger"
-	"github.com/zrepl/zrepl/util/envconst"
 	"github.com/zrepl/zrepl/zfs"
 )
 
-//go:generate stringer -type=SnapState
+type Type string
-type SnapState uint
 
 const (
-	SnapPending SnapState = 1 << iota
+	TypePeriodic Type = "periodic"
-	SnapStarted
+	TypeCron     Type = "cron"
-	SnapDone
+	TypeManual   Type = "manual"
-	SnapError
 )
 
-// All fields protected by Snapper.mtx
+type Snapper interface {
-type snapProgress struct {
+	Run(ctx context.Context, snapshotsTaken chan<- struct{})
-	state SnapState
+	Report() Report
-
-	// SnapStarted, SnapDone, SnapError
-	name     string
-	startAt  time.Time
-	hookPlan *hooks.Plan
-
-	// SnapDone
-	doneAt time.Time
-
-	// SnapErr TODO disambiguate state
-	runResults hooks.PlanReport
 }
 
-type args struct {
+type Report struct {
-	ctx            context.Context
+	Type     Type
-	log            Logger
+	Periodic *PeriodicReport
-	prefix         string
+	Cron     *CronReport
-	interval       time.Duration
+	Manual   *struct{}
-	fsf            *filters.DatasetMapFilter
-	snapshotsTaken chan<- struct{}
-	hooks          *hooks.List
-	dryRun         bool
 }
 
-type Snapper struct {
+func FromConfig(g *config.Global, fsf zfs.DatasetFilter, in config.SnapshottingEnum) (Snapper, error) {
-	args args
+	switch v := in.Ret.(type) {
-
+	case *config.SnapshottingPeriodic:
-	mtx   sync.Mutex
+		return periodicFromConfig(g, fsf, v)
-	state State
+	case *config.SnapshottingCron:
-
+		return cronFromConfig(fsf, *v)
-	// set in state Plan, used in Waiting
+	case *config.SnapshottingManual:
-	lastInvocation time.Time
+		return &manual{}, nil
-
-	// valid for state Snapshotting
-	plan map[*zfs.DatasetPath]*snapProgress
-
-	// valid for state SyncUp and Waiting
-	sleepUntil time.Time
-
-	// valid for state Err
-	err error
-}
-
-//go:generate stringer -type=State
-type State uint
-
-const (
-	SyncUp State = 1 << iota
-	SyncUpErrWait
-	Planning
-	Snapshotting
-	Waiting
-	ErrorWait
-	Stopped
-)
-
-func (s State) sf() state {
-	m := map[State]state{
-		SyncUp:        syncUp,
-		SyncUpErrWait: wait,
-		Planning:      plan,
-		Snapshotting:  snapshot,
-		Waiting:       wait,
-		ErrorWait:     wait,
-		Stopped:       nil,
-	}
-	return m[s]
-}
-
-type updater func(u func(*Snapper)) State
-type state func(a args, u updater) state
-
-type contextKey int
-
-const (
-	contextKeyLog contextKey = 0
-)
-
-type Logger = logger.Logger
-
-func WithLogger(ctx context.Context, log Logger) context.Context {
-	return context.WithValue(ctx, contextKeyLog, log)
-}
-
-func getLogger(ctx context.Context) Logger {
-	if log, ok := ctx.Value(contextKeyLog).(Logger); ok {
-		return log
-	}
-	return logger.NewNullLogger()
-}
-
-func PeriodicFromConfig(g *config.Global, fsf *filters.DatasetMapFilter, in *config.SnapshottingPeriodic) (*Snapper, error) {
-	if in.Prefix == "" {
-		return nil, errors.New("prefix must not be empty")
-	}
-	if in.Interval <= 0 {
-		return nil, errors.New("interval must be positive")
-	}
-
-	hookList, err := hooks.ListFromConfig(&in.Hooks)
-	if err != nil {
-		return nil, errors.Wrap(err, "hook config error")
-	}
-
-	args := args{
-		prefix:   in.Prefix,
-		interval: in.Interval,
-		fsf:      fsf,
-		hooks:    hookList,
-		// ctx and log is set in Run()
-	}
-
-	return &Snapper{state: SyncUp, args: args}, nil
-}
-
-func (s *Snapper) Run(ctx context.Context, snapshotsTaken chan<- struct{}) {
-
-	getLogger(ctx).Debug("start")
-	defer getLogger(ctx).Debug("stop")
-
-	s.args.snapshotsTaken = snapshotsTaken
-	s.args.ctx = ctx
-	s.args.log = getLogger(ctx)
-	s.args.dryRun = false // for future expansion
-
-	u := func(u func(*Snapper)) State {
-		s.mtx.Lock()
-		defer s.mtx.Unlock()
-		if u != nil {
-			u(s)
-		}
-		return s.state
-	}
-
-	var st state = syncUp
-
-	for st != nil {
-		pre := u(nil)
-		st = st(s.args, u)
-		post := u(nil)
-		getLogger(ctx).
-			WithField("transition", fmt.Sprintf("%s=>%s", pre, post)).
-			Debug("state transition")
-
-	}
-
-}
-
-func onErr(err error, u updater) state {
-	return u(func(s *Snapper) {
-		s.err = err
-		preState := s.state
-		switch s.state {
-		case SyncUp:
-			s.state = SyncUpErrWait
-		case Planning:
-			fallthrough
-		case Snapshotting:
-			s.state = ErrorWait
-		}
-		s.args.log.WithError(err).WithField("pre_state", preState).WithField("post_state", s.state).Error("snapshotting error")
-	}).sf()
-}
-
-func onMainCtxDone(ctx context.Context, u updater) state {
-	return u(func(s *Snapper) {
-		s.err = ctx.Err()
-		s.state = Stopped
-	}).sf()
-}
-
-func syncUp(a args, u updater) state {
-	u(func(snapper *Snapper) {
-		snapper.lastInvocation = time.Now()
-	})
-	fss, err := listFSes(a.ctx, a.fsf)
-	if err != nil {
-		return onErr(err, u)
-	}
-	syncPoint, err := findSyncPoint(a.log, fss, a.prefix, a.interval)
-	if err != nil {
-		return onErr(err, u)
-	}
-	u(func(s *Snapper) {
-		s.sleepUntil = syncPoint
-	})
-	t := time.NewTimer(time.Until(syncPoint))
-	defer t.Stop()
-	select {
-	case <-t.C:
-		return u(func(s *Snapper) {
-			s.state = Planning
-		}).sf()
-	case <-a.ctx.Done():
-		return onMainCtxDone(a.ctx, u)
-	}
-}
-
-func plan(a args, u updater) state {
-	u(func(snapper *Snapper) {
-		snapper.lastInvocation = time.Now()
-	})
-	fss, err := listFSes(a.ctx, a.fsf)
-	if err != nil {
-		return onErr(err, u)
-	}
-
-	plan := make(map[*zfs.DatasetPath]*snapProgress, len(fss))
-	for _, fs := range fss {
-		plan[fs] = &snapProgress{state: SnapPending}
-	}
-	return u(func(s *Snapper) {
-		s.state = Snapshotting
-		s.plan = plan
-		s.err = nil
-	}).sf()
-}
-
-func snapshot(a args, u updater) state {
-
-	var plan map[*zfs.DatasetPath]*snapProgress
-	u(func(snapper *Snapper) {
-		plan = snapper.plan
-	})
-
-	hookMatchCount := make(map[hooks.Hook]int, len(*a.hooks))
-	for _, h := range *a.hooks {
-		hookMatchCount[h] = 0
-	}
-
-	anyFsHadErr := false
-	// TODO channel programs -> allow a little jitter?
-	for fs, progress := range plan {
-		suffix := time.Now().In(time.UTC).Format("20060102_150405_000")
-		snapname := fmt.Sprintf("%s%s", a.prefix, suffix)
-
-		l := a.log.
-			WithField("fs", fs.ToString()).
-			WithField("snap", snapname)
-
-		hookEnvExtra := hooks.Env{
-			hooks.EnvFS:       fs.ToString(),
-			hooks.EnvSnapshot: snapname,
-		}
-
-		jobCallback := hooks.NewCallbackHookForFilesystem("snapshot", fs, func(_ context.Context) (err error) {
-			l.Debug("create snapshot")
-			err = zfs.ZFSSnapshot(a.ctx, fs, snapname, false) // TODO propagate context to ZFSSnapshot
-			if err != nil {
-				l.WithError(err).Error("cannot create snapshot")
-			}
-			return
-		})
-
-		fsHadErr := false
-		var planReport hooks.PlanReport
-		var plan *hooks.Plan
-		{
-			filteredHooks, err := a.hooks.CopyFilteredForFilesystem(fs)
-			if err != nil {
-				l.WithError(err).Error("unexpected filter error")
-				fsHadErr = true
-				goto updateFSState
-			}
-			// account for running hooks
-			for _, h := range filteredHooks {
-				hookMatchCount[h] = hookMatchCount[h] + 1
-			}
-
-			var planErr error
-			plan, planErr = hooks.NewPlan(&filteredHooks, hooks.PhaseSnapshot, jobCallback, hookEnvExtra)
-			if planErr != nil {
-				fsHadErr = true
-				l.WithError(planErr).Error("cannot create job hook plan")
-				goto updateFSState
-			}
-		}
-		u(func(snapper *Snapper) {
-			progress.name = snapname
-			progress.startAt = time.Now()
-			progress.hookPlan = plan
-			progress.state = SnapStarted
-		})
-		{
-			l := hooks.GetLogger(a.ctx).WithField("fs", fs.ToString()).WithField("snap", snapname)
-			l.WithField("report", plan.Report().String()).Debug("begin run job plan")
-			plan.Run(hooks.WithLogger(a.ctx, l), a.dryRun)
-			planReport = plan.Report()
-			fsHadErr = planReport.HadError() // not just fatal errors
-			if fsHadErr {
-				l.WithField("report", planReport.String()).Error("end run job plan with error")
-			} else {
-				l.WithField("report", planReport.String()).Info("end run job plan successful")
-			}
-		}
-
-	updateFSState:
-		anyFsHadErr = anyFsHadErr || fsHadErr
-		u(func(snapper *Snapper) {
-			progress.doneAt = time.Now()
-			progress.state = SnapDone
-			if fsHadErr {
-				progress.state = SnapError
-			}
-			progress.runResults = planReport
-		})
-	}
-
-	select {
-	case a.snapshotsTaken <- struct{}{}:
 	default:
-		if a.snapshotsTaken != nil {
+		return nil, fmt.Errorf("unknown snapshotting type %T", v)
-			a.log.Warn("callback channel is full, discarding snapshot update event")
-		}
-	}
-
-	for h, mc := range hookMatchCount {
-		if mc == 0 {
-			hookIdx := -1
-			for idx, ah := range *a.hooks {
-				if ah == h {
-					hookIdx = idx
-					break
-				}
-			}
-			a.log.WithField("hook", h.String()).WithField("hook_number", hookIdx+1).Warn("hook did not match any snapshotted filesystems")
-		}
-	}
-
-	return u(func(snapper *Snapper) {
-		if anyFsHadErr {
-			snapper.state = ErrorWait
-			snapper.err = errors.New("one or more snapshots could not be created, check logs for details")
-		} else {
-			snapper.state = Waiting
-			snapper.err = nil
-		}
-	}).sf()
-}
-
-func wait(a args, u updater) state {
-	var sleepUntil time.Time
-	u(func(snapper *Snapper) {
-		lastTick := snapper.lastInvocation
-		snapper.sleepUntil = lastTick.Add(a.interval)
-		sleepUntil = snapper.sleepUntil
-		log := a.log.WithField("sleep_until", sleepUntil).WithField("duration", a.interval)
-		logFunc := log.Debug
-		if snapper.state == ErrorWait || snapper.state == SyncUpErrWait {
-			logFunc = log.Error
-		}
-		logFunc("enter wait-state after error")
-	})
-
-	t := time.NewTimer(time.Until(sleepUntil))
-	defer t.Stop()
-
-	select {
-	case <-t.C:
-		return u(func(snapper *Snapper) {
-			snapper.state = Planning
-		}).sf()
-	case <-a.ctx.Done():
-		return onMainCtxDone(a.ctx, u)
 	}
 }
-
-func listFSes(ctx context.Context, mf *filters.DatasetMapFilter) (fss []*zfs.DatasetPath, err error) {
-	return zfs.ZFSListMapping(ctx, mf)
-}
-
-var syncUpWarnNoSnapshotUntilSyncupMinDuration = envconst.Duration("ZREPL_SNAPPER_SYNCUP_WARN_MIN_DURATION", 1*time.Second)
-
-// see docs/snapshotting.rst
-func findSyncPoint(log Logger, fss []*zfs.DatasetPath, prefix string, interval time.Duration) (syncPoint time.Time, err error) {
-
-	const (
-		prioHasVersions int = iota
-		prioNoVersions
-	)
-
-	type snapTime struct {
-		ds   *zfs.DatasetPath
-		prio int // lower is higher
-		time time.Time
-	}
-
-	if len(fss) == 0 {
-		return time.Now(), nil
-	}
-
-	snaptimes := make([]snapTime, 0, len(fss))
-	hardErrs := 0
-
-	now := time.Now()
-
-	log.Debug("examine filesystem state to find sync point")
-	for _, d := range fss {
-		l := log.WithField("fs", d.ToString())
-		syncPoint, err := findSyncPointFSNextOptimalSnapshotTime(l, now, interval, prefix, d)
-		if err == findSyncPointFSNoFilesystemVersionsErr {
-			snaptimes = append(snaptimes, snapTime{
-				ds:   d,
-				prio: prioNoVersions,
-				time: now,
-			})
-		} else if err != nil {
-			hardErrs++
-			l.WithError(err).Error("cannot determine optimal sync point for this filesystem")
-		} else {
-			l.WithField("syncPoint", syncPoint).Debug("found optimal sync point for this filesystem")
-			snaptimes = append(snaptimes, snapTime{
-				ds:   d,
-				prio: prioHasVersions,
-				time: syncPoint,
-			})
-		}
-	}
-
-	if hardErrs == len(fss) {
-		return time.Time{}, fmt.Errorf("hard errors in determining sync point for every matching filesystem")
-	}
-
-	if len(snaptimes) == 0 {
-		panic("implementation error: loop must either inc hardErrs or add result to snaptimes")
-	}
-
-	// sort ascending by (prio,time)
-	// => those filesystems with versions win over those without any
-	sort.Slice(snaptimes, func(i, j int) bool {
-		if snaptimes[i].prio == snaptimes[j].prio {
-			return snaptimes[i].time.Before(snaptimes[j].time)
-		}
-		return snaptimes[i].prio < snaptimes[j].prio
-	})
-
-	winnerSyncPoint := snaptimes[0].time
-	l := log.WithField("syncPoint", winnerSyncPoint.String())
-	l.Info("determined sync point")
-	if winnerSyncPoint.Sub(now) > syncUpWarnNoSnapshotUntilSyncupMinDuration {
-		for _, st := range snaptimes {
-			if st.prio == prioNoVersions {
-				l.WithField("fs", st.ds.ToString()).Warn("filesystem will not be snapshotted until sync point")
-			}
-		}
-	}
-
-	return snaptimes[0].time, nil
-
-}
-
-var findSyncPointFSNoFilesystemVersionsErr = fmt.Errorf("no filesystem versions")
-
-func findSyncPointFSNextOptimalSnapshotTime(l Logger, now time.Time, interval time.Duration, prefix string, d *zfs.DatasetPath) (time.Time, error) {
-
-	fsvs, err := zfs.ZFSListFilesystemVersions(context.TODO(), d, zfs.ListFilesystemVersionsOptions{
-		Types:           zfs.Snapshots,
-		ShortnamePrefix: prefix,
-	})
-	if err != nil {
-		return time.Time{}, errors.Wrap(err, "list filesystem versions")
-	}
-	if len(fsvs) <= 0 {
-		return time.Time{}, findSyncPointFSNoFilesystemVersionsErr
-	}
-
-	// 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 {
-		return time.Time{}, fmt.Errorf("snapshot %q is from the future: creation=%q now=%q", latest.ToAbsPath(d), latest.Creation, now)
-	}
-
-	return latest.Creation.Add(interval), nil
-}

daemon/snapper/snapper_all.go

@@ -1,48 +0,0 @@
-package snapper
-
-import (
-	"context"
-	"fmt"
-
-	"github.com/zrepl/zrepl/config"
-	"github.com/zrepl/zrepl/daemon/filters"
-)
-
-// FIXME: properly abstract snapshotting:
-//   - split up things that trigger snapshotting from the mechanism
-//     - timer-based trigger (periodic)
-//     - call from control socket (manual)
-//     - mixed modes?
-//   - support a `zrepl snapshot JOBNAME` subcommand for config.SnapshottingManual
-type PeriodicOrManual struct {
-	s *Snapper
-}
-
-func (s *PeriodicOrManual) Run(ctx context.Context, wakeUpCommon chan<- struct{}) {
-	if s.s != nil {
-		s.s.Run(ctx, wakeUpCommon)
-	}
-}
-
-// Returns nil if manual
-func (s *PeriodicOrManual) Report() *Report {
-	if s.s != nil {
-		return s.s.Report()
-	}
-	return nil
-}
-
-func FromConfig(g *config.Global, fsf *filters.DatasetMapFilter, in config.SnapshottingEnum) (*PeriodicOrManual, error) {
-	switch v := in.Ret.(type) {
-	case *config.SnapshottingPeriodic:
-		snapper, err := PeriodicFromConfig(g, fsf, v)
-		if err != nil {
-			return nil, err
-		}
-		return &PeriodicOrManual{snapper}, nil
-	case *config.SnapshottingManual:
-		return &PeriodicOrManual{}, nil
-	default:
-		return nil, fmt.Errorf("unknown snapshotting type %T", v)
-	}
-}

daemon/snapper/snapper_report.go

@@ -1,118 +0,0 @@
-package snapper
-
-import (
-	"fmt"
-	"sort"
-	"strings"
-	"time"
-
-	"github.com/zrepl/zrepl/daemon/hooks"
-)
-
-type Report struct {
-	State State
-	// valid in state SyncUp and Waiting
-	SleepUntil time.Time
-	// valid in state Err
-	Error string
-	// valid in state Snapshotting
-	Progress []*ReportFilesystem
-}
-
-type ReportFilesystem struct {
-	Path  string
-	State SnapState
-
-	// Valid in SnapStarted and later
-	SnapName      string
-	StartAt       time.Time
-	Hooks         string
-	HooksHadError bool
-
-	// Valid in SnapDone | SnapError
-	DoneAt time.Time
-}
-
-func errOrEmptyString(e error) string {
-	if e != nil {
-		return e.Error()
-	}
-	return ""
-}
-
-func (s *Snapper) Report() *Report {
-	s.mtx.Lock()
-	defer s.mtx.Unlock()
-
-	pReps := make([]*ReportFilesystem, 0, len(s.plan))
-	for fs, p := range s.plan {
-		var hooksStr string
-		var hooksHadError bool
-		if p.hookPlan != nil {
-			hr := p.hookPlan.Report()
-			// FIXME: technically this belongs into client
-			// but we can't serialize hooks.Step ATM
-			rightPad := func(str string, length int, pad string) string {
-				if len(str) > length {
-					return str[:length]
-				}
-				return str + strings.Repeat(pad, length-len(str))
-			}
-			hooksHadError = hr.HadError()
-			rows := make([][]string, len(hr))
-			const numCols = 4
-			lens := make([]int, numCols)
-			for i, e := range hr {
-				rows[i] = make([]string, numCols)
-				rows[i][0] = fmt.Sprintf("%d", i+1)
-				rows[i][1] = e.Status.String()
-				runTime := "..."
-				if e.Status != hooks.StepPending {
-					runTime = e.End.Sub(e.Begin).Round(time.Millisecond).String()
-				}
-				rows[i][2] = runTime
-				rows[i][3] = ""
-				if e.Report != nil {
-					rows[i][3] = e.Report.String()
-				}
-				for j, col := range lens {
-					if len(rows[i][j]) > col {
-						lens[j] = len(rows[i][j])
-					}
-				}
-			}
-			rowsFlat := make([]string, len(hr))
-			for i, r := range rows {
-				colsPadded := make([]string, len(r))
-				for j, c := range r[:len(r)-1] {
-					colsPadded[j] = rightPad(c, lens[j], " ")
-				}
-				colsPadded[len(r)-1] = r[len(r)-1]
-				rowsFlat[i] = strings.Join(colsPadded, " ")
-			}
-			hooksStr = strings.Join(rowsFlat, "\n")
-		}
-		pReps = append(pReps, &ReportFilesystem{
-			Path:          fs.ToString(),
-			State:         p.state,
-			SnapName:      p.name,
-			StartAt:       p.startAt,
-			DoneAt:        p.doneAt,
-			Hooks:         hooksStr,
-			HooksHadError: hooksHadError,
-		})
-	}
-
-	sort.Slice(pReps, func(i, j int) bool {
-		return strings.Compare(pReps[i].Path, pReps[j].Path) == -1
-	})
-
-	r := &Report{
-		State:      s.state,
-		SleepUntil: s.sleepUntil,
-		Error:      errOrEmptyString(s.err),
-		Progress:   pReps,
-	}
-
-	return r
-}

daemon/snapper/util.go

@@ -0,0 +1,21 @@
+package snapper
+
+import (
+	"context"
+
+	"github.com/zrepl/zrepl/daemon/logging"
+	"github.com/zrepl/zrepl/logger"
+)
+
+type Logger = logger.Logger
+
+func getLogger(ctx context.Context) Logger {
+	return logging.GetLogger(ctx, logging.SubsysSnapshot)
+}
+
+func errOrEmptyString(e error) string {
+	if e != nil {
+		return e.Error()
+	}
+	return ""
+}

debian

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

dist/grafana/grafana-prometheus-zrepl-0.1.json

@@ -1,595 +0,0 @@
-{
-  "annotations": {
-    "list": [
-      {
-        "$$hashKey": "object:3351",
-        "builtIn": 1,
-        "datasource": "-- Grafana --",
-        "enable": true,
-        "hide": true,
-        "iconColor": "rgba(0, 211, 255, 1)",
-        "name": "Annotations & Alerts",
-        "type": "dashboard"
-      }
-    ]
-  },
-  "editable": true,
-  "gnetId": null,
-  "graphTooltip": 1,
-  "id": 7,
-  "iteration": 1552746418826,
-  "links": [],
-  "panels": [
-    {
-      "content": "# zrepl Prometheus Metrics\n\nzrepl exposes Prometheus metrics and ships with this Grafana dashboard.\nThe exported metrics are suitable for health checks:\n\n* The log should generally be warning & error-free\n  * The `Log Messages that require attention` graph visualizes log message counts indicating problems.\n* The number of goroutines should not grow unboundedly over time.\n  * During replication, the number of goroutines can be way higher than during idle time.\n  * If the goroutine count grows with each replication, there is clearly a goroutine leak. Please open a bug report.\n* The sys memory consumption should not grow unboundedly over time.\n  * Note that the Go runtime pre-allocates some of its heap from the OS.\n  * zrepl actually uses much less memory than allocated from the OS.\n  * Since Go 1.11, Go pre-allocates more aggressively.\n* Monitor that some data is replicated, although that metric does not guarantee that replication was successful.\n\n**In general, note that the exported metrics are not stable unless declared otherwise.**",
-      "gridPos": {
-        "h": 9,
-        "w": 24,
-        "x": 0,
-        "y": 0
-      },
-      "id": 35,
-      "mode": "markdown",
-      "title": "Panel Title",
-      "type": "text"
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": null,
-      "fill": 1,
-      "gridPos": {
-        "h": 9,
-        "w": 12,
-        "x": 0,
-        "y": 9
-      },
-      "id": 15,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": true,
-      "steppedLine": false,
-      "targets": [
-        {
-          "$$hashKey": "object:3436",
-          "expr": "up{job='$prom_job_name'}",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "zrepl Instances Up",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": "5",
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": true
-        }
-      ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": null,
-      "fill": 1,
-      "gridPos": {
-        "h": 9,
-        "w": 12,
-        "x": 12,
-        "y": 9
-      },
-      "id": 22,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": false,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "increase(zrepl_daemon_log_entries{job='$prom_job_name',level=~'warn|error'}[$__interval])",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Log Messages that require attention",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": true
-        }
-      ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": null,
-      "fill": 1,
-      "gridPos": {
-        "h": 9,
-        "w": 12,
-        "x": 0,
-        "y": 18
-      },
-      "id": 33,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [
-        {
-          "alias": "/replicated bytes in last.*/",
-          "yaxis": 2
-        }
-      ],
-      "spaceLength": 10,
-      "stack": false,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "sum(rate(zrepl_replication_bytes_replicated{job='$prom_job_name'}[$__interval])) by (zrepl_job)",
-          "format": "time_series",
-          "hide": false,
-          "interval": "",
-          "intervalFactor": 1,
-          "legendFormat": "replication data rate zrepl_job={{zrepl_job}}",
-          "refId": "A"
-        },
-        {
-          "expr": "sum(increase(zrepl_replication_bytes_replicated{job='$prom_job_name'}[10m])) by (zrepl_job)",
-          "format": "time_series",
-          "hide": false,
-          "interval": "",
-          "intervalFactor": 1,
-          "legendFormat": "replicated bytes in last 10min zrepl_job={{zrepl_job}}",
-          "refId": "B"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Replication Data Rate and Volume(integrates last 10min)",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "format": "Bps",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": true
-        },
-        {
-          "format": "bytes",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": true
-        }
-      ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": null,
-      "fill": 1,
-      "gridPos": {
-        "h": 9,
-        "w": 12,
-        "x": 12,
-        "y": 18
-      },
-      "id": 23,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": false,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "sum(increase(zrepl_daemon_log_entries{job='$prom_job_name',zrepl_job=~\"^[^_].*\"}[$__interval])) by (instance,zrepl_job)",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Log Activity (without internal jobs)",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": true
-        }
-      ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": null,
-      "fill": 1,
-      "gridPos": {
-        "h": 9,
-        "w": 12,
-        "x": 0,
-        "y": 27
-      },
-      "id": 17,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": false,
-      "steppedLine": false,
-      "targets": [
-        {
-          "$$hashKey": "object:3535",
-          "expr": "go_memstats_sys_bytes{job='$prom_job_name'}",
-          "format": "time_series",
-          "hide": false,
-          "intervalFactor": 1,
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "Memory Allocated by the Go runtime from the OS (should not grow unboundedly)",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "format": "bytes",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": true
-        }
-      ]
-    },
-    {
-      "aliasColors": {},
-      "bars": false,
-      "dashLength": 10,
-      "dashes": false,
-      "datasource": null,
-      "fill": 1,
-      "gridPos": {
-        "h": 9,
-        "w": 12,
-        "x": 12,
-        "y": 27
-      },
-      "id": 19,
-      "legend": {
-        "avg": false,
-        "current": false,
-        "max": false,
-        "min": false,
-        "show": true,
-        "total": false,
-        "values": false
-      },
-      "lines": true,
-      "linewidth": 1,
-      "links": [],
-      "nullPointMode": "null",
-      "percentage": false,
-      "pointradius": 5,
-      "points": false,
-      "renderer": "flot",
-      "seriesOverrides": [],
-      "spaceLength": 10,
-      "stack": false,
-      "steppedLine": false,
-      "targets": [
-        {
-          "expr": "go_goroutines{job='$prom_job_name'}",
-          "format": "time_series",
-          "intervalFactor": 1,
-          "refId": "A"
-        }
-      ],
-      "thresholds": [],
-      "timeFrom": null,
-      "timeShift": null,
-      "title": "number of goroutines (should not grow unboundedly)",
-      "tooltip": {
-        "shared": true,
-        "sort": 0,
-        "value_type": "individual"
-      },
-      "type": "graph",
-      "xaxis": {
-        "buckets": null,
-        "mode": "time",
-        "name": null,
-        "show": true,
-        "values": []
-      },
-      "yaxes": [
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": "0",
-          "show": true
-        },
-        {
-          "format": "short",
-          "label": null,
-          "logBase": 1,
-          "max": null,
-          "min": null,
-          "show": true
-        }
-      ]
-    }
-  ],
-  "refresh": "1m",
-  "schemaVersion": 16,
-  "style": "dark",
-  "tags": [],
-  "templating": {
-    "list": [
-      {
-        "allValue": null,
-        "current": {
-          "text": "zrepl",
-          "value": "zrepl"
-        },
-        "datasource": "prometheus",
-        "hide": 0,
-        "includeAll": false,
-        "label": "Prometheus Job Name",
-        "multi": false,
-        "name": "prom_job_name",
-        "options": [],
-        "query": "label_values(up, job)",
-        "refresh": 1,
-        "regex": "",
-        "sort": 1,
-        "tagValuesQuery": "",
-        "tags": [],
-        "tagsQuery": "",
-        "type": "query",
-        "useTags": false
-      }
-    ]
-  },
-  "time": {
-    "from": "now-2d",
-    "to": "now"
-  },
-  "timepicker": {
-    "refresh_intervals": [
-      "5s",
-      "10s",
-      "30s",
-      "1m",
-      "5m",
-      "15m",
-      "30m",
-      "1h",
-      "2h",
-      "1d"
-    ],
-    "time_options": [
-      "5m",
-      "15m",
-      "1h",
-      "6h",
-      "12h",
-      "24h",
-      "2d",
-      "7d",
-      "30d"
-    ]
-  },
-  "timezone": "",
-  "title": "zrepl 0.1",
-  "uid": "xTljn4qmk",
-  "version": 6
-}

dist/openrc/zrepl

@@ -0,0 +1,23 @@
+#!/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

@@ -9,6 +9,9 @@ ExecStart=/usr/local/bin/zrepl --config /etc/zrepl/zrepl.yml daemon
 RuntimeDirectory=zrepl zrepl/stdinserver
 RuntimeDirectoryMode=0700
 
+# Make Go produce coredumps
+Environment=GOTRACEBACK='crash'
+
 ProtectSystem=strict
 #PrivateDevices=yes # TODO ZFS needs access to /dev/zfs, could we limit this?
 ProtectKernelTunables=yes
@@ -27,7 +30,8 @@ ProtectHome=read-only
 # SystemCallFilter
 #   ~@privileged doesn't work with Ubuntu 18.04 ssh
 SystemCallFilter=~ @mount @cpu-emulation @keyring @module @obsolete @raw-io @debug @clock @resources
-
+# Go1.19 added automatic RLIMIT_NOFILE changes, so, we need to allow that
+SystemCallFilter= setrlimit
 
 [Install]
 WantedBy=multi-user.target