WIP

generalize trigger kinds
WIP: generic activation through + new interval-based replication trigger
2024-05-09 13:25:31 +00:00 · 2023-12-22 14:40:53 +00:00 · 2023-12-22 14:01:24 +00:00 · 2023-12-22 13:38:14 +01:00 · 2023-11-01 00:12:54 +01:00 · 2023-11-01 00:12:23 +01:00
97 changed files with 5686 additions and 2545 deletions
@@ -1,5 +1,8 @@
 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:
@@ -24,18 +27,12 @@ commands:

  apt-update-and-install-common-deps:
    steps:
-      - run: sudo apt update && sudo apt install gawk make
-
-  restore-cache-gomod:
-    steps:
-      - restore_cache:
-          key: go-mod-v4-{{ checksum "go.sum" }}
-  save-cache-gomod:
-    steps:
-      - save_cache:
-          key: go-mod-v4-{{ checksum "go.sum" }}
-          paths:
-            - "/go/pkg/mod"
+      - 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:
@@ -50,94 +47,48 @@ commands:
      - invoke-lazy-sh:
          subcommand: docdep

-  download-and-install-minio-client:
-    steps:
-      - setup-home-local-bin
-      - restore_cache:
-          key: minio-client-v2
-      - run:
-          shell: /bin/bash -eo pipefail
-          command: |
-            if which mc; then exit 0; fi
-            sudo curl -sSL https://dl.min.io/client/mc/release/linux-amd64/archive/mc.RELEASE.2020-08-20T00-23-01Z \
-              -o "$HOME/.local/bin/mc"
-            sudo chmod +x "$HOME/.local/bin/mc"
-      - save_cache:
-          key: minio-client-v2
-          paths:
-            - "$HOME/.local/bin/mc"
-
-  upload-minio:
+  docs-publish-sh:
    parameters:
-      src:
-        type: string
-      dst:
-        type: string
+      push:
+        type: boolean
    steps:
+      - checkout
      - run:
-          shell: /bin/bash -eo pipefail
-          when: always
          command: |
-            if [ -n "$CIRCLE_PR_NUMBER" ]; then # CIRCLE_PR_NUMBER is guaranteed to be only present in forked PRs (external)
-              echo "Forked PR detected. Sry, can't trust you with credentials to external artifact store, use CircleCI's instead."
-              exit 0
-            fi
-            set -u # from now on
+            git config --global user.email "zreplbot@cschwarz.com"
+            git config --global user.name "zrepl-github-io-ci"

-            mc config host add --api s3v4 zrepl-minio https://minio.cschwarz.com ${MINIO_ACCESS_KEY} ${MINIO_SECRET_KEY}
+      # if we're pushing, we need to add the deploy key
+      # which is stored as "Additional SSH Keys" in the CircleCI project settings.
+      # We can't use the CircleCI-manage deploy key because we're pushing
+      # to a different repo than the one we're building.
+      - when:
+          condition: << parameters.push >>
+          steps:
+            # 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"

-            # keep in sync with set-github-minio-status
-            jobprefix=zrepl-ci-artifacts/${CIRCLE_SHA1}-pipeline-<<pipeline.number>>/${CIRCLE_JOB}
-
-            # Upload artifacts
-            mkdir -p ./artifacts
-            mc cp -r <<parameters.src>> "zrepl-minio/$jobprefix/<<parameters.dst>>"
-
-  set-github-minio-status:
-    parameters:
-      context:
-        type: string
-      description:
-        type: string
-      minio-dst:
-        type: string
-    steps:
-      - run:
-          shell: /bin/bash -eo pipefail
-          command: |
-            if [ -n "$CIRCLE_PR_NUMBER" ]; then # CIRCLE_PR_NUMBER is guaranteed to be only present in forked PRs (external)
-              echo "Forked PR detected. Sry, can't trust you with credentials to external artifact store, use CircleCI's instead."
-              exit 0
-            fi
-            set -u # from now on
-
-            # keep in sync with with upload-minio command
-            jobprefix=zrepl-ci-artifacts/${CIRCLE_SHA1}-pipeline-<<pipeline.number>>/${CIRCLE_JOB}
-            # Push Artifact Link to GitHub
-            REPO="zrepl/zrepl"
-            COMMIT="${CIRCLE_SHA1}"
-            JOB_NAME="${CIRCLE_JOB}"
-            CONTEXT="<<parameters.context>>"
-            DESCRIPTION="<<parameters.description>>"
-            TARGETURL=https://minio.cschwarz.com/minio/"$jobprefix"/"<<parameters.minio-dst>>"
-            curl "https://api.github.com/repos/$REPO/statuses/$COMMIT" \
-              -H "Content-Type: application/json" \
-              -H "Authorization: token $GITHUB_COMMIT_STATUS_TOKEN" \
-              -X POST \
-              -d '{"context":"'"$CONTEXT"'", "state": "success", "description":"'"$DESCRIPTION"'", "target_url":"'"$TARGETURL"'"}'
-
-
-  trigger-pipeline:
-    parameters:
-      body_no_shell_subst:
-        type: string
-    steps:
-      - run: |
-          curl -X POST https://circleci.com/api/v2/project/github/zrepl/zrepl/pipeline \
-            -H 'Content-Type: application/json' \
-            -H 'Accept: application/json' \
-            -H "Circle-Token: $ZREPL_BOT_CIRCLE_TOKEN" \
-            --data '<<parameters.body_no_shell_subst>>'
+      # 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:
@@ -150,7 +101,7 @@ parameters:

  release_docker_baseimage_tag:
    type: string
-    default: "1.17"
+    default: "1.21"

 workflows:
  version: 2
@@ -160,19 +111,19 @@ workflows:
    jobs:
      - quickcheck-docs
      - quickcheck-go: &quickcheck-go-smoketest
-          name: quickcheck-go-amd64-linux-1.17
-          goversion: &latest-go-release "1.17"
+          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-1.17 #quickcheck-go-smoketest.name
+            - quickcheck-go-amd64-linux-1.21 #quickcheck-go-smoketest.name
          matrix: &quickcheck-go-matrix
            alias: quickcheck-go-matrix
            parameters:
-              goversion: [*latest-go-release, "1.12"]
+              goversion: [*latest-go-release, "1.20"]
              goos: ["linux", "freebsd"]
              goarch: ["amd64", "arm64"]
            exclude:
@@ -180,10 +131,14 @@ workflows:
              - goversion: *latest-go-release
                goos: linux
                goarch: amd64
-              # not supported by Go 1.12
-              - goversion: "1.12"
-                goos: freebsd
-                goarch: arm64
+      - 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 >>
@@ -201,20 +156,7 @@ workflows:
            - release-deb
            - release-rpm

-  periodic:
-    triggers:
-      - schedule:
-          cron: "00 17 * * *"
-          filters:
-            branches:
-              only:
-                - master
-                - stable
-                - problame/circleci-build
-    jobs:
-      - periodic-full-pipeline-run
-
-  zrepl.github.io:
+  publish-zrepl.github.io:
    jobs:
      - publish-zrepl-github-io:
          filters:
@@ -225,20 +167,15 @@ workflows:
 jobs:
  quickcheck-docs:
    docker:
-      - image: cimg/base:2020.08
+      - image: cimg/base:2023.09
    steps:
      - checkout
      - install-docdep
+      # do the current docs build
      - run: make docs
-
-      - download-and-install-minio-client
-      - upload-minio:
-          src: artifacts
-          dst: ""
-      - set-github-minio-status:
-          context: artifacts/${CIRCLE_JOB}
-          description: artifacts of CI job ${CIRCLE_JOB}
-          minio-dst: ""
+      # does the publish.sh script still work?
+      - docs-publish-sh:
+          push: false

  quickcheck-go:
    parameters:
@@ -249,7 +186,7 @@ jobs:
      goarch:
        type: string
    docker:
-      - image: circleci/golang:<<parameters.goversion>>
+      - image: cimg/go:<<parameters.goversion>>
    environment:
      GOOS: <<parameters.goos>>
      GOARCH: <<parameters.goarch>>
@@ -257,41 +194,62 @@ jobs:
    steps:
      - checkout

-      - restore-cache-gomod
+      - go/load-cache:
+          key: quickcheck-<<parameters.goversion>>
+      - install-godep
      - run: go mod download
      - run: cd build && go mod download
-      - save-cache-gomod
+      - go/save-cache:
+          key: quickcheck-<<parameters.goversion>>

-      - install-godep
      - run: make formatcheck
      - run: make generate-platform-test-list
      - run: make zrepl-bin test-platform-bin
      - run: make vet
      - run: make lint

-      - download-and-install-minio-client
      - run: rm -f artifacts/generate-platform-test-list
      - store_artifacts:
          path: artifacts
-      - upload-minio:
-          src: artifacts
-          dst: ""
-      - set-github-minio-status:
-          context: artifacts/${CIRCLE_JOB}
-          description: artifacts of CI job ${CIRCLE_JOB}
-          minio-dst: ""
+      - persist_to_workspace:
+          root: .
+          paths: [.]
+
+  platformtest:
+    parameters:
+      goversion:
+        type: string
+      goos:
+        type: string
+      goarch:
+        type: string
+    machine:
+      image: ubuntu-2204:current
+    resource_class: medium
+    environment:
+      GOOS: <<parameters.goos>>
+      GOARCH: <<parameters.goarch>>
+    steps:
+      - attach_workspace:
+          at: .
+      - run: sudo apt-get update
+      - run: sudo apt-get install -y zfsutils-linux
+      - run: sudo zfs version
+      - run: sudo make test-platform GOOS="$GOOS" GOARCH="$GOARCH"

  test-go-on-latest-go-release:
    parameters:
      goversion:
        type: string
    docker:
-      - image: circleci/golang:<<parameters.goversion>>
+      - image: cimg/go:<<parameters.goversion>>
    steps:
      - checkout
-      - restore-cache-gomod
+      - go/load-cache:
+          key: make-test-go
      - run: make test-go
-      # don't save-cache-gomod here, test-go doesn't pull all the dependencies
+      - go/save-cache:
+          key: make-test-go

  release-build:
    machine:
@@ -332,48 +290,15 @@ jobs:
    steps:
      - attach_workspace:
          at: .
+      - run: make wrapup-and-checksum
      - store_artifacts:
          path: artifacts
-      - download-and-install-minio-client
-      - upload-minio:
-          src: artifacts
-          dst: ""
-      - set-github-minio-status:
-          context: artifacts/release
-          description: CI-generated release artifacts
-          minio-dst: ""
-
-  periodic-full-pipeline-run:
-    docker:
-      - image: cimg/base:2020.08
-    steps:
-      - trigger-pipeline:
-          body_no_shell_subst: '{"branch":"<<pipeline.git.branch>>", "parameters": { "do_ci": true, "do_release": true }}'

  publish-zrepl-github-io:
    docker:
-      - image: cimg/python:3.7
+      - image: cimg/base:2023.09
    steps:
      - checkout
-      - invoke-lazy-sh:
-          subcommand: docdep
-      - 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"
-
-      - run: bash -x docs/publish.sh -c -a
+      - install-docdep
+      - docs-publish-sh:
+          push: true
@@ -0,0 +1,84 @@
+import argparse
+from pathlib import Path
+import re
+
+import requests
+
+import time
+
+import os
+import argparse
+from pathlib import Path
+
+
+circle_token = os.environ.get('CIRCLE_TOKEN')
+if not circle_token:
+    raise ValueError('CIRCLE_TOKEN environment variable not set')
+
+parser = argparse.ArgumentParser(description='Download artifacts from CircleCI')
+parser.add_argument('build_num', type=str, help='Build number')
+parser.add_argument('dst', type=Path, help='Destination directory')
+parser.add_argument('--prefix', type=str, default='', help='Filter for prefix')
+parser.add_argument('--match', type=str, default='.*', help='Only include paths matching the given regex')
+
+args = parser.parse_args()
+
+res = requests.get(
+    f"https://circleci.com/api/v1.1/project/github/zrepl/zrepl/{args.build_num}/artifacts",
+    headers={
+        "Circle-Token": circle_token,
+    },
+)
+res.raise_for_status()
+
+# https://circleci.com/docs/api/v1/index.html#artifacts-of-a-job
+# [ {
+#   "path" : "raw-test-output/go-test-report.xml",
+#   "pretty_path" : "raw-test-output/go-test-report.xml",
+#   "node_index" : 0,
+#   "url" : "https://24-88881093-gh.circle-artifacts.com/0/raw-test-output/go-test-report.xml"
+# }, {
+#   "path" : "raw-test-output/go-test.out",
+#   "pretty_path" : "raw-test-output/go-test.out",
+#   "node_index" : 0,
+#   "url" : "https://24-88881093-gh.circle-artifacts.com/0/raw-test-output/go-test.out"
+# } ]
+res = res.json()
+
+for artifact in res:
+    if not artifact["pretty_path"].startswith(args.prefix):
+        continue
+    if not re.match(args.match, artifact["pretty_path"]):
+        continue
+    stripped = artifact["pretty_path"][len(args.prefix):]
+    print(f"Downloading {artifact['pretty_path']} to {args.dst / stripped}")
+    artifact_rel = Path(stripped)
+    artifact_dst = args.dst / artifact_rel
+    artifact_dst.parent.mkdir(parents=True, exist_ok=True)
+
+    res = requests.get(
+        artifact["url"],
+        headers={
+            "Circle-Token": circle_token,
+        },
+        stream=True,
+    )
+    res.raise_for_status()
+
+    total_size = int(res.headers.get("Content-Length", 0))
+    block_size = 128 * 1024
+    with open(artifact_dst, "wb") as f:
+        progress = 0
+        start_time = time.time()
+        for chunk in res.iter_content(chunk_size=block_size):
+            f.write(chunk)
+            progress += len(chunk)
+            percent = progress / total_size * 100
+            elapsed_time = time.time() - start_time
+            if elapsed_time >= 5:
+                print(f"Downloaded {progress}/{total_size} bytes ({percent:.2f}%)", end="\r")
+                start_time = time.time()
+        print(f"Downloaded {progress}/{total_size} bytes ({percent:.2f}%)")
+    print("Download complete!")
+
+print("All files downloaded")
@@ -1,5 +1,5 @@
 .PHONY: generate build test vet cover release docs docs-clean clean format lint platformtest
-.PHONY: release bins-all release-noarch
+.PHONY: release release-noarch
 .DEFAULT_GOAL := zrepl-bin

 ARTIFACTDIR := artifacts
@@ -14,13 +14,15 @@ ifndef _ZREPL_VERSION
    endif
 endif

+ZREPL_PACKAGE_RELEASE := 1
+
 GO := go
 GOOS ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOOS"')
 GOARCH ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOARCH"')
 GOARM ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOARM"')
 GOHOSTOS ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOHOSTOS"')
 GOHOSTARCH ?= $(shell bash -c 'source <($(GO) env) && echo "$$GOHOSTARCH"')
-GO_ENV_VARS := GO111MODULE=on
+GO_ENV_VARS := GO111MODULE=on CGO_ENABLED=0
 GO_LDFLAGS := "-X github.com/zrepl/zrepl/version.zreplVersion=$(_ZREPL_VERSION)"
 GO_MOD_READONLY := -mod=readonly
 GO_EXTRA_BUILDFLAGS :=
@@ -28,7 +30,7 @@ GO_BUILDFLAGS := $(GO_MOD_READONLY) $(GO_EXTRA_BUILDFLAGS)
 GO_BUILD := $(GO_ENV_VARS) $(GO) build $(GO_BUILDFLAGS) -ldflags $(GO_LDFLAGS)
 GOLANGCI_LINT := golangci-lint
 GOCOVMERGE := gocovmerge
-RELEASE_DOCKER_BASEIMAGE_TAG ?= 1.17
+RELEASE_DOCKER_BASEIMAGE_TAG ?= 1.21
 RELEASE_DOCKER_BASEIMAGE ?= golang:$(RELEASE_DOCKER_BASEIMAGE_TAG)

 ifneq ($(GOARM),)
@@ -50,21 +52,20 @@ printvars:
 release: clean
 	# no cross-platform support for target test
 	$(MAKE) test-go
-	$(MAKE) bins-all
+	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="vet"
+	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="lint"
+	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="zrepl-bin"
+	$(MAKE) _run_make_foreach_target_tuple RUN_MAKE_FOREACH_TARGET_TUPLE_ARG="test-platform-bin"
 	$(MAKE) noarch
-	$(MAKE) wrapup-and-checksum
-	$(MAKE) check-git-clean
-ifeq (SIGN, 1)
-	$(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=$(GOARM)
+		make release \
+			GOOS=$(GOOS) GOARCH=$(GOARCH) GOARM=$(GOARM) \
+			ZREPL_VERSION=$(ZREPL_VERSION) ZREPL_PACKAGE_RELEASE=$(ZREPL_PACKAGE_RELEASE)

 debs-docker:
 	$(MAKE) _debs_or_rpms_docker _DEB_OR_RPM=deb
@@ -81,9 +82,14 @@ rpm: $(ARTIFACTDIR) # artifacts/_zrepl.zsh_completion artifacts/bash_completion
 	$(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 >  $(_ZREPL_RPM_TOPDIR_ABS)/SPECS/zrepl.spec
+	for d in BUILD BUILDROOT RPMS SOURCES SPECS SRPMS; do \
+		mkdir -p "$(_ZREPL_RPM_TOPDIR_ABS)/$$d"; \
+	done
+	sed \
+		-e "s/^Version:.*/Version:          $(_ZREPL_RPM_VERSION)/g" \
+		-e "s/^Release:.*/Release:          $(ZREPL_PACKAGE_RELEASE)/g" \
+	    packaging/rpm/zrepl.spec \
+		>  $(_ZREPL_RPM_TOPDIR_ABS)/SPECS/zrepl.spec

 	# see /usr/lib/rpm/platform
 ifeq ($(GOARCH),amd64)
@@ -110,13 +116,16 @@ 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=$(GOARM)
+		make rpm \
+			GOOS=$(GOOS) GOARCH=$(GOARCH) GOARM=$(GOARM) \
+			ZREPL_VERSION=$(ZREPL_VERSION) ZREPL_PACKAGE_RELEASE=$(ZREPL_PACKAGE_RELEASE)
+

 deb: $(ARTIFACTDIR) # artifacts/_zrepl.zsh_completion artifacts/bash_completion docs zrepl-bin

 	cp packaging/deb/debian/changelog.template packaging/deb/debian/changelog
 	sed -i 's/DATE_DASH_R_OUTPUT/$(shell date -R)/' packaging/deb/debian/changelog
-	VERSION="$(subst -,.,$(_ZREPL_VERSION))"; \
+	VERSION="$(subst -,.,$(_ZREPL_VERSION))-$(ZREPL_PACKAGE_RELEASE)"; \
 		export VERSION="$${VERSION#v}"; \
 		sed -i 's/VERSION/'"$$VERSION"'/' packaging/deb/debian/changelog

@@ -134,11 +143,19 @@ endif

 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=$(GOARM)
+		make deb \
+			GOOS=$(GOOS) GOARCH=$(GOARCH) GOARM=$(GOARM) \
+			ZREPL_VERSION=$(ZREPL_VERSION) ZREPL_PACKAGE_RELEASE=$(ZREPL_PACKAGE_RELEASE)

-# expects `release` target to have run before
+# expects `release`, `deb` & `rpm` targets to have run before
 NOARCH_TARBALL := $(ARTIFACTDIR)/zrepl-noarch.tar
 wrapup-and-checksum:
 	rm -f $(NOARCH_TARBALL)
@@ -155,7 +172,7 @@ wrapup-and-checksum:
 		config/samples
 	rm -rf "$(ARTIFACTDIR)/release"
 	mkdir -p "$(ARTIFACTDIR)/release"
-	cp -l $(ARTIFACTDIR)/zrepl-* \
+	cp -l $(ARTIFACTDIR)/zrepl* \
 		$(ARTIFACTDIR)/platformtest-* \
 		"$(ARTIFACTDIR)/release"
 	cd "$(ARTIFACTDIR)/release" && sha512sum $$(ls | sort) > sha512sum.txt
@@ -175,39 +192,42 @@ check-git-clean:

 tag-release:
 	test -n "$(ZREPL_TAG_VERSION)" || exit 1
-	git tag -u E27CA5FC -m "$(ZREPL_TAG_VERSION)" "$(ZREPL_TAG_VERSION)"
+	git tag -u '328A6627FA98061D!' -m "$(ZREPL_TAG_VERSION)" "$(ZREPL_TAG_VERSION)"

 sign:
-	gpg -u "89BC 5D89 C845 568B F578  B306 CDBD 8EC8 E27C A5FC" \
+	gpg -u '328A6627FA98061D!' \
 		--armor \
 		--detach-sign $(ARTIFACTDIR)/release/sha512sum.txt

 clean: docs-clean
 	rm -rf "$(ARTIFACTDIR)"

-##################### BINARIES #####################
-.PHONY: bins-all lint test-go test-platform cover-merge cover-html vet zrepl-bin test-platform-bin generate-platform-test-list
+download-circleci-release:
+	rm -rf "$(ARTIFACTDIR)"
+	mkdir -p "$(ARTIFACTDIR)/release"
+	python3 .circleci/download_artifacts.py --prefix 'artifacts/release/' "$(BUILD_NUM)" "$(ARTIFACTDIR)/release"

-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
-	$(MAKE) $(BINS_ALL_TARGETS) GOOS=linux     GOARCH=386
-	$(MAKE) $(BINS_ALL_TARGETS) GOOS=darwin    GOARCH=amd64
-	$(MAKE) $(BINS_ALL_TARGETS) GOOS=solaris   GOARCH=amd64
-ifeq ($(GO_SUPPORTS_ILLUMOS), illumos)
-	$(MAKE) $(BINS_ALL_TARGETS) GOOS=illumos   GOARCH=amd64
-else ifeq ($(GO_SUPPORTS_ILLUMOS), noillumos)
-	@echo "SKIPPING ILLUMOS BUILD BECAUSE GO VERSION DOESN'T SUPPORT IT"
-else
-	@echo "CANNOT DETERMINE WHETHER GO VERSION SUPPORTS GOOS=illumos"; exit 1
-endif
+##################### MULTI-ARCH HELPERS #####################
+
+_run_make_foreach_target_tuple:
+	if [ "$(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG)" = "" ]; then \
+		echo "RUN_MAKE_FOREACH_TARGET_TUPLE_ARG must be set"; \
+		exit 1; \
+	fi
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=amd64
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=386
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=arm	  GOARM=7
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=freebsd   GOARCH=arm64
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=amd64
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=arm64
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=arm     GOARM=7
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=linux     GOARCH=386
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=darwin    GOARCH=amd64
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=solaris   GOARCH=amd64
+	$(MAKE) $(RUN_MAKE_FOREACH_TARGET_TUPLE_ARG) GOOS=illumos   GOARCH=amd64
+
+##################### REGULAR TARGETS #####################
+.PHONY: lint test-go test-platform cover-merge cover-html vet zrepl-bin test-platform-bin generate-platform-test-list

 lint:
 	$(GO_ENV_VARS) $(GOLANGCI_LINT) run ./...
@@ -2,12 +2,18 @@ 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
@@ -4,13 +4,37 @@ go 1.12

 require (
 	github.com/alvaroloes/enumer v1.1.1
-	github.com/golangci/golangci-lint v1.35.2
-	github.com/golangci/misspell v0.3.4 // indirect
-	github.com/golangci/revgrep v0.0.0-20180812185044-276a5c0a1039 // indirect
-	github.com/spf13/afero v1.2.2 // indirect
-	github.com/spf13/jwalterweatherman v1.1.0 // indirect
+	github.com/breml/bidichk v0.2.6 // indirect
+	github.com/breml/errchkjson v0.3.5 // indirect
+	github.com/chavacava/garif v0.1.0 // indirect
+	github.com/daixiang0/gci v0.11.1 // indirect
+	github.com/golangci/golangci-lint v1.54.2
+	github.com/golangci/revgrep v0.5.0 // indirect
+	github.com/hashicorp/errwrap v1.1.0 // indirect
+	github.com/jgautheron/goconst v1.6.0 // indirect
+	github.com/mattn/go-isatty v0.0.19 // indirect
+	github.com/mattn/go-runewidth v0.0.15 // indirect
+	github.com/mgechev/revive v1.3.3 // indirect
+	github.com/pelletier/go-toml/v2 v2.1.0 // indirect
+	github.com/polyfloyd/go-errorlint v1.4.5 // indirect
+	github.com/prometheus/client_golang v1.16.0 // indirect
+	github.com/prometheus/common v0.44.0 // indirect
+	github.com/prometheus/procfs v0.11.1 // indirect
+	github.com/rivo/uniseg v0.4.4 // indirect
+	github.com/ryanrolds/sqlclosecheck v0.5.1 // indirect
+	github.com/spf13/viper v1.16.0 // indirect
+	github.com/stretchr/objx v0.5.1 // indirect
+	github.com/subosito/gotenv v1.6.0 // indirect
+	github.com/tetafro/godot v1.4.15 // indirect
 	github.com/wadey/gocovmerge v0.0.0-20160331181800-b5bfa59ec0ad
-	golang.org/x/tools v0.0.0-20210105210202-9ed45478a130
-	google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.1.0 // indirect
-	google.golang.org/protobuf v1.25.0
+	github.com/xen0n/gosmopolitan v1.2.2 // indirect
+	gitlab.com/bosi/decorder v0.4.1 // indirect
+	go.uber.org/multierr v1.11.0 // indirect
+	go.uber.org/zap v1.25.0 // 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
+	google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.1.0
+	google.golang.org/protobuf v1.31.0
+	mvdan.cc/unparam v0.0.0-20230815095028-f7c6fb1088f0 // indirect
 )
@@ -374,13 +374,15 @@ func renderReplicationReport(t *stringbuilder.B, rep *report.Report, history *by
 			eta = time.Duration((float64(expected)-float64(replicated))/float64(rate)) * time.Second
 		}

-		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)))
+		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()
 		}
-		t.Newline()
 		if containsInvalidSizeEstimates {
 			t.Write("NOTE: not all steps could be size-estimated, total estimate is likely imprecise!")
 			t.Newline()
@@ -493,15 +495,17 @@ func renderPrunerReport(t *stringbuilder.B, r *pruner.Report, fsfilter FilterFun
 	}

 	// global progress bar
-	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()
+	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
@@ -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))

@@ -6,8 +6,6 @@ import (
 	"log/syslog"
 	"os"
 	"reflect"
-	"regexp"
-	"strconv"
 	"time"

 	"github.com/pkg/errors"
@@ -67,7 +65,6 @@ type ActiveJob struct {
 	Name               string                `yaml:"name"`
 	Connect            ConnectEnum           `yaml:"connect"`
 	Pruning            PruningSenderReceiver `yaml:"pruning"`
-	Debug              JobDebugSettings      `yaml:"debug,optional"`
 	Replication        *Replication          `yaml:"replication,optional,fromdefaults"`
 	ConflictResolution *ConflictResolution   `yaml:"conflict_resolution,optional,fromdefaults"`
 }
@@ -77,17 +74,15 @@ type ConflictResolution struct {
 }

 type PassiveJob struct {
-	Type  string           `yaml:"type"`
-	Name  string           `yaml:"name"`
-	Serve ServeEnum        `yaml:"serve"`
-	Debug JobDebugSettings `yaml:"debug,optional"`
+	Type  string    `yaml:"type"`
+	Name  string    `yaml:"name"`
+	Serve ServeEnum `yaml:"serve"`
 }

 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"`
 }
@@ -126,6 +121,7 @@ type BandwidthLimit struct {
 }

 type Replication struct {
+	Triggers    []*ReplicationTriggerEnum
 	Protection  *ReplicationOptionsProtection  `yaml:"protection,optional,fromdefaults"`
 	Concurrency *ReplicationOptionsConcurrency `yaml:"concurrency,optional,fromdefaults"`
 }
@@ -140,6 +136,32 @@ type ReplicationOptionsConcurrency struct {
 	SizeEstimates int `yaml:"size_estimates,optional,default=4"`
 }

+type ReplicationTriggerEnum struct {
+	Ret interface{}
+}
+
+func (t *ReplicationTriggerEnum) UnmarshalYAML(u func(interface{}, bool) error) (err error) {
+	t.Ret, err = enumUnmarshal(u, map[string]interface{}{
+		"manual":   &ReplicationTriggerManual{},
+		"periodic": &ReplicationTriggerPeriodic{},
+	})
+	return
+}
+
+type ReplicationTriggerManual struct {
+	Type string `yaml:"type"`
+}
+
+type ReplicationTriggerPeriodic struct {
+	Type     string            `yaml:"type"`
+	Interval *PositiveDuration `yaml:"interval"`
+}
+
+type ReplicationTriggerCron struct {
+	Type string   `yaml:"type"`
+	Cron CronSpec `yaml:"cron"`
+}
+
 type PropertyRecvOptions struct {
 	Inherit  []zfsprop.Property          `yaml:"inherit,optional"`
 	Override map[zfsprop.Property]string `yaml:"override,optional"`
@@ -162,7 +184,6 @@ func (j *PushJob) GetSendOptions() *SendOptions      { return j.Send }
 type PullJob struct {
 	ActiveJob `yaml:",inline"`
 	RootFS    string                   `yaml:"root_fs"`
-	Interval  PositiveDurationOrManual `yaml:"interval"`
 	Recv      *RecvOptions             `yaml:"recv,fromdefaults,optional"`
 }

@@ -190,13 +211,10 @@ 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
 }
@@ -228,10 +246,11 @@ type SnapshottingEnum struct {
 }

 type SnapshottingPeriodic struct {
-	Type     string        `yaml:"type"`
-	Prefix   string        `yaml:"prefix"`
-	Interval time.Duration `yaml:"interval,positive"`
-	Hooks    HookList      `yaml:"hooks,optional"`
+	Type            string            `yaml:"type"`
+	Prefix          string            `yaml:"prefix"`
+	Interval        *PositiveDuration `yaml:"interval"`
+	Hooks           HookList          `yaml:"hooks,optional"`
+	TimestampFormat string            `yaml:"timestamp_format,optional,default=dense"`
 }

 type CronSpec struct {
@@ -260,10 +279,11 @@ func (s *CronSpec) UnmarshalYAML(unmarshal func(v interface{}, not_strict bool)
 }

 type SnapshottingCron struct {
-	Type   string   `yaml:"type"`
-	Prefix string   `yaml:"prefix"`
-	Cron   CronSpec `yaml:"cron"`
-	Hooks  HookList `yaml:"hooks,optional"`
+	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 {
@@ -483,14 +503,6 @@ 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 {
@@ -715,41 +727,3 @@ 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
-}
@@ -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
+}
@@ -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
+	})
+}
@@ -44,7 +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 {
@@ -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) {
@@ -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
@@ -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 another shell, execute mount vxfs snapshot.
-//    	From the first client, execute UNLOCK TABLES.
-//    	Copy files from the snapshot.
-//    	Unmount the snapshot.
+//	   	From a client program, execute FLUSH TABLES WITH READ LOCK.
+//	   	From another shell, execute mount vxfs snapshot.
+//	   	From the first client, execute UNLOCK TABLES.
+//	   	Copy files from 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
@@ -15,6 +15,7 @@ import (

 	"github.com/zrepl/zrepl/config"
 	"github.com/zrepl/zrepl/daemon/job/reset"
+	"github.com/zrepl/zrepl/daemon/job/trigger"
 	"github.com/zrepl/zrepl/daemon/job/wakeup"
 	"github.com/zrepl/zrepl/daemon/pruner"
 	"github.com/zrepl/zrepl/daemon/snapper"
@@ -44,6 +45,8 @@ type ActiveSide struct {
 	promReplicationErrors prometheus.Gauge
 	promLastSuccessful    prometheus.Gauge

+	triggers *trigger.Triggers
+
 	tasksMtx sync.Mutex
 	tasks    activeSideTasks
 }
@@ -90,7 +93,7 @@ type activeMode interface {
 	SenderReceiver() (logic.Sender, logic.Receiver)
 	Type() Type
 	PlannerPolicy() logic.PlannerPolicy
-	RunPeriodic(ctx context.Context, wakeUpCommon chan<- struct{})
+	RunPeriodic(ctx context.Context, wakeReplication *trigger.Manual)
 	SnapperReport() *snapper.Report
 	ResetConnectBackoff()
 }
@@ -132,8 +135,8 @@ func (m *modePush) Type() Type { return TypePush }

 func (m *modePush) PlannerPolicy() logic.PlannerPolicy { return *m.plannerPolicy }

-func (m *modePush) RunPeriodic(ctx context.Context, wakeUpCommon chan<- struct{}) {
-	m.snapper.Run(ctx, wakeUpCommon)
+func (m *modePush) RunPeriodic(ctx context.Context, trigger *trigger.Manual) {
+	m.snapper.Run(ctx, trigger)
 }

 func (m *modePush) SnapperReport() *snapper.Report {
@@ -221,7 +224,7 @@ func (*modePull) Type() Type { return TypePull }

 func (m *modePull) PlannerPolicy() logic.PlannerPolicy { return *m.plannerPolicy }

-func (m *modePull) RunPeriodic(ctx context.Context, wakeUpCommon chan<- struct{}) {
+func (m *modePull) RunPeriodic(ctx context.Context, wakeReplication *trigger.Manual) {
 	if m.interval.Manual {
 		GetLogger(ctx).Info("manual pull configured, periodic pull disabled")
 		// "waiting for wakeups" is printed in common ActiveSide.do
@@ -232,14 +235,7 @@ func (m *modePull) RunPeriodic(ctx context.Context, wakeUpCommon chan<- struct{}
 	for {
 		select {
 		case <-t.C:
-			select {
-			case wakeUpCommon <- struct{}{}:
-			default:
-				GetLogger(ctx).
-					WithField("pull_interval", m.interval).
-					Warn("pull job took longer than pull interval")
-				wakeUpCommon <- struct{}{} // block anyways, to queue up the wakeup
-			}
+			wakeReplication.Fire()
 		case <-ctx.Done():
 			return
 		}
@@ -370,6 +366,11 @@ func activeSide(g *config.Global, in *config.ActiveJob, configJob interface{}, p
 		return nil, errors.Wrap(err, "cannot build replication driver config")
 	}

+	j.triggers, err = trigger.FromConfig(in.Replication.Triggers)
+	if err != nil {
+		return nil, errors.Wrap(err, "cannot build triggers")
+	}
+
 	return j, nil
 }

@@ -444,12 +445,17 @@ func (j *ActiveSide) Run(ctx context.Context) {

 	defer log.Info("job exiting")

-	periodicDone := make(chan struct{})
 	ctx, cancel := context.WithCancel(ctx)
 	defer cancel()
-	periodicCtx, endTask := trace.WithTask(ctx, "periodic")
+
+	periodCtx, endTask := trace.WithTask(ctx, "periodic")
+	defer endTask()
+	go j.mode.RunPeriodic(periodCtx, periodicTrigger)
+
+	wakeupTrigger := wakeup.Trigger(ctx)
+
+	triggered, endTask := j.triggers.Spawn(ctx, []*trigger.Trigger{periodicTrigger, wakeupTrigger})
 	defer endTask()
-	go j.mode.RunPeriodic(periodicCtx, periodicDone)

 	invocationCount := 0
 outer:
@@ -459,10 +465,15 @@ outer:
 		case <-ctx.Done():
 			log.WithError(ctx.Err()).Info("context")
 			break outer
-
-		case <-wakeup.Wait(ctx):
-			j.mode.ResetConnectBackoff()
-		case <-periodicDone:
+		case trigger := <-triggered:
+			log :=
+				log.WithField("trigger_id", trigger.ID())
+			log.Info("triggered")
+			switch trigger {
+			case wakeupTrigger:
+				log.Info("trigger is wakeup command, resetting connection backoff")
+				j.mode.ResetConnectBackoff()
+			}
 		}
 		invocationCount++
 		invocationCtx, endSpan := trace.WithSpan(ctx, fmt.Sprintf("invocation-%d", invocationCount))
@@ -24,21 +24,6 @@ func JobsFromConfig(c *config.Config, parseFlags config.ParseFlags) ([]Job, erro
 		js[i] = j
 	}

-	// receiving-side root filesystems must not overlap
-	{
-		rfss := make([]string, 0, len(js))
-		for _, j := range js {
-			jrfs, ok := j.OwnedDatasetSubtreeRoot()
-			if !ok {
-				continue
-			}
-			rfss = append(rfss, jrfs.ToString())
-		}
-		if err := validateReceivingSidesDoNotOverlap(rfss); err != nil {
-			return nil, err
-		}
-	}
-
 	return js, nil
 }

@@ -15,6 +15,7 @@ import (

 	"github.com/zrepl/zrepl/config"
 	"github.com/zrepl/zrepl/daemon/filters"
+	"github.com/zrepl/zrepl/daemon/job/trigger"
 	"github.com/zrepl/zrepl/daemon/job/wakeup"
 	"github.com/zrepl/zrepl/daemon/pruner"
 	"github.com/zrepl/zrepl/daemon/snapper"
@@ -104,12 +105,18 @@ func (j *SnapJob) Run(ctx context.Context) {

 	defer log.Info("job exiting")

-	periodicDone := make(chan struct{})
+	wakeupTrigger := wakeup.Trigger(ctx)
+
+	snapshottingTrigger := trigger.New("periodic")
 	ctx, cancel := context.WithCancel(ctx)
 	defer cancel()
 	periodicCtx, endTask := trace.WithTask(ctx, "snapshotting")
 	defer endTask()
-	go j.snapper.Run(periodicCtx, periodicDone)
+	go j.snapper.Run(periodicCtx, snapshottingTrigger)
+
+	triggers := trigger.Empty()
+	triggered, endTask := triggers.Spawn(ctx, []trigger.Trigger{snapshottingTrigger, wakeupTrigger})
+	defer endTask()

 	invocationCount := 0
 outer:
@@ -119,9 +126,7 @@ outer:
 		case <-ctx.Done():
 			log.WithError(ctx.Err()).Info("context")
 			break outer
-
-		case <-wakeup.Wait(ctx):
-		case <-periodicDone:
+		case <-triggered:
 		}
 		invocationCount++

@@ -138,7 +143,7 @@ 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
@@ -0,0 +1,23 @@
+package trigger
+
+import (
+	"context"
+
+	"github.com/robfig/cron/v3"
+)
+
+type Cron struct {
+	spec cron.Schedule
+}
+
+var _ Trigger = &Cron{}
+
+func NewCron(spec cron.Schedule) *Cron {
+	return &Cron{spec: spec}
+}
+
+func (t *Cron) ID() string { return "cron" }
+
+func (t *Cron) run(ctx context.Context, signal chan<- struct{}) {
+	panic("unimpl: extract from cron snapper")
+}
@@ -0,0 +1,30 @@
+package trigger
+
+import (
+	"fmt"
+
+	"github.com/zrepl/zrepl/config"
+)
+
+func FromConfig(in []*config.ReplicationTriggerEnum) (*Triggers, error) {
+	triggers := make([]Trigger, len(in))
+	for i, e := range in {
+		var t Trigger = nil
+		switch te := e.Ret.(type) {
+		case *config.ReplicationTriggerManual:
+			// not a trigger
+			t = NewManual("manual")
+		case *config.ReplicationTriggerPeriodic:
+			t = NewPeriodic(te.Interval.Duration())
+		case *config.ReplicationTriggerCron:
+			t = NewCron(te.Cron.Schedule)
+		default:
+			return nil, fmt.Errorf("unknown trigger type %T", te)
+		}
+		triggers[i] = t
+	}
+	return &Triggers{
+		spawned:  false,
+		triggers: triggers,
+	}, nil
+}
@@ -0,0 +1,12 @@
+package trigger
+
+import (
+	"context"
+
+	"github.com/zrepl/zrepl/logger"
+)
+
+
+func getLogger(ctx context.Context) logger.Logger {
+	panic("unimpl")
+}
@@ -0,0 +1,33 @@
+package trigger
+
+import "context"
+
+type Manual struct {
+	id     string
+	signal chan<- struct{}
+}
+
+var _ Trigger = &Manual{}
+
+func NewManual(id string) *Manual {
+	return &Manual{
+		id:     id,
+		signal: nil,
+	}
+}
+
+func (t *Manual) ID() string {
+	return t.id
+}
+
+func (t *Manual) run(ctx context.Context, signal chan<- struct{}) {
+	if t.signal != nil {
+		panic("run must only be called once")
+	}
+	t.signal = signal
+}
+
+// Panics if called before the trigger has been spanwed as part of a `Triggers`.
+func (t *Manual) Fire() {
+	t.signal <- struct{}{}
+}
@@ -0,0 +1,33 @@
+package trigger
+
+import (
+	"context"
+	"time"
+)
+
+type Periodic struct {
+	interval time.Duration
+}
+
+var _ Trigger = &Periodic{}
+
+func NewPeriodic(interval time.Duration) *Periodic {
+	return &Periodic{
+		interval: interval,
+	}
+}
+
+func (p *Periodic) ID() string { return "periodic" }
+
+func (p *Periodic) run(ctx context.Context, signal chan<- struct{}) {
+	t := time.NewTicker(p.interval)
+	defer t.Stop()
+	for {
+		select {
+		case <-t.C:
+			signal <- struct{}{}
+		case <-ctx.Done():
+			return
+		}
+	}
+}
@@ -0,0 +1,91 @@
+package trigger
+
+import (
+	"context"
+	"github.com/zrepl/zrepl/daemon/logging/trace"
+)
+
+type Triggers struct {
+	spawned  bool
+	triggers []Trigger
+}
+
+type Trigger interface {
+	ID() string
+	run(context.Context, chan<- struct{})
+}
+
+func Empty() *Triggers {
+	return &Triggers{
+		spawned:  false,
+		triggers: nil,
+	}
+}
+
+func (t *Triggers) Spawn(ctx context.Context, additionalTriggers []Trigger) (chan Trigger, trace.DoneFunc) {
+	if t.spawned {
+		panic("must only spawn once")
+	}
+	t.spawned = true
+	t.triggers = append(t.triggers, additionalTriggers...)
+	sink := make(chan Trigger)
+	endTask := t.spawn(ctx, sink)
+	return sink, endTask
+}
+
+type triggering struct {
+	trigger Trigger
+	handled chan struct{}
+}
+
+func (t *Triggers) spawn(ctx context.Context, sink chan Trigger) trace.DoneFunc {
+	ctx, endTask := trace.WithTask(ctx, "triggers")
+	ctx, add, wait := trace.WithTaskGroup(ctx, "trigger-tasks")
+	triggered := make(chan triggering, len(t.triggers))
+	for _, t := range t.triggers {
+		t := t
+		signal := make(chan struct{})
+		go add(func(ctx context.Context) {
+			t.run(ctx, signal)
+		})
+		go func() {
+			for {
+				select {
+				case <-ctx.Done():
+					return
+				case <-signal:
+					handled := make(chan struct{})
+					select {
+					case triggered <- triggering{trigger: t, handled: handled}:
+					default:
+						panic("this funtion ensures that there's always room in the channel")
+					}
+					select {
+					case <-handled:
+					case <-ctx.Done():
+						return
+					}
+				}
+			}
+		}()
+	}
+	go func() {
+		defer wait()
+		for {
+			select {
+			case <-ctx.Done():
+				return
+			case triggering := <-triggered:
+				select {
+				case sink <- triggering.trigger:
+				default:
+					getLogger(ctx).
+						WithField("trigger_id", triggering.trigger.ID()).
+						Warn("dropping triggering because job is busy")
+				}
+				close(triggering.handled)
+			}
+		}
+	}()
+	return endTask
+}
@@ -3,6 +3,8 @@ package wakeup
 import (
 	"context"
 	"errors"
+
+	"github.com/zrepl/zrepl/daemon/job/trigger"
 )

 type contextKey int
@@ -17,6 +19,10 @@ func Wait(ctx context.Context) <-chan struct{} {
 	return wc
 }

+func Trigger(ctx context.Context) trigger.Trigger {
+	panic("unimpl")
+}
+
 type Func func() error

 var AlreadyWokenUp = errors.New("already woken up")
@@ -63,6 +63,7 @@ type Subsystem string
 const (
 	SubsysMeta         Subsystem = "meta"
 	SubsysJob          Subsystem = "job"
+	SubsysTrigger      Subsystem = "trigger"
 	SubsysReplication  Subsystem = "repl"
 	SubsysEndpoint     Subsystem = "endpoint"
 	SubsysPruning      Subsystem = "pruning"
@@ -1,6 +1,6 @@
 // package trace provides activity tracing via ctx through Tasks and Spans
 //
-// Basic Concepts
+// # Basic Concepts
 //
 // Tracing can be used to identify where a piece of code spends its time.
 //
@@ -10,51 +10,50 @@
 // 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.
-//
+// - 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)
+//	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)
-//    }()
-//  }
+//	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)
-//        }()
-//      }
-//    }()
-//  }
+//
+//	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,
@@ -65,8 +64,7 @@
 //
 // Recovering from endSpan() or endTask() panics will corrupt the trace stack and lead to corrupt tracefile output.
 //
-//
-// Best Practices For Naming Tasks And Spans
+// # 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.
@@ -74,8 +72,7 @@
 // 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
+// # 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 .
@@ -11,8 +11,6 @@ import (
 // use like this:
 //
 //	defer WithSpanFromStackUpdateCtx(&existingCtx)()
-//
-//
 func WithSpanFromStackUpdateCtx(ctx *context.Context) DoneFunc {
 	childSpanCtx, end := WithSpan(*ctx, getMyCallerOrPanic())
 	*ctx = childSpanCtx
@@ -199,6 +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() {
@@ -10,6 +10,8 @@ import (

 	"github.com/zrepl/zrepl/config"
 	"github.com/zrepl/zrepl/daemon/hooks"
+	"github.com/zrepl/zrepl/daemon/job/trigger"
+	"github.com/zrepl/zrepl/util/suspendresumesafetimer"
 	"github.com/zrepl/zrepl/zfs"
 )

@@ -20,8 +22,9 @@ func cronFromConfig(fsf zfs.DatasetFilter, in config.SnapshottingCron) (*Cron, e
 		return nil, errors.Wrap(err, "hook config error")
 	}
 	planArgs := planArgs{
-		prefix: in.Prefix,
-		hooks:  hooksList,
+		prefix:          in.Prefix,
+		timestampFormat: in.TimestampFormat,
+		hooks:           hooksList,
 	}
 	return &Cron{config: in, fsf: fsf, planArgs: planArgs}, nil
 }
@@ -40,69 +43,41 @@ type Cron struct {
 	wakeupWhileRunningCount int
 }

-func (s *Cron) Run(ctx context.Context, snapshotsTaken chan<- struct{}) {
+func (s *Cron) Run(ctx context.Context, snapshotsTaken *trigger.Manual) {

-	t := time.NewTimer(0)
-	defer func() {
-		if !t.Stop() {
-			select {
-			case <-t.C:
-			default:
-			}
-		}
-	}()
 	for {
 		now := time.Now()
 		s.mtx.Lock()
 		s.wakeupTime = s.config.Cron.Schedule.Next(now)
 		s.mtx.Unlock()

-		// Re-arm the timer.
-		// Need to Stop before Reset, see docs.
-		if !t.Stop() {
-			// Use non-blocking read from timer channel
-			// because, except for the first loop iteration,
-			// the channel is already drained
-			select {
-			case <-t.C:
-			default:
-			}
-		}
-		t.Reset(s.wakeupTime.Sub(now))
-
-		select {
-		case <-ctx.Done():
+		ctxDone := suspendresumesafetimer.SleepUntil(ctx, s.wakeupTime)
+		if ctxDone != nil {
 			return
-		case <-t.C:
-			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")
-					}
-				}
-			}()
 		}
+
+		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()
+
+			snapshotsTaken.Fire()
+		}()
 	}

 }
@@ -4,6 +4,7 @@ import (
 	"context"
 	"fmt"
 	"sort"
+	"strconv"
 	"strings"
 	"time"

@@ -14,8 +15,9 @@ import (
 )

 type planArgs struct {
-	prefix string
-	hooks  *hooks.List
+	prefix          string
+	timestampFormat string
+	hooks           *hooks.List
 }

 type plan struct {
@@ -58,6 +60,21 @@ type snapProgress struct {
 	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))
@@ -68,7 +85,7 @@ func (plan *plan) execute(ctx context.Context, dryRun bool) (ok bool) {
 	anyFsHadErr := false
 	// TODO channel programs -> allow a little jitter?
 	for fs, progress := range plan.snaps {
-		suffix := time.Now().In(time.UTC).Format("20060102_150405_000")
+		suffix := plan.formatNow(plan.args.timestampFormat)
 		snapname := fmt.Sprintf("%s%s", plan.args.prefix, suffix)

 		ctx := logging.WithInjectedField(ctx, "fs", fs.ToString())
@@ -2,11 +2,13 @@ package snapper

 import (
 	"context"
+
+	"github.com/zrepl/zrepl/daemon/job/trigger"
 )

 type manual struct{}

-func (s *manual) Run(ctx context.Context, wakeUpCommon chan<- struct{}) {
+func (s *manual) Run(ctx context.Context, snapshotsTaken *trigger.Manual) {
 	// nothing to do
 }

@@ -9,12 +9,14 @@ import (

 	"github.com/pkg/errors"

+	"github.com/zrepl/zrepl/daemon/job/trigger"
 	"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"
 )

@@ -22,7 +24,7 @@ func periodicFromConfig(g *config.Global, fsf zfs.DatasetFilter, in *config.Snap
 	if in.Prefix == "" {
 		return nil, errors.New("prefix must not be empty")
 	}
-	if in.Interval <= 0 {
+	if in.Interval.Duration() <= 0 {
 		return nil, errors.New("interval must be positive")
 	}

@@ -32,11 +34,12 @@ func periodicFromConfig(g *config.Global, fsf zfs.DatasetFilter, in *config.Snap
 	}

 	args := periodicArgs{
-		interval: in.Interval,
+		interval: in.Interval.Duration(),
 		fsf:      fsf,
 		planArgs: planArgs{
-			prefix: in.Prefix,
-			hooks:  hookList,
+			prefix:          in.Prefix,
+			timestampFormat: in.TimestampFormat,
+			hooks:           hookList,
 		},
 		// ctx and log is set in Run()
 	}
@@ -49,7 +52,7 @@ type periodicArgs struct {
 	interval       time.Duration
 	fsf            zfs.DatasetFilter
 	planArgs       planArgs
-	snapshotsTaken chan<- struct{}
+	snapshotsTaken *trigger.Manual
 	dryRun         bool
 }

@@ -101,7 +104,7 @@ func (s State) sf() state {
 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{}) {
+func (s *Periodic) Run(ctx context.Context, snapshotsTaken *trigger.Manual) {
 	defer trace.WithSpanFromStackUpdateCtx(&ctx)()
 	getLogger(ctx).Debug("start")
 	defer getLogger(ctx).Debug("stop")
@@ -171,16 +174,13 @@ func periodicStateSyncUp(a periodicArgs, u updater) state {
 	u(func(s *Periodic) {
 		s.sleepUntil = syncPoint
 	})
-	t := time.NewTimer(time.Until(syncPoint))
-	defer t.Stop()
-	select {
-	case <-t.C:
-		return u(func(s *Periodic) {
-			s.state = Planning
-		}).sf()
-	case <-a.ctx.Done():
+	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 {
@@ -208,13 +208,7 @@ func periodicStateSnapshot(a periodicArgs, u updater) state {

 	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")
-		}
-	}
+	a.snapshotsTaken.Fire()

 	return u(func(snapper *Periodic) {
 		if !ok {
@@ -241,17 +235,13 @@ func periodicStateWait(a periodicArgs, u updater) state {
 		logFunc("enter wait-state after error")
 	})

-	t := time.NewTimer(time.Until(sleepUntil))
-	defer t.Stop()
-
-	select {
-	case <-t.C:
-		return u(func(snapper *Periodic) {
-			snapper.state = Planning
-		}).sf()
-	case <-a.ctx.Done():
+	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) {
@@ -5,6 +5,7 @@ import (
 	"fmt"

 	"github.com/zrepl/zrepl/config"
+	"github.com/zrepl/zrepl/daemon/job/trigger"
 	"github.com/zrepl/zrepl/zfs"
 )

@@ -17,7 +18,7 @@ const (
 )

 type Snapper interface {
-	Run(ctx context.Context, snapshotsTaken chan<- struct{})
+	Run(ctx context.Context, snapshotsTaken *trigger.Manual)
 	Report() Report
 }

@@ -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
@@ -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
@@ -10,11 +10,11 @@ BUILDDIR      = _build

 # Put it first so that "make" without argument is like "make help".
 help:
-	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" -c sphinxconf $(SPHINXOPTS) $(O)
+	@$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)

 .PHONY: help Makefile

 # Catch-all target: route all unknown targets to Sphinx using the new
 # "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
 %: Makefile
-	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" -c sphinxconf $(SPHINXOPTS) $(O)
+	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
@@ -0,0 +1,43 @@
+/* https://github.com/sphinx-contrib/sphinxcontrib-versioning/blob/0b56210959c6b21bbb730072e81876491b4e1371/sphinxcontrib/versioning/_static/banner.css */
+
+.scv-banner {
+    padding: 3px;
+    border-radius: 2px;
+    font-size: 80%;
+    text-align: center;
+    color: white;
+    background: #d40 linear-gradient(-45deg,
+        rgba(255, 255, 255, 0.2) 0%,
+        rgba(255, 255, 255, 0.2) 25%,
+        transparent 25%,
+        transparent 50%,
+        rgba(255, 255, 255, 0.2) 50%,
+        rgba(255, 255, 255, 0.2) 75%,
+        transparent 75%,
+        transparent
+    );
+    background-size: 28px 28px;
+}
+.scv-banner > a {
+    color: white;
+}
+
+
+.scv-sphinx_rtd_theme {
+    background-color: #2980B9;
+}
+
+
+.scv-bizstyle {
+    background-color: #336699;
+}
+
+
+.scv-classic {
+    text-align: center !important;
+}
+
+
+.scv-traditional {
+    text-align: center !important;
+}
@@ -0,0 +1,17 @@
+{% extends "!page.html" %}
+{% block body %}
+{% if current_version and latest_version and current_version != latest_version %}
+<p class="scv-banner scv-sphinx_rtd_theme">
+  <strong>
+    {% if current_version.is_released %}
+    You're reading an old version of this documentation.
+    If you want up-to-date information, please have a look at <a href="{{ vpathto(latest_version.name) }}">{{latest_version.name}}</a>.
+    {% else %}
+    You're reading the documentation for a development version.
+    For the latest released version, please have a look at <a href="{{ vpathto(latest_version.name) }}">{{latest_version.name}}</a>.
+    {% endif %}
+  </strong>
+</p>
+{% endif %}
+{{ super() }}
+{% endblock %}%
@@ -0,0 +1,27 @@
+{%- if current_version %}
+<div class="rst-versions" data-toggle="rst-versions" role="note" aria-label="versions">
+  <span class="rst-current-version" data-toggle="rst-current-version">
+    <span class="fa fa-book"> Other Versions</span>
+    v: {{ current_version.name }}
+    <span class="fa fa-caret-down"></span>
+  </span>
+  <div class="rst-other-versions">
+    {%- if versions.tags %}
+    <dl>
+      <dt>Tags</dt>
+      {%- for item in versions.tags %}
+      <dd><a href="{{ item.url }}">{{ item.name }}</a></dd>
+      {%- endfor %}
+    </dl>
+    {%- endif %}
+    {%- if versions.branches %}
+    <dl>
+      <dt>Branches</dt>
+      {%- for item in versions.branches %}
+      <dd><a href="{{ item.url }}">{{ item.name }}</a></dd>
+      {%- endfor %}
+    </dl>
+    {%- endif %}
+  </div>
+</div>
+{%- endif %}
@@ -16,13 +16,63 @@ Changelog
 The changelog summarizes bugfixes that are deemed relevant for users and package maintainers.
 Developers should consult the git commit log or GitHub issue tracker.

-0.6 (Unreleased)
----------------
+Next Release
+------------

-* `Feature Wishlist on GitHub <https://github.com/zrepl/zrepl/discussions/547>`_
+The plan for the next release is to revisit how zrepl does snapshot management.
+High-level goals:
+
+- Make it easy to decouple snapshot management (snapshotting, pruning) from replication.
+- Ability to include/exclude snapshots from replication.
+  This is useful for aforementioned decoupling, e.g., separate snapshot prefixes for local & remote replication.
+  Also, it makes explicit that by default, zrepl replicates all snapshots, and that
+  replication has no concept of "zrepl-created snapshots", which is a common misconception.
+- Use of ``zfs snapshot`` comma syntax or channel programs to take snapshots of multiple
+  datasets atomically.
+- Provide an alternative to the ``grid`` pruning policy.
+  Most likely something based on hourly/daily/weekly/monthly "trains" plus a count.
+- Ability to prune at the granularity of the **group** of snapshots created at a given
+  time, as opposed to the individual snapshots within a dataset.
+  Maybe this will be addressed by the alternative to the ``grid`` pruning policy,
+  as it will likely be more predictable.
+
+Those changes will likely come with some breakage in the config.
+However, I want to avoid breaking **use cases** that are satisfied by the current design.
+There will be beta/RC releases to give users a chance to evaluate.
+
+0.6.1
+-----
+
+* |feature| add metric to detect filesystems rules that don't match any local dataset (thanks, `@gmekicaxcient <https://github.com/gmekicaxcient>`_).
+* |bugfix| ``zrepl status``: hide progress bar once all filesystems reach terminal state (thanks, `@0x3333 <https://github.com/0x3333>`_).
+* |bugfix| handling of tenative cursor presence if protection strategy doesn't use it (:issue:`714`).
+* |docs| address setup with two or more external disks (thanks, `@se-jaeger <https://github.com/se-jaeger>`_).
+* |docs| document ``replication`` and ``conflict_resolution`` options (thanks, `@InsanePrawn <https://github.com/InsanePrawn>`_).
+* |docs| docs: talks: add note on keep_bookmarks option (thanks, `@skirmess <https://github.com/skirmess>`_).
+* |maint| dist: add openrc service file (thanks, `@gramosg <https://github.com/gramosg>`_).
+* |maint| grafana: update dashboard to Grafana 9.3.6.
+* |maint| run platform tests as part of CI.
+* |maint| build: upgrade to Go 1.21 and update golangci-lint; minimum Go version for builds is now 1.20
+
+.. NOTE::
+  |  zrepl is a spare-time project primarily developed by `Christian Schwarz <https://cschwarz.com>`_.
+  |  You can support maintenance and feature development through one of the following services:
+  |  |Donate via Patreon| |Donate via GitHub Sponsors| |Donate via Liberapay| |Donate via PayPal|
+  |  Note that PayPal processing fees are relatively high for small donations.
+  |  For SEPA wire transfer and **commercial support**, please `contact Christian directly <https://cschwarz.com>`_.
+
+0.6
+---

 * |feature| :ref:`Schedule-based snapshotting<job-snapshotting--cron>` using ``cron`` syntax instead of an interval.
-* |feature| Add ``ZREPL_DESTROY_MAX_BATCH_SIZE`` env var (default 0=unlimited).
+* |feature| Configurable initial replication policy.
+  When a filesystem is first replicated to a receiver, this control whether just the newest
+  snapshot will be replicated vs. all existing snapshots. Learn more :ref:`in the docs <conflict_resolution-initial_replication>`.
+* |feature| Configurable timestamp format for snapshot names via :ref:`timestamp_format<job-snapshotting-timestamp_format>`
+  (Thanks, `@ydylla <https://github.com/ydylla>`_).
+* |feature| Add ``ZREPL_DESTROY_MAX_BATCH_SIZE`` env var (default 0=unlimited)
+  (Thanks, `@3nprob <https://github.com/3nprob>`_).
+* |feature| Add ``zrepl configcheck --skip-cert-check`` flag (Thanks, `@cole-h <https://github.com/cole-h>`_).
 * |bugfix| Fix resuming from interrupted replications that use ``send.raw`` on unencrypted datasets.

  * The send options introduced in zrepl 0.4 allow users to specify additional zfs send flags for zrepl to use.
@@ -34,7 +84,7 @@ Developers should consult the git commit log or GitHub issue tracker.
  * However, this means that the ``zrepl status`` UI no longer indicates whether a replication step uses encrypted sends or not.
    The setting is still effective though.

-* |break| |feature| convert Prometheus metric ``zrepl_version_daemon`` to ``zrepl_start_time`` metric
+* |break| convert Prometheus metric ``zrepl_version_daemon`` to ``zrepl_start_time`` metric

  * The metric still reports the zrepl version in a label.
    But the metric *value* is now the Unix timestamp at the time the daemon was started.
@@ -43,6 +93,13 @@ Developers should consult the git commit log or GitHub issue tracker.
 * |bugfix| transient zrepl status error: ``Post "http://unix/status": EOF``
 * |bugfix| don't treat receive-side bookmarks as a replication conflict.
  This facilitates chaining of replication jobs. See :issue:`490`.
+* |bugfix| workaround for Go/gRPC problem on Illumos where zrepl would
+  crash when using the ``local`` transport type (:issue:`598`).
+* |bugfix| fix active child tasks panic that cold occur during replication plannig (:issue:`193abbe`)
+* |bugfix| ``zrepl status`` off-by-one error in display of completed step count (:commit:`ce6701f`)
+* |bugfix| Allow using day & week units for ``snapshotting.interval`` (:commit:`ffb1d89`)
+* |docs| ``docs/overview`` improvements (Thanks, `@jtagcat <https://github.com/jtagcat>`_).
+* |maint| Update to Go 1.19.

 0.5
 ---
@@ -75,14 +132,6 @@ Note to all users: please read up on the following OpenZFS bugs, as you might be

 Finally, I'd like to point you to the `GitHub discussion <https://github.com/zrepl/zrepl/discussions/547>`_ about which bugfixes and features should be prioritized in zrepl 0.6 and beyond!

-.. NOTE::
-  |  zrepl is a spare-time project primarily developed by `Christian Schwarz <https://cschwarz.com>`_.
-  |  You can support maintenance and feature development through one of the following services:
-  |  |Donate via Patreon| |Donate via GitHub Sponsors| |Donate via Liberapay| |Donate via PayPal|
-  |  Note that PayPal processing fees are relatively high for small donations.
-  |  For SEPA wire transfer and **commercial support**, please `contact Christian directly <https://cschwarz.com>`_.
-
-
 0.4.0
 -----

@@ -1 +0,0 @@
-sphinxconf/conf.py
@@ -0,0 +1,187 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+#
+# zrepl documentation build configuration file, created by
+# sphinx-quickstart on Wed Nov  8 22:28:10 2017.
+#
+# This file is execfile()d with the current directory set to its
+# containing dir.
+#
+# Note that not all possible configuration values are present in this
+# autogenerated file.
+#
+# All configuration values have a default; values that are commented out
+# serve to show the default.
+
+# If extensions (or modules to document with autodoc) are in another directory,
+# add these directories to sys.path here. If the directory is relative to the
+# documentation root, use os.path.abspath to make it absolute, like shown here.
+#
+# import os
+# import sys
+# sys.path.insert(0, os.path.abspath('.'))
+
+
+# -- General configuration ------------------------------------------------
+
+# If your documentation needs a minimal Sphinx version, state it here.
+#
+# needs_sphinx = '1.0'
+
+# Add any Sphinx extension module names here, as strings. They can be
+# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
+# ones.
+extensions = ['sphinx.ext.todo',
+    'sphinx.ext.githubpages',
+    'sphinx.ext.extlinks',
+    "sphinx_multiversion",
+]
+
+# suppress_warnings = ['image.nonlocal_uri']
+
+# Add any paths that contain templates here, relative to this directory.
+templates_path = ['./_templates']
+
+# The suffix(es) of source filenames.
+# You can specify multiple suffix as a list of string:
+#
+# source_suffix = ['.rst', '.md']
+source_suffix = '.rst'
+
+# The master toctree document.
+master_doc = 'index'
+
+# General information about the project.
+project = 'zrepl'
+copyright = '2017-2023, Christian Schwarz'
+author = 'Christian Schwarz'
+
+# The version info for the project you're documenting, acts as replacement for
+# |version| and |release|, also used in various other places throughout the
+# built documents.
+#
+# The short X.Y version.
+#version = set by sphinxcontrib-versioning
+# The full version, including alpha/beta/rc tags.
+#release = version
+
+# The language for content autogenerated by Sphinx. Refer to documentation
+# for a list of supported languages.
+#
+# This is also used if you do content translation via gettext catalogs.
+# Usually you set "language" from the command line for these cases.
+language = 'en'
+
+# List of patterns, relative to source directory, that match files and
+# directories to ignore when looking for source files.
+# This patterns also effect to html_static_path and html_extra_path
+exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
+
+# The name of the Pygments (syntax highlighting) style to use.
+pygments_style = 'sphinx'
+
+# If true, `todo` and `todoList` produce output, else they produce nothing.
+todo_include_todos = True
+
+
+# -- Options for HTML output ----------------------------------------------
+
+# The theme to use for HTML and HTML Help pages.  See the documentation for
+# a list of builtin themes.
+#
+html_theme = 'sphinx_rtd_theme'
+
+# Theme options are theme-specific and customize the look and feel of a theme
+# further.  For a list of options available for each theme, see the
+# documentation.
+#
+# html_theme_options = {}
+
+# Add any paths that contain custom static files (such as style sheets) here,
+# relative to this directory. They are copied after the builtin static files,
+# so a file named "default.css" will overwrite the builtin "default.css".
+html_static_path = ['_static']
+
+html_css_files = [
+    'banner.css',
+]
+
+html_logo = '_static/zrepl.svg'
+
+html_context = {
+  # https://github.com/rtfd/sphinx_rtd_theme/issues/205
+  # Add 'Edit on Github' link instead of 'View page source'
+  "display_github": True,
+  "github_user": "zrepl",
+  "github_repo": "zrepl",
+  "github_version": "master",
+  "conf_py_path": "/docs/",
+  "source_suffix": source_suffix,
+}
+
+# -- Options for HTMLHelp output ------------------------------------------
+
+# Output file base name for HTML help builder.
+htmlhelp_basename = 'zrepldoc'
+
+
+# -- Options for LaTeX output ---------------------------------------------
+
+latex_elements = {
+    # The paper size ('letterpaper' or 'a4paper').
+    #
+    # 'papersize': 'letterpaper',
+
+    # The font size ('10pt', '11pt' or '12pt').
+    #
+    # 'pointsize': '10pt',
+
+    # Additional stuff for the LaTeX preamble.
+    #
+    # 'preamble': '',
+
+    # Latex figure (float) alignment
+    #
+    # 'figure_align': 'htbp',
+}
+
+# Grouping the document tree into LaTeX files. List of tuples
+# (source start file, target name, title,
+#  author, documentclass [howto, manual, or own class]).
+latex_documents = [
+    (master_doc, 'zrepl.tex', 'zrepl Documentation',
+     'Christian Schwarz', 'manual'),
+]
+
+
+# -- Options for manual page output ---------------------------------------
+
+# One entry per manual page. List of tuples
+# (source start file, name, description, authors, manual section).
+man_pages = [
+    (master_doc, 'zrepl', 'zrepl Documentation',
+     [author], 1)
+]
+
+
+# -- Options for Texinfo output -------------------------------------------
+
+# Grouping the document tree into Texinfo files. List of tuples
+# (source start file, target name, title, author,
+#  dir menu entry, description, category)
+texinfo_documents = [
+    (master_doc, 'zrepl', 'zrepl Documentation',
+     author, 'zrepl', 'One line description of project.',
+     'Miscellaneous'),
+]
+
+
+# -- Options for the extlinks extension -----------------------------------
+# http://www.sphinx-doc.org/en/stable/ext/extlinks.html
+extlinks = {
+    'issue':('https://github.com/zrepl/zrepl/issues/%s', 'issue #%s'),
+    'repomasterlink':('https://github.com/zrepl/zrepl/blob/master/%s', '%s'),
+    'sampleconf':('https://github.com/zrepl/zrepl/blob/master/config/samples%s', 'config/samples%s'),
+    'commit':('https://github.com/zrepl/zrepl/commit/%s', 'commit %s'),
+        }
+
@@ -1,5 +1,6 @@
 .. include:: ../global.rst.inc

+.. _conflict_resolution-options:

 Conflict Resolution Options
 ===========================
@@ -15,7 +16,7 @@ Conflict Resolution Options

     ...

-.. _conflict_resolution-initial_replication-option-send_all_snapshots:
+.. _conflict_resolution-initial_replication:


 ``initial_replication`` option
@@ -30,6 +30,10 @@ Job Type ``push``
      - |snapshotting-spec|
    * - ``pruning``
      - |pruning-spec|
+    * - ``replication``
+      - |replication-options|
+    * - ``conflict_resolution``
+      - |conflict-resolution-options|

 Example config: :sampleconf:`/push.yml`

@@ -81,6 +85,10 @@ Job Type ``pull``
        | ``manual`` disables periodic pulling, replication then only happens on :ref:`wakeup <cli-signal-wakeup>`.
    * - ``pruning``
      - |pruning-spec|
+    * - ``replication``
+      - |replication-options|
+    * - ``conflict_resolution``
+      - |conflict-resolution-options|

 Example config: :sampleconf:`/pull.yml`

@@ -44,27 +44,3 @@ Interval & duration fields in job definitions, pruning configurations, etc. must

    var durationStringRegex *regexp.Regexp = regexp.MustCompile(`^\s*(\d+)\s*(s|m|h|d|w)\s*$`)
    // s = second, m = minute, h = hour, d = day, w = week (7 days)
-
-Super-Verbose Job Debugging
---------------------------
-
-You have probably landed here because you opened an issue on GitHub and some developer told you to do this...
-So just read the annotated comments ;)
-
-::
-
-    job:
-    - name: ...
-      ...
-     # JOB DEBUGGING OPTIONS
-      # should be equal for all job types, but each job implements the debugging itself
-      debug:
-        conn: # debug the io.ReadWriteCloser connection
-          read_dump: /tmp/connlog_read   # dump results of Read() invocations to this file
-          write_dump: /tmp/connlog_write # dump results of Write() invocations to this file
-        rpc: # debug the RPC protocol implementation
-          log: true # log output from rpc layer to the job log
-
-.. ATTENTION::
-
-    Connection dumps will almost certainly contain your or other's private data. Do not share it in a bug report.
@@ -130,7 +130,7 @@ The following high-level steps take place during replication and can be monitore
  * Move the **replication cursor** bookmark on the sending side (see below).
  * Move the **last-received-hold** on the receiving side (see below).
  * Release the send-side step-holds.
-   
+
 The idea behind the execution order of replication steps is that if the sender snapshots all filesystems simultaneously at fixed intervals, the receiver will have all filesystems snapshotted at time ``T1`` before the first snapshot at ``T2 = T1 + $interval`` is replicated.

 ZFS Background Knowledge
@@ -258,6 +258,9 @@ On your setup, ensure that

 * all ``filesystems`` filter specifications are disjoint
 * no ``root_fs`` is a prefix or equal to another ``root_fs``
+
+  * For ``sink`` jobs, consider all possible ``root_fs/${client_identity}``.
+
 * no ``filesystems`` filter matches any ``root_fs``

 **Exceptions to the rule**:
@@ -1,5 +1,6 @@
 .. include:: ../global.rst.inc

+.. _replication-options:

 Replication Options
 ===================
@@ -197,11 +197,8 @@ Mount behaviour
 * ``canmount``
 * ``overlay``

-Note: inheriting or overriding the ``mountpoint`` property on ZVOLs fails in ``zfs recv``.
-This is an `issue in OpenZFS <https://github.com/openzfs/zfs/issues/11416>`_ .
-As a workaround, consider creating separate zrepl jobs for your ZVOL and filesystem datasets.
-Please comment at zrepl :issue:`430` if you encounter this issue and/or would like zrepl to automatically work around it.
-
+Note: Before `OpenZFS 2.0.5 <https://github.com/openzfs/zfs/issues/11416>`_, inheriting or overriding the ``mountpoint`` property on ZVOLs fails in ``zfs recv``.
+If you are on such an older version, consider creating separate zrepl jobs for your ZVOL and filesystem datasets.

 Systemd
 -------
@@ -62,6 +62,9 @@ The ``periodic`` and ``cron`` snapshotting types share some common options and b
       type: periodic
       prefix: zrepl_
       interval: 10m
+       # Timestamp format that is used as snapshot suffix.
+       # Can be any of "dense" (default), "human", "iso-8601", "unix-seconds" or a custom Go time format (see https://go.dev/src/time/format.go)
+       timestamp_format: dense
       hooks: ...
    pruning: ...

@@ -91,6 +94,9 @@ The snapshotter uses the ``prefix`` to identify which snapshots it created.
       # (second, optional) minute hour day-of-month month day-of-week
       # This example takes snapshots daily at 3:00.
       cron: "0 3 * * *"
+       # Timestamp format that is used as snapshot suffix.
+       # Can be any of "dense" (default), "human", "iso-8601", "unix-seconds" or a custom Go time format (see https://go.dev/src/time/format.go)
+       timestamp_format: dense
     pruning: ...

 In ``cron`` mode, the snapshotter takes snaphots at fixed points in time.
@@ -98,6 +104,21 @@ See https://en.wikipedia.org/wiki/Cron for details on the syntax.
 zrepl uses the ``the github.com/robfig/cron/v3`` Go package for parsing.
 An optional field for "seconds" is supported to take snapshots at sub-minute frequencies.

+.. _job-snapshotting-timestamp_format:
+
+Timestamp Format
+~~~~~~~~~~~~~~~~
+
+The ``cron`` and ``periodic`` snapshotter support configuring a custom timestamp format that is used as suffix for the snapshot name.
+It can be used by setting ``timestamp_format`` to any of the following values:
+
+* ``dense`` (default) looks like ``20060102_150405_000``
+* ``human`` looks like ``2006-01-02_15:04:05``
+* ``iso-8601`` looks like ``2006-01-02T15:04:05.000Z``
+* ``unix-seconds`` looks like ``1136214245``
+* Any custom Go time format accepted by `time.Time#Format <https://go.dev/src/time/format.go>`_.
+
+
 ``manual`` Snapshotting
 -----------------------

@@ -26,6 +26,8 @@
 .. |connect-transport| replace:: :ref:`connect specification<transport>`
 .. |send-options| replace:: :ref:`send options<job-send-options>`, e.g. for encrypted sends
 .. |recv-options| replace:: :ref:`recv options<job-recv-options>`
+.. |replication-options| replace:: :ref:`replication options<replication-options>`
+.. |conflict-resolution-options| replace:: :ref:`conflict resolution options<conflict_resolution-options>`
 .. |snapshotting-spec| replace:: :ref:`snapshotting specification <job-snapshotting-spec>`
 .. |pruning-spec| replace:: :ref:`pruning specification <prune>`
 .. |filter-spec| replace:: :ref:`filter specification<pattern-filter>`
@@ -15,3 +15,5 @@ Talks & Presentations
  `Event <https://wiki.freebsd.org/DevSummit/201709>`__
  )

+  * Note: The remarks on ``keep_bookmarks`` are irrelevant as of zrepl 0.1 which introduced the zrepl-managed replication cursor bookmark.
+    Read the `Overview <overview-how-replication-works>`_ section to learn more.
@@ -3,7 +3,8 @@ set -euo pipefail

 NON_INTERACTIVE=false
 DO_CLONE=false
-while getopts "ca" arg; do
+PUSH=false
+while getopts "caP" arg; do
    case "$arg" in
        "a")
            NON_INTERACTIVE=true
@@ -11,6 +12,9 @@ while getopts "ca" arg; do
        "c")
            DO_CLONE=true
            ;;
+        "P")
+            PUSH=true
+            ;;
        *)
            echo "invalid option '-$arg'"
            exit 1
@@ -26,8 +30,8 @@ checkout_repo_msg() {
    echo "clone ${GHPAGESREPO} to ${PUBLICDIR}:"
 }

-if ! type sphinx-versioning >/dev/null; then
-    echo "install sphinx-versioning and come back"
+if ! type sphinx-multiversion >/dev/null; then
+    echo "install sphinx-multiversion and come back"
    exit 1
 fi

@@ -49,11 +53,11 @@ else
    read -r
 fi

-pushd "$PUBLICDIR" 
+pushd "$PUBLICDIR"

 echo "verify we're in the GitHub pages repo..."
 git remote get-url origin | grep -E "^${GHPAGESREPO}\$"
-if [ "$?" -ne "0" ] ;then 
+if [ "$?" -ne "0" ] ;then
    checkout_repo_msg
    echo "finished checkout, please run again"
    exit 1
@@ -73,22 +77,15 @@ popd

 echo "building site"

-flags="$(python3 gen-sphinx-versioning-flags.py)"
-set -e
-sphinx-versioning build \
-   $flags \
-   docs ./public_git \
-   -- -c sphinxconf # older conf.py throw errors because they used
-                    # version = subprocess.show_output(["git", "describe"])
-                    # which fails when building with sphinxcontrib-versioning
-set +e
+python3 run-sphinx-multiversion.py . ./public_git
+

 CURRENT_COMMIT=$(git rev-parse HEAD)
 git status --porcelain
 if [[ "$(git status --porcelain)" != "" ]]; then
-    CURRENT_COMMIT="${CURRENT_COMMIT}(dirty)" 
+    CURRENT_COMMIT="${CURRENT_COMMIT}(dirty)"
 fi
-COMMIT_MSG="sphinx-versioning render from publish.sh - $(date -u) - ${CURRENT_COMMIT}"
+COMMIT_MSG="render from publish.sh - $(date -u) - ${CURRENT_COMMIT}"

 pushd "$PUBLICDIR"

@@ -100,6 +97,11 @@ if [ "$(git status --porcelain)" != "" ]; then
 else
    echo "nothing to commit"
 fi
-echo "pushing to GitHub pages repo"
-git push origin master
+
+if $PUSH; then
+    echo "pushing to GitHub pages repo"
+    git push origin master
+else
+    echo "not pushing to GitHub pages repo, set -P flag to push"
+fi

@@ -33,4 +33,17 @@ You will likely want to customize some aspects mentioned in the top comment in t

 .. literalinclude:: ../../config/samples/quickstart_backup_to_external_disk.yml

+
+Offline Backups with two (or more) External Disks
+-------------------------------------------------
+
+It can be desirable to have multiple disk-based backups of the same machine.
+To accomplish this,
+
+* create one zpool per external HDD, each with a unique name, and
+* define a pair of ``push`` and ``sink`` job **for each** of these zpools, each with a unique ``name``, ``listener_name``, and ``root_fs``.
+
+The unique names ensure that the jobs don't step on each others' toes when managing :ref:`zrepl's ZFS abstractions <zrepl-zfs-abstractions>` . 
+
+
 :ref:`Click here <quickstart-apply-config>` to go back to the quickstart guide.
@@ -1,30 +1,24 @@
-alabaster==0.7.12
-attrs==19.1.0
-Babel==2.7.0
-certifi==2019.6.16
-chardet==3.0.4
-Click==7.0
-colorclass==2.2.0
-docutils==0.15.2
-idna==2.8
-imagesize==1.1.0
-Jinja2==2.10.1
-MarkupSafe==1.1.1
-packaging==19.1
-Pygments==2.4.2
-pyparsing==2.4.2
-pytz==2019.2
-requests==2.22.0
-six==1.12.0
-snowballstemmer==1.9.1
-Sphinx==1.8.5
-sphinx-rtd-theme==0.4.3
-sphinxcontrib-applehelp==1.0.1
-sphinxcontrib-devhelp==1.0.1
-sphinxcontrib-htmlhelp==1.0.2
+alabaster==0.7.13
+Babel==2.12.1
+certifi==2023.7.22
+charset-normalizer==3.2.0
+docutils==0.18.1
+idna==3.4
+imagesize==1.4.1
+Jinja2==3.1.2
+MarkupSafe==2.1.3
+packaging==23.1
+Pygments==2.16.1
+requests==2.31.0
+snowballstemmer==2.2.0
+Sphinx==7.2.5
+sphinx-multiversion @ git+https://github.com/zrepl/sphinx-multiversion/@52c915d7ad898d9641ec48c8bbccb7d4f079db93
+sphinx-rtd-theme==1.3.0
+sphinxcontrib-applehelp==1.0.7
+sphinxcontrib-devhelp==1.0.5
+sphinxcontrib-htmlhelp==2.0.4
+sphinxcontrib-jquery==4.1
 sphinxcontrib-jsmath==1.0.1
-sphinxcontrib-qthelp==1.0.2
-sphinxcontrib-serializinghtml==1.1.3
-git+https://github.com/rwblair/sphinxcontrib-versioning.git@7e3885a389a809e17ea55261316b7b0e98dbf98f#egg=sphinxcontrib-versioning
-sphinxcontrib-websupport==1.1.2
-urllib3==1.25.3
+sphinxcontrib-qthelp==1.0.6
+sphinxcontrib-serializinghtml==1.1.9
+urllib3==2.0.4
@@ -1,7 +1,15 @@
 #!/usr/bin/env python3

+from pathlib import Path
 import subprocess
 import re
+import argparse
+import distutils
+
+argparser = argparse.ArgumentParser()
+argparser.add_argument("docsroot")
+argparser.add_argument("outdir")
+args = argparser.parse_args()

 output = subprocess.run(["git", "tag", "-l"], capture_output=True, check=True, text=True)
 tagRE = re.compile(r"^v(?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(-rc(?P<rc>\d+))?$")
@@ -27,8 +35,8 @@ for line in output.stdout.split("\n"):

    t = Tag()
    t.orig = line
-    t.major = int(m["major"])   
-    t.minor = int(m["minor"])   
+    t.major = int(m["major"])
+    t.minor = int(m["minor"])
    t.patch = int(m["patch"])
    t.rc = int(m["rc"] if m["rc"] is not None else 0)

@@ -49,26 +57,16 @@ for (mm, l) in by_major_minor.items():
    latest_by_major_minor.append(l[-1])
 latest_by_major_minor.sort(key=lambda tag: (tag.major, tag.minor))

-# print(by_major_minor)
-# print(latest_by_major_minor)
-
-cmdline = []
-
-for latest_patch in latest_by_major_minor:
-    cmdline.append("--whitelist-tags")
-    cmdline.append(f"^{re.escape(latest_patch.orig)}$")
-
-# we want flexibility to update docs for the latest stable release
-# => we have a branch for that, called `stable` which we move manually
-# TODO: in the future, have f"stable-{latest_by_major_minor[-1]}"
-default_version = "stable"
-cmdline.extend(["--whitelist-branches", default_version])
-
-cmdline.extend(["--root-ref", f"{default_version}"])
-cmdline.extend(["--banner-main-ref", f"{default_version}"])
-cmdline.extend(["--show-banner"])
-cmdline.extend(["--sort", "semver"])
-
-cmdline.extend(["--whitelist-branches", "master"])
-
-print(" ".join(cmdline))
+cmdline = [
+    "sphinx-multiversion",
+    "-D", "smv_tag_whitelist=^({})$".format("|".join([re.escape(tag.orig) for tag in latest_by_major_minor])),
+    "-D", "smv_branch_whitelist=^(master|stable)$",
+    "-D", "smv_remote_whitelist=^.*$",
+    "-D", "smv_latest_version=stable",
+    "-D", r"smv_released_pattern=^refs/(tags|heads|remotes/[^/]+)/(?!master).*$", # treat everything except master as released, that way, the banner message makes sense
+    # "--dump-metadata", # for debugging
+    args.docsroot,
+    args.outdir,
+]
+print(cmdline)
+subprocess.run(cmdline, check=True)
@@ -1,182 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-#
-# zrepl documentation build configuration file, created by
-# sphinx-quickstart on Wed Nov  8 22:28:10 2017.
-#
-# This file is execfile()d with the current directory set to its
-# containing dir.
-#
-# Note that not all possible configuration values are present in this
-# autogenerated file.
-#
-# All configuration values have a default; values that are commented out
-# serve to show the default.
-
-# If extensions (or modules to document with autodoc) are in another directory,
-# add these directories to sys.path here. If the directory is relative to the
-# documentation root, use os.path.abspath to make it absolute, like shown here.
-#
-# import os
-# import sys
-# sys.path.insert(0, os.path.abspath('.'))
-
-
-# -- General configuration ------------------------------------------------
-
-# If your documentation needs a minimal Sphinx version, state it here.
-#
-# needs_sphinx = '1.0'
-
-# Add any Sphinx extension module names here, as strings. They can be
-# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
-# ones.
-extensions = ['sphinx.ext.todo',
-    'sphinx.ext.githubpages',
-    'sphinx.ext.extlinks']
-
-suppress_warnings = ['image.nonlocal_uri']
-
-# Add any paths that contain templates here, relative to this directory.
-templates_path = ['../_templates']
-
-# The suffix(es) of source filenames.
-# You can specify multiple suffix as a list of string:
-#
-# source_suffix = ['.rst', '.md']
-source_suffix = '.rst'
-
-# The master toctree document.
-master_doc = 'index'
-
-# General information about the project.
-project = 'zrepl'
-copyright = '2017-2019, Christian Schwarz'
-author = 'Christian Schwarz'
-
-# The version info for the project you're documenting, acts as replacement for
-# |version| and |release|, also used in various other places throughout the
-# built documents.
-#
-# The short X.Y version.
-#version = set by sphinxcontrib-versioning
-# The full version, including alpha/beta/rc tags.
-#release = version
-
-# The language for content autogenerated by Sphinx. Refer to documentation
-# for a list of supported languages.
-#
-# This is also used if you do content translation via gettext catalogs.
-# Usually you set "language" from the command line for these cases.
-language = None
-
-# List of patterns, relative to source directory, that match files and
-# directories to ignore when looking for source files.
-# This patterns also effect to html_static_path and html_extra_path
-exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
-
-# The name of the Pygments (syntax highlighting) style to use.
-pygments_style = 'sphinx'
-
-# If true, `todo` and `todoList` produce output, else they produce nothing.
-todo_include_todos = True
-
-
-# -- Options for HTML output ----------------------------------------------
-
-# The theme to use for HTML and HTML Help pages.  See the documentation for
-# a list of builtin themes.
-#
-html_theme = 'sphinx_rtd_theme'
-
-# Theme options are theme-specific and customize the look and feel of a theme
-# further.  For a list of options available for each theme, see the
-# documentation.
-#
-# html_theme_options = {}
-
-# Add any paths that contain custom static files (such as style sheets) here,
-# relative to this directory. They are copied after the builtin static files,
-# so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['../_static']
-
-
-html_logo = '../_static/zrepl.svg'
-
-html_context = {
-  # https://github.com/rtfd/sphinx_rtd_theme/issues/205
-  # Add 'Edit on Github' link instead of 'View page source'
-  "display_github": True, 
-  "github_user": "zrepl",
-  "github_repo": "zrepl",
-  "github_version": "master",
-  "conf_py_path": "/docs/",
-  "source_suffix": source_suffix,
-}
-
-# -- Options for HTMLHelp output ------------------------------------------
-
-# Output file base name for HTML help builder.
-htmlhelp_basename = 'zrepldoc'
-
-
-# -- Options for LaTeX output ---------------------------------------------
-
-latex_elements = {
-    # The paper size ('letterpaper' or 'a4paper').
-    #
-    # 'papersize': 'letterpaper',
-
-    # The font size ('10pt', '11pt' or '12pt').
-    #
-    # 'pointsize': '10pt',
-
-    # Additional stuff for the LaTeX preamble.
-    #
-    # 'preamble': '',
-
-    # Latex figure (float) alignment
-    #
-    # 'figure_align': 'htbp',
-}
-
-# Grouping the document tree into LaTeX files. List of tuples
-# (source start file, target name, title,
-#  author, documentclass [howto, manual, or own class]).
-latex_documents = [
-    (master_doc, 'zrepl.tex', 'zrepl Documentation',
-     'Christian Schwarz', 'manual'),
-]
-
-
-# -- Options for manual page output ---------------------------------------
-
-# One entry per manual page. List of tuples
-# (source start file, name, description, authors, manual section).
-man_pages = [
-    (master_doc, 'zrepl', 'zrepl Documentation',
-     [author], 1)
-]
-
-
-# -- Options for Texinfo output -------------------------------------------
-
-# Grouping the document tree into Texinfo files. List of tuples
-# (source start file, target name, title, author,
-#  dir menu entry, description, category)
-texinfo_documents = [
-    (master_doc, 'zrepl', 'zrepl Documentation',
-     author, 'zrepl', 'One line description of project.',
-     'Miscellaneous'),
-]
-
-
-# -- Options for the extlinks extension -----------------------------------
-# http://www.sphinx-doc.org/en/stable/ext/extlinks.html
-extlinks = {
-    'issue':('https://github.com/zrepl/zrepl/issues/%s', 'issue #'),
-    'repomasterlink':('https://github.com/zrepl/zrepl/blob/master/%s', ''),
-    'sampleconf':('https://github.com/zrepl/zrepl/blob/master/config/samples%s', 'config/samples'),
-    'commit':('https://github.com/zrepl/zrepl/commit/%s', 'commit '),
-        }
-
@@ -32,6 +32,7 @@ We would like to thank the following people and organizations for supporting zre

   <div class="fa fa-code" style="width: 1em;"></div>

+* |supporter-std| `Max Christian Pohle <https://coderonline.de>`_
 * |supporter-gold| Prominic.NET, Inc.
 * |supporter-std| Torsten Blum
 * |supporter-gold| Cyberiada GmbH
@@ -7,6 +7,7 @@ import (
 	"fmt"
 	"io"
 	"path"
+	"strings"

 	"github.com/kr/pretty"
 	"github.com/pkg/errors"
@@ -233,6 +234,21 @@ func (s *Sender) Send(ctx context.Context, r *pdu.SendReq) (*pdu.SendRes, io.Rea
 	//
 	// Note further that a resuming send, due to the idempotent nature of func CreateReplicationCursor and HoldStep,
 	// will never lose its step holds because we just (idempotently re-)created them above, before attempting the cleanup.
+	destroyTypes := AbstractionTypeSet{
+		AbstractionStepHold:                           true,
+		AbstractionTentativeReplicationCursorBookmark: true,
+	}
+	// The replication planner can also pick an endpoint zfs abstraction as FromVersion.
+	// Keep it, so that the replication will succeed.
+	//
+	// NB: there is no abstraction for snapshots, so, we only need to check bookmarks.
+	if sendArgs.FromVersion != nil && sendArgs.FromVersion.IsBookmark() {
+		dp, err := zfs.NewDatasetPath(sendArgs.FS)
+		if err != nil {
+			panic(err) // sendArgs is validated, this shouldn't happen
+		}
+		liveAbs = append(liveAbs, destroyTypes.ExtractBookmark(dp, sendArgs.FromVersion))
+	}
 	func() {
 		ctx, endSpan := trace.WithSpan(ctx, "cleanup-stale-abstractions")
 		defer endSpan()
@@ -245,35 +261,45 @@ func (s *Sender) Send(ctx context.Context, r *pdu.SendReq) (*pdu.SendRes, io.Rea
 			return keep
 		}
 		check := func(obsoleteAbs []Abstraction) {
-			// last line of defense: check that we don't destroy the incremental `from` and `to`
-			// if we did that, we might be about to blow away the last common filesystem version between sender and receiver
-			mustLiveVersions := []zfs.FilesystemVersion{sendArgs.ToVersion}
-			if sendArgs.FromVersion != nil {
-				mustLiveVersions = append(mustLiveVersions, *sendArgs.FromVersion)
+			// Ensure that we don't delete `From` or `To`.
+			// Regardless of whether they are in AbstractionTypeSet or not.
+			// And produce a nice error message in case we do, to aid debugging the resulting panic.
+			//
+			// This is especially important for `From`. We could break incremental replication
+			// if we deleted the last common filesystem version between sender and receiver.
+			type Problem struct {
+				sendArgsWhat string
+				fullpath     string
+				obsoleteAbs  Abstraction
 			}
-			for _, staleVersion := range obsoleteAbs {
-				for _, mustLiveVersion := range mustLiveVersions {
-					isSendArg := zfs.FilesystemVersionEqualIdentity(mustLiveVersion, staleVersion.GetFilesystemVersion())
-					stepHoldBasedGuaranteeStrategy := false
-					k := replicationGuaranteeStrategy.Kind()
-					switch k {
-					case ReplicationGuaranteeKindResumability:
-						stepHoldBasedGuaranteeStrategy = true
-					case ReplicationGuaranteeKindIncremental:
-					case ReplicationGuaranteeKindNone:
-					default:
-						panic(fmt.Sprintf("this is supposed to be an exhaustive match, got %v", k))
-					}
-					isSnapshot := mustLiveVersion.IsSnapshot()
-					if isSendArg && (!isSnapshot || stepHoldBasedGuaranteeStrategy) {
-						panic(fmt.Sprintf("impl error: %q would be destroyed because it is considered stale but it is part of of sendArgs=%s", mustLiveVersion.String(), pretty.Sprint(sendArgs)))
+			problems := make([]Problem, 0)
+			checkFullpaths := make(map[string]string, 2)
+			checkFullpaths["ToVersion"] = sendArgs.ToVersion.FullPath(sendArgs.FS)
+			if sendArgs.FromVersion != nil {
+				checkFullpaths["FromVersion"] = sendArgs.FromVersion.FullPath(sendArgs.FS)
+			}
+			for _, a := range obsoleteAbs {
+				for what, fullpath := range checkFullpaths {
+					if a.GetFullPath() == fullpath && a.GetType().IsSnapshotOrBookmark() {
+						problems = append(problems, Problem{
+							sendArgsWhat: what,
+							fullpath:     fullpath,
+							obsoleteAbs:  a,
+						})
 					}
 				}
 			}
-		}
-		destroyTypes := AbstractionTypeSet{
-			AbstractionStepHold:                           true,
-			AbstractionTentativeReplicationCursorBookmark: true,
+			if len(problems) == 0 {
+				return
+			}
+			var msg strings.Builder
+			fmt.Fprintf(&msg, "cleaning up send stale would destroy send args:\n")
+			fmt.Fprintf(&msg, "  SendArgs: %s\n", pretty.Sprint(sendArgs))
+			for _, check := range problems {
+				fmt.Fprintf(&msg, "would delete %s %s because it was deemed an obsolete abstraction: %s\n",
+					check.sendArgsWhat, check.fullpath, check.obsoleteAbs)
+			}
+			panic(msg.String())
 		}
 		abstractionsCacheSingleton.TryBatchDestroy(ctx, s.jobId, sendArgs.FS, destroyTypes, keep, check)
 	}()
@@ -428,6 +454,7 @@ func (p *Sender) Receive(ctx context.Context, r *pdu.ReceiveReq, _ io.ReadCloser

 type FSFilter interface { // FIXME unused
 	Filter(path *zfs.DatasetPath) (pass bool, err error)
+	UserSpecifiedDatasets() zfs.UserSpecifiedDatasetsSet
 }

 // FIXME: can we get away without error types here?
@@ -439,7 +466,7 @@ type FSMap interface { // FIXME unused
 }

 // NOTE: when adding members to this struct, remember
-//       to add them to `ReceiverConfig.copyIn()`
+// to add them to `ReceiverConfig.copyIn()`
 type ReceiverConfig struct {
 	JobID JobID

@@ -587,6 +614,12 @@ func (f subroot) Filter(p *zfs.DatasetPath) (pass bool, err error) {
 	return p.HasPrefix(f.localRoot) && !p.Equal(f.localRoot), nil
 }

+func (f subroot) UserSpecifiedDatasets() zfs.UserSpecifiedDatasetsSet {
+	return zfs.UserSpecifiedDatasetsSet{
+		f.localRoot.ToString(): true,
+	}
+}
+
 func (f subroot) MapToLocal(fs string) (*zfs.DatasetPath, error) {
 	p, err := zfs.NewDatasetPath(fs)
 	if err != nil {
@@ -89,6 +89,8 @@ func ReplicationGuaranteeFromKind(k ReplicationGuaranteeKind) ReplicationGuarant

 type ReplicationGuaranteeNone struct{}

+func (g ReplicationGuaranteeNone) String() string { return "none" }
+
 func (g ReplicationGuaranteeNone) Kind() ReplicationGuaranteeKind {
 	return ReplicationGuaranteeKindNone
 }
@@ -107,6 +109,8 @@ func (g ReplicationGuaranteeNone) SenderPostRecvConfirmed(ctx context.Context, j

 type ReplicationGuaranteeIncremental struct{}

+func (g ReplicationGuaranteeIncremental) String() string { return "incremental" }
+
 func (g ReplicationGuaranteeIncremental) Kind() ReplicationGuaranteeKind {
 	return ReplicationGuaranteeKindIncremental
 }
@@ -144,6 +148,8 @@ func (g ReplicationGuaranteeIncremental) SenderPostRecvConfirmed(ctx context.Con

 type ReplicationGuaranteeResumability struct{}

+func (g ReplicationGuaranteeResumability) String() string { return "resumability" }
+
 func (g ReplicationGuaranteeResumability) Kind() ReplicationGuaranteeKind {
 	return ReplicationGuaranteeKindResumability
 }
@@ -31,7 +31,7 @@ const (
 	AbstractionReplicationCursorBookmarkV2        AbstractionType = "replication-cursor-bookmark-v2"
 )

-var AbstractionTypesAll = map[AbstractionType]bool{
+var AbstractionTypesAll = AbstractionTypeSet{
 	AbstractionStepHold:                           true,
 	AbstractionLastReceivedHold:                   true,
 	AbstractionTentativeReplicationCursorBookmark: true,
@@ -168,7 +168,7 @@ func (s AbstractionTypeSet) String() string {
 	for i := range s {
 		sts = append(sts, string(i))
 	}
-	sts = sort.StringSlice(sts)
+	sort.Strings(sts)
 	return strings.Join(sts, ",")
 }

@@ -181,6 +181,38 @@ func (s AbstractionTypeSet) Validate() error {
 	return nil
 }

+// Use the `BookmarkExtractor()` method of each abstraction type in this set
+// to try extract an abstraction from the given FilesystemVersion.
+//
+// Abstraction types in this set that don't have a bookmark extractor are skipped.
+//
+// Panics if more than one abstraction type matches.
+func (s AbstractionTypeSet) ExtractBookmark(dp *zfs.DatasetPath, v *zfs.FilesystemVersion) Abstraction {
+	matched := make(AbstractionTypeSet, 1)
+	var matchedAbs Abstraction
+	for absType := range s {
+		extractor := absType.BookmarkExtractor()
+		if extractor == nil {
+			continue
+		}
+		abstraction := extractor(dp, *v)
+		if abstraction != nil {
+			matched[absType] = true
+			matchedAbs = abstraction
+		}
+	}
+	if len(matched) == 0 {
+		return nil
+	}
+	if len(matched) == 1 {
+		if matchedAbs == nil {
+			panic("loop above should always set matchedAbs if there is a match")
+		}
+		return matchedAbs
+	}
+	panic(fmt.Sprintf("abstraction types extractors should not overlap: %s", matched))
+}
+
 type BookmarkExtractor func(fs *zfs.DatasetPath, v zfs.FilesystemVersion) Abstraction

 // returns nil if the abstraction type is not bookmark-based
@@ -238,6 +270,23 @@ func (t AbstractionType) BookmarkNamer() func(fs string, guid uint64, jobId JobI
 	}
 }

+func (t AbstractionType) IsSnapshotOrBookmark() bool {
+	switch t {
+	case AbstractionTentativeReplicationCursorBookmark:
+		return true
+	case AbstractionReplicationCursorBookmarkV1:
+		return true
+	case AbstractionReplicationCursorBookmarkV2:
+		return true
+	case AbstractionStepHold:
+		return false
+	case AbstractionLastReceivedHold:
+		return false
+	default:
+		panic(fmt.Sprintf("unimpl: %q", t))
+	}
+}
+
 type ListZFSHoldsAndBookmarksQuery struct {
 	FS ListZFSHoldsAndBookmarksQueryFilesystemFilter
 	// What abstraction types should match (any contained in the set)
@@ -23,7 +23,7 @@ func replicationCursorBookmarkNameImpl(fs string, guid uint64, jobid string) (st

 var ErrV1ReplicationCursor = fmt.Errorf("bookmark name is a v1-replication cursor")

-//err != nil always means that the bookmark is not a valid replication bookmark
+// err != nil always means that the bookmark is not a valid replication bookmark
 //
 // Returns ErrV1ReplicationCursor as error if the bookmark is a v1 replication cursor
 func ParseReplicationCursorBookmarkName(fullname string) (uint64, JobID, error) {
@@ -150,9 +150,7 @@ func NewOutlets() *Outlets {
 func (os *Outlets) DeepCopy() (copy *Outlets) {
 	copy = NewOutlets()
 	for level := range os.outs {
-		for i := range os.outs[level] {
-			copy.outs[level] = append(copy.outs[level], os.outs[level][i])
-		}
+		copy.outs[level] = append(copy.outs[level], os.outs[level]...)
 	}
 	return copy
 }
@@ -7,7 +7,8 @@ RUN apt-get update && apt-get install -y \
    dh-exec \
 	binutils-aarch64-linux-gnu \
 	binutils-arm-linux-gnueabihf  \
-	binutils-i686-linux-gnu
+	binutils-i686-linux-gnu \
+	binutils-x86-64-linux-gnu

 RUN mkdir -p /build/src && chmod -R 0777 /build

@@ -16,7 +16,7 @@ zrepl is a one-stop, integrated solution for ZFS replication.}
 %define __strip /usr/bin/true

 Name:             zrepl
-Release:          1
+Release:          SUBSTITUTED_BY_MAKEFILE
 Summary:          One-stop, integrated solution for ZFS replication
 License:          MIT
 URL:              https://zrepl.github.io/
@@ -4,6 +4,21 @@ export ZREPL_MOCK_ZFS_COMMAND_LOG="$1"
 shift
 export ZREPL_MOCK_ZFS_PATH="$1"
 shift
-export PATH="$(dirname "${BASH_SOURCE[0]}" )":"$PATH"
+dirname="$(dirname "${BASH_SOURCE[0]}")"
+# If we invoke this script from the top zrepl source tree, like so:
+#    ./platformtest/logmockzfs/logzfsenv ...
+# then dirname is relative, i.e., ./platformtest/logmockzfs.
+# If we put that relative dir in PATH, then Go >= 1.19 will refuse to
+# exec the `./platformtest/logmockzfs/zfs` wrapper script if it finds
+# it via PATH lookup. For Example:
+#   cmd := exec.Command("zfs")
+#   err := cmd.Run()
+# will fail with an error that errors.Is(err, exec.ErrDot), message:
+#   cannot run executable found relative to current directory
+# The solution is to use an abspath.
+# Learn more at https://pkg.go.dev/os/exec#hdr-Executables_in_the_current_directory
+# and https://go-review.googlesource.com/c/go/+/381374
+absdirname="$(readlink -e "$dirname")"
+export PATH="$absdirname":"$PATH"
 args=("$@")
 exec "${args[@]}"
@@ -20,6 +20,7 @@ var Cases = []Case{BatchDestroy,
 	ReplicationIncrementalCleansUpStaleAbstractionsWithCacheOnSecondReplication,
 	ReplicationIncrementalCleansUpStaleAbstractionsWithoutCacheOnSecondReplication,
 	ReplicationIncrementalDestroysStepHoldsIffIncrementalStepHoldsAreDisabledButStepHoldsExist,
+	ReplicationIncrementalHandlesFromVersionEqTentativeCursorCorrectly,
 	ReplicationIncrementalIsPossibleIfCommonSnapshotIsDestroyed,
 	ReplicationInitialAll,
 	ReplicationInitialFail,
@@ -248,7 +248,10 @@ func implReplicationIncrementalCleansUpStaleAbstractions(ctx *platformtest.Conte
 		require.NoError(ctx, err)
 		snap2Hold, err := endpoint.HoldStep(ctx, sfs, snap2, jobId) // no shadow
 		require.NoError(ctx, err)
-		return []endpoint.Abstraction{snap2Cursor, snap1Hold, snap2Hold}
+		// create artificial tentative cursor
+		snap3TentativeCursor, err := endpoint.CreateTentativeReplicationCursor(ctx, sfs, snap3, jobId)
+		require.NoError(ctx, err)
+		return []endpoint.Abstraction{snap2Cursor, snap1Hold, snap2Hold, snap3TentativeCursor}
 	}
 	createArtificalStaleAbstractions(sjid)
 	ojidSendAbstractions := createArtificalStaleAbstractions(ojid)
@@ -333,21 +336,29 @@ func implReplicationIncrementalCleansUpStaleAbstractions(ctx *platformtest.Conte
 		require.NoError(ctx, err)
 		snap2OjidCursorName, err := endpoint.ReplicationCursorBookmarkName(sfs, snap2.Guid, ojid)
 		require.NoError(ctx, err)
+		snap3SjidTentativeCursorName, err := endpoint.TentativeReplicationCursorBookmarkName(sfs, snap3.Guid, sjid)
+		require.NoError(ctx, err)
+		snap3OjidTentativeCursorName, err := endpoint.TentativeReplicationCursorBookmarkName(sfs, snap3.Guid, ojid)
+		require.NoError(ctx, err)
 		var bmNames []string
 		for _, bm := range sBms {
 			bmNames = append(bmNames, bm.Name)
 		}

 		if invalidateCacheBeforeSecondReplication {
-			require.Len(ctx, sBms, 3)
+			require.Len(ctx, sBms, 4)
 			require.Contains(ctx, bmNames, snap5SjidCursorName)
 			require.Contains(ctx, bmNames, snap2OjidCursorName)
+			require.Contains(ctx, bmNames, snap3OjidTentativeCursorName)
 			require.Contains(ctx, bmNames, "2")
 		} else {
-			require.Len(ctx, sBms, 4)
+			require.Len(ctx, sBms, 6)
+			ctx.Logf("%s", pretty.Sprint(sBms))
 			require.Contains(ctx, bmNames, snap5SjidCursorName)
 			require.Contains(ctx, bmNames, snap2SjidCursorName)
 			require.Contains(ctx, bmNames, snap2OjidCursorName)
+			require.Contains(ctx, bmNames, snap3SjidTentativeCursorName)
+			require.Contains(ctx, bmNames, snap3OjidTentativeCursorName)
 			require.Contains(ctx, bmNames, "2")
 		}
 	}
@@ -370,6 +381,84 @@ func implReplicationIncrementalCleansUpStaleAbstractions(ctx *platformtest.Conte

 }

+func ReplicationIncrementalHandlesFromVersionEqTentativeCursorCorrectly(ctx *platformtest.Context) {
+
+	platformtest.Run(ctx, platformtest.PanicErr, ctx.RootDataset, `
+		CREATEROOT
+		+  "sender"
+		+  "sender@1"
+		+  "receiver"
+		R  zfs create -p "${ROOTDS}/receiver/${ROOTDS}"
+	`)
+
+	sjid := endpoint.MustMakeJobID("sender-job")
+	rjid := endpoint.MustMakeJobID("receiver-job")
+
+	sfs := ctx.RootDataset + "/sender"
+	rfsRoot := ctx.RootDataset + "/receiver"
+
+	rep := replicationInvocation{
+		sjid:    sjid,
+		rjid:    rjid,
+		sfs:     sfs,
+		rfsRoot: rfsRoot,
+		// It doesn't really matter what guarantee we use here, as the second replication will configure another.
+		// But, in the real world, the only way for a stale tentative cursor to appear is if the guarantee is set to
+		// incremental replication and we crash before converting the tentative cursor into a regular cursor.
+		guarantee: pdu.ReplicationConfigProtectionWithKind(pdu.ReplicationGuaranteeKind_GuaranteeIncrementalReplication),
+	}
+
+	// Do initial replication to set up the test.
+	rep1 := rep.Do(ctx)
+	ctx.Logf("\n%s", pretty.Sprint(rep1))
+	sfsDs := mustDatasetPath(sfs)
+	snap1_sender := mustGetFilesystemVersion(ctx, sfs+"@1")
+	snap1_replicationCursor_name, err := endpoint.ReplicationCursorBookmarkName(sfs, snap1_sender.Guid, sjid)
+	require.NoError(ctx, err)
+	snap1_replicationCursor := mustGetFilesystemVersion(ctx, sfs+"#"+snap1_replicationCursor_name)
+
+	// The second replication will be done with a guarantee kind that doesn't create tentative cursors by itself.
+	// So, it would generally be right to clean up any tentative cursors on sfs since they're stale abstractions.
+	// However, if the cursor is used as the `from` version in any send step, we must not destroy it, as that
+	// would break incremental replication.
+	// NB: we only need to test the first step as all subsequent steps will be snapshot->snapshot.
+	rep.guarantee = pdu.ReplicationConfigProtectionWithKind(pdu.ReplicationGuaranteeKind_GuaranteeNothing)
+	// create the artificial cursor
+	snap1_tentativeCursor, err := endpoint.CreateTentativeReplicationCursor(ctx, sfs, snap1_sender, sjid)
+	require.NoError(ctx, err)
+	endpoint.AbstractionsCacheInvalidate(sfs)
+	// remove other bookmarks of snap1, and snap1 itself, to force the replication planner to use the tentative cursor
+	err = zfs.ZFSDestroyFilesystemVersion(ctx, sfsDs, &snap1_sender)
+	require.NoError(ctx, err)
+	err = zfs.ZFSDestroyFilesystemVersion(ctx, sfsDs, &snap1_replicationCursor)
+	require.NoError(ctx, err)
+	versions, err := zfs.ZFSListFilesystemVersions(ctx, sfsDs, zfs.ListFilesystemVersionsOptions{})
+	require.NoError(ctx, err)
+	require.Len(ctx, versions, 1)
+	require.Equal(ctx, versions[0].Guid, snap1_tentativeCursor.GetFilesystemVersion().Guid)
+	// create another snapshot so that replication does one incremental step `tentative_cursor` -> `@2`
+	mustSnapshot(ctx, sfs+"@2")
+	mustGetFilesystemVersion(ctx, sfs+"@2")
+	// do the replication
+	rep2 := rep.Do(ctx)
+	ctx.Logf("\n%s", pretty.Sprint(rep2))
+
+	// Ensure that the tentative cursor was used.
+	require.Len(ctx, rep2.Attempts, 1)
+	require.Equal(ctx, rep2.Attempts[0].State, report.AttemptDone)
+	require.Len(ctx, rep2.Attempts[0].Filesystems, 1)
+	require.Nil(ctx, rep2.Attempts[0].Filesystems[0].Error())
+	require.Len(ctx, rep2.Attempts[0].Filesystems[0].Steps, 1)
+	require.EqualValues(ctx, rep2.Attempts[0].Filesystems[0].CurrentStep, 1)
+	require.Len(ctx, rep2.Attempts[0].Filesystems[0].Steps, 1)
+	require.Equal(ctx, rep2.Attempts[0].Filesystems[0].Steps[0].Info.From, snap1_tentativeCursor.GetFilesystemVersion().RelName())
+
+	// Ensure that the tentative cursor was destroyed as part of SendPost.
+	_, err = zfs.ZFSGetFilesystemVersion(ctx, snap1_replicationCursor.FullPath(sfs))
+	_, ok := err.(*zfs.DatasetDoesNotExist)
+	require.True(ctx, ok)
+}
+
 type PartialSender struct {
 	*endpoint.Sender
 	failAfterByteCount int64
@@ -33,11 +33,6 @@ func NewKeepLastN(n int, regex string) (*KeepLastN, error) {
 }

 func (k KeepLastN) KeepRule(snaps []Snapshot) (destroyList []Snapshot) {
-
-	if k.n > len(snaps) {
-		return []Snapshot{}
-	}
-
 	matching, notMatching := partitionSnapList(snaps, func(snapshot Snapshot) bool {
 		return k.re.MatchString(snapshot.Name())
 	})
@@ -90,7 +90,7 @@ func TestKeepLastN(t *testing.T) {
 				stubSnap{"a2", false, o(12)},
 			},
 			rules: []KeepRule{
-				MustKeepLastN(3, "a"),
+				MustKeepLastN(4, "a"),
 			},
 			expDestroy: map[string]bool{
 				"b1": true,
@@ -797,7 +797,7 @@ func (a *attempt) errorReport() *errorReport {
 			r.byClass[class] = errs
 		}
 		for _, err := range r.flattened {
-			if neterr, ok := err.Err.(net.Error); ok && neterr.Temporary() {
+			if neterr, ok := err.Err.(net.Error); ok && neterr.Timeout() {
 				putClass(err, errorClassTemporaryConnectivityRelated)
 				continue
 			}
@@ -13,7 +13,7 @@ func init() {
 	}
 }

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func debug(format string, args ...interface{}) {
 	if debugEnabled {
 		fmt.Fprintf(os.Stderr, "repl: driver: %s\n", fmt.Sprintf(format, args...))
@@ -22,7 +22,7 @@ func debug(format string, args ...interface{}) {

 type debugFunc func(format string, args ...interface{})

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func debugPrefix(prefixFormat string, prefixFormatArgs ...interface{}) debugFunc {
 	prefix := fmt.Sprintf(prefixFormat, prefixFormatArgs...)
 	return func(format string, args ...interface{}) {
@@ -192,3 +192,14 @@ func (r *Report) GetFailedFilesystemsCountInLatestAttempt() int {
 		return 0
 	}
 }
+
+// Returns true in case the AttemptState is a terminal
+// state(AttemptPlanningError, AttemptFanOutError, AttemptDone)
+func (a AttemptState) IsTerminal() bool {
+	switch a {
+	case AttemptPlanningError, AttemptFanOutError, AttemptDone:
+		return true
+	default:
+		return false
+	}
+}
@@ -14,7 +14,7 @@ func init() {
 	}
 }

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func debug(format string, args ...interface{}) {
 	if debugEnabled {
 		fmt.Fprintf(os.Stderr, "rpc/dataconn: %s\n", fmt.Sprintf(format, args...))
@@ -24,6 +24,9 @@ func (e HeartbeatTimeout) Error() string {
 	return "heartbeat timeout"
 }

+// This function is deprecated in net.Error and since this
+// function is not involved in .Accept() code path, nothing
+// really needs this method to be here.
 func (e HeartbeatTimeout) Temporary() bool { return true }

 func (e HeartbeatTimeout) Timeout() bool { return true }
@@ -13,7 +13,7 @@ func init() {
 	}
 }

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func debug(format string, args ...interface{}) {
 	if debugEnabled {
 		fmt.Fprintf(os.Stderr, "rpc/dataconn/heartbeatconn: %s\n", fmt.Sprintf(format, args...))
@@ -6,12 +6,12 @@
 // In commit 082335df5d85e1b0b9faa35ff182c71886142d3e and earlier, heartbeatconn would fail
 // this benchmark with a writev I/O timeout (here the ss(8) output at the time of failure)
 //
-//  ESTAB        33369    0                                        127.0.0.1:12345                                                127.0.0.1:57282                    users:(("heartbeatconn_i",pid=25953,fd=5))
-//	 cubic wscale:7,7 rto:203 rtt:2.992/5.849 ato:162 mss:32768 pmtu:65535 rcvmss:32741 advmss:65483 cwnd:10 bytes_sent:48 bytes_acked:48 bytes_received:195401 segs_out:44 segs_in:57 data_segs_out:6 data_segs_in:34 send 876.1Mbps lastsnd:125 lastrcv:9390 lastack:125 pacing_rate 1752.0Mbps delivery_rate 6393.8Mbps delivered:7 app_limited busy:42ms rcv_rtt:1 rcv_space:65483 rcv_ssthresh:65483 minrtt:0.029
-//	 --
-//	 ESTAB        0        3956805                                  127.0.0.1:57282                                                127.0.0.1:12345                    users:(("heartbeatconn_i",pid=26100,fd=3))
-//		  cubic wscale:7,7 rto:211 backoff:5 rtt:10.38/16.937 ato:40 mss:32768 pmtu:65535 rcvmss:536 advmss:65483 cwnd:10 bytes_sent:195401 bytes_acked:195402 bytes_received:48 segs_out:57 segs_in:45 data_segs_out:34 data_segs_in:6 send 252.5Mbps lastsnd:9390 lastrcv:125 lastack:125 pacing_rate 505.1Mbps delivery_rate 1971.0Mbps delivered:35 busy:30127ms rwnd_limited:30086ms(99.9%) rcv_space:65495 rcv_ssthresh:65495 notsent:3956805 minrtt:0.007
-//	 panic: writev tcp 127.0.0.1:57282->127.0.0.1:12345: i/o timeout
+//	 ESTAB        33369    0                                        127.0.0.1:12345                                                127.0.0.1:57282                    users:(("heartbeatconn_i",pid=25953,fd=5))
+//		 cubic wscale:7,7 rto:203 rtt:2.992/5.849 ato:162 mss:32768 pmtu:65535 rcvmss:32741 advmss:65483 cwnd:10 bytes_sent:48 bytes_acked:48 bytes_received:195401 segs_out:44 segs_in:57 data_segs_out:6 data_segs_in:34 send 876.1Mbps lastsnd:125 lastrcv:9390 lastack:125 pacing_rate 1752.0Mbps delivery_rate 6393.8Mbps delivered:7 app_limited busy:42ms rcv_rtt:1 rcv_space:65483 rcv_ssthresh:65483 minrtt:0.029
+//		 --
+//		 ESTAB        0        3956805                                  127.0.0.1:57282                                                127.0.0.1:12345                    users:(("heartbeatconn_i",pid=26100,fd=3))
+//			  cubic wscale:7,7 rto:211 backoff:5 rtt:10.38/16.937 ato:40 mss:32768 pmtu:65535 rcvmss:536 advmss:65483 cwnd:10 bytes_sent:195401 bytes_acked:195402 bytes_received:48 segs_out:57 segs_in:45 data_segs_out:34 data_segs_in:6 send 252.5Mbps lastsnd:9390 lastrcv:125 lastack:125 pacing_rate 505.1Mbps delivery_rate 1971.0Mbps delivered:35 busy:30127ms rwnd_limited:30086ms(99.9%) rcv_space:65495 rcv_ssthresh:65495 notsent:3956805 minrtt:0.007
+//		 panic: writev tcp 127.0.0.1:57282->127.0.0.1:12345: i/o timeout
 //
 // The assumed reason for those writev timeouts is the following:
 // - Sporadic server stalls (sever data handling, usually I/O) cause TCP exponential backoff on the client for client->server
@@ -22,27 +22,23 @@
 // The fix contained in the commit this message was committed with resets the deadline whenever
 // a heartbeat is received from the server.
 //
-//
 // How to run this integration test:
 //
+//	Terminal 1:
+//	   $ ZREPL_RPC_DATACONN_HEARTBEATCONN_DEBUG=1 go run heartbeatconn_integration_variablereceiverate.go  -mode server -addr 127.0.0.1:12345
+//	   rpc/dataconn/heartbeatconn: send heartbeat
+//	   rpc/dataconn/heartbeatconn: send heartbeat
+//	   ...
 //
-//  Terminal 1:
-//     $ ZREPL_RPC_DATACONN_HEARTBEATCONN_DEBUG=1 go run heartbeatconn_integration_variablereceiverate.go  -mode server -addr 127.0.0.1:12345
-//     rpc/dataconn/heartbeatconn: send heartbeat
-//     rpc/dataconn/heartbeatconn: send heartbeat
-//     ...
-//
-//  Terminal 2:
-//     $ ZREPL_RPC_DATACONN_HEARTBEATCONN_DEBUG=1 go run heartbeatconn_integration_variablereceiverate.go  -mode  client -addr 127.0.0.1:12345
-//     rpc/dataconn/heartbeatconn: received heartbeat, resetting write timeout
-//     rpc/dataconn/heartbeatconn: renew frameconn write timeout returned errT=<nil> err=%!s(<nil>)
-//     rpc/dataconn/heartbeatconn: send heartbeat
-//     rpc/dataconn/heartbeatconn: received heartbeat, resetting write timeout
-//     rpc/dataconn/heartbeatconn: renew frameconn write timeout returned errT=<nil> err=%!s(<nil>)
-//     rpc/dataconn/heartbeatconn: received heartbeat, resetting write timeout
-//     ...
-//
-//  You should observe
+//	Terminal 2:
+//	   $ ZREPL_RPC_DATACONN_HEARTBEATCONN_DEBUG=1 go run heartbeatconn_integration_variablereceiverate.go  -mode  client -addr 127.0.0.1:12345
+//	   rpc/dataconn/heartbeatconn: received heartbeat, resetting write timeout
+//	   rpc/dataconn/heartbeatconn: renew frameconn write timeout returned errT=<nil> err=%!s(<nil>)
+//	   rpc/dataconn/heartbeatconn: send heartbeat
+//	   rpc/dataconn/heartbeatconn: received heartbeat, resetting write timeout
+//	   rpc/dataconn/heartbeatconn: renew frameconn write timeout returned errT=<nil> err=%!s(<nil>)
+//	   rpc/dataconn/heartbeatconn: received heartbeat, resetting write timeout
+//	   ...
 package main

 import (
@@ -2,15 +2,13 @@
 //
 // With stdin / stdout on client and server, simulating zfs send|recv piping
 //
-//   ./microbenchmark -appmode server  | pv -r > /dev/null
-//   ./microbenchmark -appmode client -direction recv < /dev/zero
-//
+//	./microbenchmark -appmode server  | pv -r > /dev/null
+//	./microbenchmark -appmode client -direction recv < /dev/zero
 //
 // Without the overhead of pipes (just protocol performance, mostly useful with perf bc no bw measurement)
 //
-//   ./microbenchmark -appmode client -direction recv -devnoopWriter -devnoopReader
-//   ./microbenchmark -appmode server -devnoopReader -devnoopWriter
-//
+//	./microbenchmark -appmode client -direction recv -devnoopWriter -devnoopReader
+//	./microbenchmark -appmode server -devnoopReader -devnoopWriter
 package main

 import (
@@ -29,7 +29,7 @@ func WithLogger(ctx context.Context, log Logger) context.Context {
 	return context.WithValue(ctx, contextKeyLogger, log)
 }

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func getLog(ctx context.Context) Logger {
 	log, ok := ctx.Value(contextKeyLogger).(Logger)
 	if !ok {
@@ -181,23 +181,19 @@ func (e *ReadStreamError) Error() string {

 var _ net.Error = &ReadStreamError{}

-func (e ReadStreamError) netErr() net.Error {
-	if netErr, ok := e.Err.(net.Error); ok {
-		return netErr
-	}
-	return nil
-}
-
 func (e ReadStreamError) Timeout() bool {
-	if netErr := e.netErr(); netErr != nil {
+	if netErr, ok := e.Err.(net.Error); ok {
 		return netErr.Timeout()
 	}
 	return false
 }

+// This function is deprecated in net.Error and since this
+// function is not involved in .Accept() code path, nothing
+// really needs this method to be here.
 func (e ReadStreamError) Temporary() bool {
-	if netErr := e.netErr(); netErr != nil {
-		return netErr.Temporary()
+	if te, ok := e.Err.(interface{ Temporary() bool }); ok {
+		return te.Temporary()
 	}
 	return false
 }
@@ -232,7 +232,12 @@ var _ net.Error = (*closeStateErrConnectionClosed)(nil)
 func (e *closeStateErrConnectionClosed) Error() string {
 	return "connection closed"
 }
-func (e *closeStateErrConnectionClosed) Timeout() bool   { return false }
+
+func (e *closeStateErrConnectionClosed) Timeout() bool { return false }
+
+// This function is deprecated in net.Error and since this
+// function is not involved in .Accept() code path, nothing
+// really needs this method to be here.
 func (e *closeStateErrConnectionClosed) Temporary() bool { return false }

 func (s *closeState) CloseEntry() error {
@@ -13,7 +13,7 @@ func init() {
 	}
 }

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func debug(format string, args ...interface{}) {
 	if debugEnabled {
 		fmt.Fprintf(os.Stderr, "rpc/dataconn/stream: %s\n", fmt.Sprintf(format, args...))
@@ -107,34 +107,31 @@ func (c *Conn) RenewWriteDeadline() error {
 	return c.SetWriteDeadline(time.Now().Add(c.idleTimeout))
 }

-func (c *Conn) Read(p []byte) (n int, err error) {
+func (c *Conn) Read(p []byte) (n int, _ error) {
 	n = 0
-	err = nil
 restart:
 	if err := c.renewReadDeadline(); err != nil {
 		return n, err
 	}
 	var nCurRead int
-	nCurRead, err = c.Wire.Read(p[n:])
+	nCurRead, err := c.Wire.Read(p[n:])
 	n += nCurRead
 	if netErr, ok := err.(net.Error); ok && netErr.Timeout() && nCurRead > 0 {
-		err = nil
 		goto restart
 	}
 	return n, err
 }

-func (c *Conn) Write(p []byte) (n int, err error) {
+func (c *Conn) Write(p []byte) (n int, _ error) {
 	n = 0
 restart:
 	if err := c.RenewWriteDeadline(); err != nil {
 		return n, err
 	}
 	var nCurWrite int
-	nCurWrite, err = c.Wire.Write(p[n:])
+	nCurWrite, err := c.Wire.Write(p[n:])
 	n += nCurWrite
 	if netErr, ok := err.(net.Error); ok && netErr.Timeout() && nCurWrite > 0 {
-		err = nil
 		goto restart
 	}
 	return n, err
@@ -144,17 +141,16 @@ restart:
 // but is guaranteed to use the writev system call if the wrapped Wire
 // support it.
 // Note the Conn does not support writev through io.Copy(aConn, aNetBuffers).
-func (c *Conn) WritevFull(bufs net.Buffers) (n int64, err error) {
+func (c *Conn) WritevFull(bufs net.Buffers) (n int64, _ error) {
 	n = 0
 restart:
 	if err := c.RenewWriteDeadline(); err != nil {
 		return n, err
 	}
 	var nCurWrite int64
-	nCurWrite, err = io.Copy(c.Wire, &bufs)
+	nCurWrite, err := io.Copy(c.Wire, &bufs)
 	n += nCurWrite
 	if netErr, ok := err.(net.Error); ok && netErr.Timeout() && nCurWrite > 0 {
-		err = nil
 		goto restart
 	}
 	return n, err
@@ -1,25 +1,5 @@
 // Package netadaptor implements an adaptor from
 // transport.AuthenticatedListener to net.Listener.
-//
-// In contrast to transport.AuthenticatedListener,
-// net.Listener is commonly expected (e.g. by net/http.Server.Serve),
-// to return errors that fulfill the Temporary interface:
-//   interface Temporary { Temporary() bool }
-// Common behavior of packages consuming net.Listener is to return
-// from the serve-loop if an error returned by Accept is not Temporary,
-// i.e., does not implement the interface or is !net.Error.Temporary().
-//
-// The zrepl transport infrastructure was written with the
-// idea that Accept() may return any kind of error, and the consumer
-// would just log the error and continue calling Accept().
-// We have to adapt these listeners' behavior to the expectations
-// of net/http.Server.
-//
-// Hence, Listener does not return an error at all but blocks the
-// caller of Accept() until we get a (successfully authenticated)
-// connection without errors from the transport.
-// Accept errors returned from the transport are logged as errors
-// to the logger passed on initialization.
 package netadaptor

 import (
@@ -188,8 +188,8 @@ func (c *Client) WaitForConnectivity(ctx context.Context) error {
 			data, dataErr := c.dataClient.ReqPing(ctx, &req)
 			// dataClient uses transport.Connecter, which doesn't expose WaitForReady(true)
 			// => we need to mask dial timeouts
-			if err, ok := dataErr.(interface{ Temporary() bool }); ok && err.Temporary() {
-				// Rate-limit pings here in case Temporary() is a mis-classification
+			if err, ok := dataErr.(interface{ Timeout() bool }); ok && err.Timeout() {
+				// Rate-limit pings here in case Timeout() == true is a mis-classification
 				// or returns immediately (this is a tight loop in that case)
 				// TODO keep this in lockstep with controlClient
 				// 		=> don't use FailFast for control, but check that both control and data worked
@@ -14,7 +14,7 @@ func init() {
 	}
 }

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func debug(format string, args ...interface{}) {
 	if debugEnabled {
 		fmt.Fprintf(os.Stderr, "rpc: %s\n", fmt.Sprintf(format, args...))
@@ -3,7 +3,7 @@
 // The zrepl documentation refers to the client as the
 // `active side` and to the server as the `passive side`.
 //
-// Design Considerations
+// # Design Considerations
 //
 // zrepl has non-standard requirements to remote procedure calls (RPC):
 // whereas the coordination of replication (the planning phase) mostly
@@ -35,7 +35,7 @@
 //
 // Hence, this package attempts to combine the best of both worlds:
 //
-// GRPC for Coordination and Dataconn for Bulk Data Transfer
+// # GRPC for Coordination and Dataconn for Bulk Data Transfer
 //
 // This package's Client uses its transport.Connecter to maintain
 // separate control and data connections to the Server.
@@ -47,68 +47,66 @@
 // The following ASCII diagram gives an overview of how the individual
 // building blocks are glued together:
 //
-//                    +------------+
-//                    | rpc.Client |
-//                    +------------+
-//                      |         |
-//             +--------+         +------------+
-//             |                               |
-//   +---------v-----------+          +--------v------+
-//   |pdu.ReplicationClient|          |dataconn.Client|
-//   +---------------------+          +--------v------+
-//             | label:            label:      |
-//             | zrepl_control      zrepl_data |
-//             +--------+         +------------+
-//                      |         |
-//                   +--v---------v---+
-//                   |  transportmux  |
-//                   +-------+--------+
-//                           | uses
-//                   +-------v--------+
-//                   |versionhandshake|
-//                   +-------+--------+
-//                           | uses
-//                    +------v------+
-//                    |  transport  |
-//                    +------+------+
-//                           |
-//                        NETWORK
-//                           |
-//                    +------+------+
-//                    |  transport  |
-//                    +------^------+
-//                           | uses
-//                   +-------+--------+
-//                   |versionhandshake|
-//                   +-------^--------+
-//                           | uses
-//                   +-------+--------+
-//                   |  transportmux  |
-//                   +--^--------^----+
-//                      |        |
-//             +--------+        --------------+       ---
-//             |                               |        |
-//             | label:            label:      |        |
-//             | zrepl_control      zrepl_data |        |
-//       +-----+----+              +-----------+---+    |
-//       |netadaptor|              |dataconn.Server|    | rpc.Server
-//       |     +    |              +------+--------+    |
-//       |grpcclient|                     |             |
-//       |identity  |                     |             |
-//       +-----+----+                     |             |
-//             |                          |             |
-//   +---------v-----------+              |             |
-//   |pdu.ReplicationServer|              |             |
-//   +---------+-----------+              |             |
-//             |                          |            ---
-//             +----------+  +------------+
-//                        |  |
-//                    +---v--v-----+
-//                    |  Handler   |
-//                    +------------+
-//           (usually endpoint.{Sender,Receiver})
-//
-//
+//	                 +------------+
+//	                 | rpc.Client |
+//	                 +------------+
+//	                   |         |
+//	          +--------+         +------------+
+//	          |                               |
+//	+---------v-----------+          +--------v------+
+//	|pdu.ReplicationClient|          |dataconn.Client|
+//	+---------------------+          +--------v------+
+//	          | label:            label:      |
+//	          | zrepl_control      zrepl_data |
+//	          +--------+         +------------+
+//	                   |         |
+//	                +--v---------v---+
+//	                |  transportmux  |
+//	                +-------+--------+
+//	                        | uses
+//	                +-------v--------+
+//	                |versionhandshake|
+//	                +-------+--------+
+//	                        | uses
+//	                 +------v------+
+//	                 |  transport  |
+//	                 +------+------+
+//	                        |
+//	                     NETWORK
+//	                        |
+//	                 +------+------+
+//	                 |  transport  |
+//	                 +------^------+
+//	                        | uses
+//	                +-------+--------+
+//	                |versionhandshake|
+//	                +-------^--------+
+//	                        | uses
+//	                +-------+--------+
+//	                |  transportmux  |
+//	                +--^--------^----+
+//	                   |        |
+//	          +--------+        --------------+       ---
+//	          |                               |        |
+//	          | label:            label:      |        |
+//	          | zrepl_control      zrepl_data |        |
+//	    +-----+----+              +-----------+---+    |
+//	    |netadaptor|              |dataconn.Server|    | rpc.Server
+//	    |     +    |              +------+--------+    |
+//	    |grpcclient|                     |             |
+//	    |identity  |                     |             |
+//	    +-----+----+                     |             |
+//	          |                          |             |
+//	+---------v-----------+              |             |
+//	|pdu.ReplicationServer|              |             |
+//	+---------+-----------+              |             |
+//	          |                          |            ---
+//	          +----------+  +------------+
+//	                     |  |
+//	                 +---v--v-----+
+//	                 |  Handler   |
+//	                 +------------+
+//	        (usually endpoint.{Sender,Receiver})
 package rpc

 // edit trick for the ASCII art above:
@@ -9,7 +9,7 @@ import (

 type Logger = logger.Logger

-/// All fields must be non-nil
+// All fields must be non-nil
 type Loggers struct {
 	General Logger
 	Control Logger
@@ -33,8 +33,47 @@ var _ net.Error = &HandshakeError{}

 func (e HandshakeError) Error() string { return e.msg }

-// Like with net.OpErr (Go issue 6163), a client failing to handshake
-// should be a temporary Accept error toward the Listener .
+// When a net.Listener.Accept() returns an error, the server must
+// decide whether to retry calling Accept() or not.
+// On some platforms (e.g., Linux), Accept() can return errors
+// related to the specific protocol connection that was supposed
+// to be returned as asocket FD. Obviously, we want to ignore,
+// maybe log, those errors and retry Accept() immediately to
+// serve other connections.
+// But there are also conditions where we get Accept() errors because
+// the process has run out of file descriptors. In that case, retrying
+// won't help. We need to close some file descriptor to make progress.
+// Note that there could be lots of open file descriptors because we
+// have accepted, and not yet closed, lots of connections in the past.
+// And then, of course there can be errors where we just want
+// to return, e.g., if there's a programming error and we're getting
+// an EBADFD or whatever.
+//
+// So, the serve loops in net/http.Server.Serve() or gRPC's server.Serve()
+// must inspect the error and decide what to do.
+// The vehicle for this is the
+//
+//	interface { Temporary() bool }
+//
+// Behavior in both of the aforementioned Serve() loops:
+//
+//   - if the error doesn't implement the interface, stop serving and return
+//   - `Temporary() == true`: retry with back-off
+//   - `Temporary() == false`: stop serving and return
+//
+// So, to make this package's HandshakeListener work with these
+// Serve() loops, we return Temporary() == true if the handshake fails.
+// In the aforementioned categories, that's the case of a per-connection
+// protocol error.
+//
+// Note: the net.Error interface has deprecated the Temporary() method
+// in go.dev/issue/45729, but there is no replacement for users of .Accept().
+// Existing users of .Accept() continue to check for the interface.
+// So, we need to continue supporting Temporary() until there's a different
+// mechanism for serve loops to decide whether to retry or not.
+// The following mailing list post proposes to eliminate the retries
+// completely, but it seems like the effort has stalled.
+// https://groups.google.com/g/golang-nuts/c/-JcZzOkyqYI/m/xwaZzjCgAwAJ
 func (e HandshakeError) Temporary() bool {
 	if e.isAcceptError {
 		return true
@@ -3,13 +3,12 @@
 //
 // Intended Usage
 //
-//   defer s.lock().unlock()
-//   // drop lock while waiting for wait group
-//   func() {
-//	     defer a.l.Unlock().Lock()
-//	     fssesDone.Wait()
-//	 }()
-//
+//	  defer s.lock().unlock()
+//	  // drop lock while waiting for wait group
+//	  func() {
+//		     defer a.l.Unlock().Lock()
+//		     fssesDone.Wait()
+//		 }()
 package chainlock

 import "sync"
@@ -7,26 +7,25 @@
 //
 // Example:
 //
-//  type Config struct {
-//      // This field must be set to a non-nil value,
-//      // forcing the caller to make their mind up
-//      // about this field.
-//      CriticalSetting   *nodefault.Bool
-//  }
+//	type Config struct {
+//	    // This field must be set to a non-nil value,
+//	    // forcing the caller to make their mind up
+//	    // about this field.
+//	    CriticalSetting   *nodefault.Bool
+//	}
 //
 // An function that takes such a Config should _not_ check for nil-ness:
 // and instead unconditionally dereference:
 //
-//  func f(c Config) {
-//      if (c.CriticalSetting) { }
-//  }
+//	func f(c Config) {
+//	    if (c.CriticalSetting) { }
+//	}
 //
 // If the caller of f forgot to specify the .CriticalSetting
 // field, the Go runtime will issue a nil-pointer deref panic
 // and it'll be clear that the caller did not read the docs of Config.
 //
-//  f(Config{}) // crashes
-//
-//  f Config{ CriticalSetting: &nodefault.Bool{B: false}} // doesn't crash
+//	f(Config{}) // crashes
 //
+//	f Config{ CriticalSetting: &nodefault.Bool{B: false}} // doesn't crash
 package nodefault
@@ -0,0 +1,96 @@
+package suspendresumesafetimer
+
+import (
+	"context"
+	"time"
+
+	"github.com/zrepl/zrepl/util/envconst"
+)
+
+// The returned error is guaranteed to be the ctx.Err()
+func SleepUntil(ctx context.Context, sleepUntil time.Time) error {
+
+	// We use .Round(0) to strip the monotonic clock reading from the time.Time
+	// returned by time.Now(). That will make the before/after check in the ticker
+	// for-loop compare wall-clock times instead of monotonic time.
+	// Comparing wall clock time is necessary because monotonic time does not progress
+	// while the system is suspended.
+	//
+	// Background
+	//
+	// A time.Time carries a wallclock timestamp and optionally a monotonic clock timestamp.
+	// time.Now() returns a time.Time that carries both.
+	// time.Time.Add() applies the same delta to both timestamps in the time.Time.
+	// x.Sub(y) will return the *monotonic* delta if both x and y carry a monotonic timestamp.
+	// time.Until(x) == x.Sub(now) where `now` will have a monotonic timestamp.
+	// So, time.Until(x) with an `x` that has monotonic timestamp will return monotonic delta.
+	//
+	// Why Do We Care?
+	//
+	// On systems that suspend/resume, wall clock time progresses during suspend but
+	// monotonic time does not.
+	//
+	// So, suppose the following sequence of events:
+	//   x <== time.Now()
+	//   System suspends for 1 hour
+	//   delta <== time.Now().Sub(x)
+	// `delta` will be near 0 because time.Until() subtracts the monotonic
+	// timestamps, and monotonic time didn't progress during suspend.
+	//
+	// Now strip the timestamp using .Round(0)
+	//   x <== time.Now().Round(0)
+	//   System suspends for 1 hour
+	//   delta <== time.Now().Sub(x)
+	// `delta` will be 1 hour because time.Sub() subtracted wallclock timestamps
+	// because x didn't have a monotonic timestamp because we stripped it using .Round(0).
+	//
+	//
+	sleepUntil = sleepUntil.Round(0)
+
+	// Set up a timer so that, if the system doesn't suspend/resume,
+	// we get a precise wake-up time from the native Go timer.
+	monotonicClockTimer := time.NewTimer(time.Until(sleepUntil))
+	defer func() {
+		if !monotonicClockTimer.Stop() {
+			// non-blocking read since we can come here when
+			// we've already drained the channel through
+			//		case <-monotonicClockTimer.C
+			// in the `for` loop below.
+			select {
+			case <-monotonicClockTimer.C:
+			default:
+			}
+		}
+	}()
+
+	// Set up a ticker so that we're guaranteed to wake up periodically.
+	// We'll then get the current wall-clock time and check ourselves
+	// whether we're past the requested expiration time.
+	// Pick a 10 second check interval by default since it's rare that
+	// suspend/resume is done more frequently.
+	ticker := time.NewTicker(envconst.Duration("ZREPL_WALLCLOCKTIMER_MAX_DELAY", 10*time.Second))
+	defer ticker.Stop()
+
+	for {
+		select {
+		case <-monotonicClockTimer.C:
+			return nil
+		case <-ticker.C:
+			now := time.Now()
+			if now.Before(sleepUntil) {
+				// Continue waiting.
+
+				// Reset the monotonic timer to reset drift.
+				if !monotonicClockTimer.Stop() {
+					<-monotonicClockTimer.C
+				}
+				monotonicClockTimer.Reset(time.Until(sleepUntil))
+
+				continue
+			}
+			return nil
+		case <-ctx.Done():
+			return ctx.Err()
+		}
+	}
+}
@@ -3,12 +3,20 @@ package zfs
 import (
 	"context"
 	"fmt"
+
+	"github.com/zrepl/zrepl/zfs/zfscmd"
 )

 type DatasetFilter interface {
 	Filter(p *DatasetPath) (pass bool, err error)
+	// The caller owns the returned set.
+	// Implementations should return a copy.
+	UserSpecifiedDatasets() UserSpecifiedDatasetsSet
 }

+// A set of dataset names that the user specified in the configuration file.
+type UserSpecifiedDatasetsSet map[string]bool
+
 // Returns a DatasetFilter that does not filter (passes all paths)
 func NoFilter() DatasetFilter {
 	return noFilter{}
@@ -18,7 +26,8 @@ type noFilter struct{}

 var _ DatasetFilter = noFilter{}

-func (noFilter) Filter(p *DatasetPath) (pass bool, err error) { return true, nil }
+func (noFilter) Filter(p *DatasetPath) (pass bool, err error)    { return true, nil }
+func (noFilter) UserSpecifiedDatasets() UserSpecifiedDatasetsSet { return nil }

 func ZFSListMapping(ctx context.Context, filter DatasetFilter) (datasets []*DatasetPath, err error) {
 	res, err := ZFSListMappingProperties(ctx, filter, nil)
@@ -61,6 +70,7 @@ func ZFSListMappingProperties(ctx context.Context, filter DatasetFilter, propert

 	go ZFSListChan(ctx, rchan, properties, nil, "-r", "-t", "filesystem,volume")

+	unmatchedUserSpecifiedDatasets := filter.UserSpecifiedDatasets()
 	datasets = make([]ZFSListMappingPropertiesResult, 0)
 	for r := range rchan {

@@ -74,6 +84,8 @@ func ZFSListMappingProperties(ctx context.Context, filter DatasetFilter, propert
 			return
 		}

+		delete(unmatchedUserSpecifiedDatasets, path.ToString())
+
 		pass, filterErr := filter.Filter(path)
 		if filterErr != nil {
 			return nil, fmt.Errorf("error calling filter: %s", filterErr)
@@ -87,5 +99,9 @@ func ZFSListMappingProperties(ctx context.Context, filter DatasetFilter, propert

 	}

+	jobid := zfscmd.GetJobIDOrDefault(ctx, "__nojobid")
+	metric := prom.ZFSListUnmatchedUserSpecifiedDatasetCount.WithLabelValues(jobid)
+	metric.Add(float64(len(unmatchedUserSpecifiedDatasets)))
+
 	return
 }
@@ -53,7 +53,6 @@ var componentValidChar = regexp.MustCompile(`^[0-9a-zA-Z-_\.: ]+$`)
 // characters:
 //
 //	[-_.: ]
-//
 func ComponentNamecheck(datasetPathComponent string) error {
 	if len(datasetPathComponent) == 0 {
 		return fmt.Errorf("path component must not be empty")
@@ -91,8 +90,8 @@ func (e *PathValidationError) Error() string {

 // combines
 //
-//    lib/libzfs/libzfs_dataset.c: zfs_validate_name
-//    module/zcommon/zfs_namecheck.c: entity_namecheck
+//	lib/libzfs/libzfs_dataset.c: zfs_validate_name
+//	module/zcommon/zfs_namecheck.c: entity_namecheck
 //
 // The '%' character is not allowed because it's reserved for zfs-internal use
 func EntityNamecheck(path string, t EntityType) (err *PathValidationError) {
@@ -3,10 +3,11 @@ package zfs
 import "github.com/prometheus/client_golang/prometheus"

 var prom struct {
-	ZFSListFilesystemVersionDuration *prometheus.HistogramVec
-	ZFSSnapshotDuration              *prometheus.HistogramVec
-	ZFSBookmarkDuration              *prometheus.HistogramVec
-	ZFSDestroyDuration               *prometheus.HistogramVec
+	ZFSListFilesystemVersionDuration          *prometheus.HistogramVec
+	ZFSSnapshotDuration                       *prometheus.HistogramVec
+	ZFSBookmarkDuration                       *prometheus.HistogramVec
+	ZFSDestroyDuration                        *prometheus.HistogramVec
+	ZFSListUnmatchedUserSpecifiedDatasetCount *prometheus.GaugeVec
 }

 func init() {
@@ -34,6 +35,15 @@ func init() {
 		Name:      "destroy_duration",
 		Help:      "Duration it took to destroy a dataset",
 	}, []string{"dataset_type", "filesystem"})
+	prom.ZFSListUnmatchedUserSpecifiedDatasetCount = prometheus.NewGaugeVec(prometheus.GaugeOpts{
+		Namespace: "zrepl",
+		Subsystem: "zfs",
+		Name:      "list_unmatched_user_specified_dataset_count",
+		Help: "When evaluating a DatsetFilter against zfs list output, this counter " +
+			"is incremented for every DatasetFilter rule that did not match any " +
+			"filesystem name in the zfs list output. Monitor for increases to detect filesystem " +
+			"filter rules that have no effect because they don't match any local filesystem.",
+	}, []string{"jobid"})
 }

 func PrometheusRegister(registry prometheus.Registerer) error {
@@ -49,5 +59,8 @@ func PrometheusRegister(registry prometheus.Registerer) error {
 	if err := registry.Register(prom.ZFSDestroyDuration); err != nil {
 		return err
 	}
+	if err := registry.Register(prom.ZFSListUnmatchedUserSpecifiedDatasetCount); err != nil {
+		return err
+	}
 	return nil
 }
@@ -343,7 +343,6 @@ const (

 type SendStream struct {
 	cmd          *zfscmd.Cmd
-	kill         context.CancelFunc
 	stdoutReader io.ReadCloser // not *os.File for mocking during platformtest
 	stderrBuf    *circlog.CircularLog

@@ -416,8 +415,29 @@ func (s *SendStream) killAndWait() error {
 	debug("sendStream: killAndWait enter")
 	defer debug("sendStream: killAndWait leave")

+	// ensure this function is only called once
+	if s.state != sendStreamOpen {
+		panic(s.state)
+	}
+
 	// send SIGKILL
-	s.kill()
+	// In an earlier version, we used the starting context.Context's cancel function
+	// for this, but in Go > 1.19, doing so will cause .Wait() to return the
+	// context cancel error instead of the *exec.ExitError.
+	err := s.cmd.Process().Kill()
+	if err != nil {
+		if err == os.ErrProcessDone {
+			// This can happen if
+			// (1) the process has already been .Wait()ed, or
+			// (2) some other goroutine cancels the context, likely further up
+			//     the context tree.
+			// Case (1) can't happen to us because we only call
+			// this function in sendStreamOpen state.
+			// In Case (2), it's still our job to .Wait(), so, fallthrough.
+		} else {
+			return err
+		}
+	}

 	// Close our read-end of the pipe.
 	//
@@ -963,10 +983,10 @@ func ZFSSend(ctx context.Context, sendArgs ZFSSendArgsValidated) (*SendStream, e

 	stream := &SendStream{
 		cmd:          cmd,
-		kill:         cancel,
 		stdoutReader: stdoutReader,
 		stderrBuf:    stderrBuf,
 	}
+	_ = cancel // the SendStream.killAndWait() will kill the process

 	return stream, nil
 }
@@ -1022,8 +1042,10 @@ func (s *DrySendInfo) unmarshalZFSOutput(output []byte) (err error) {
 }

 // unmarshal info line, looks like this:
-//   full	zroot/test/a@1	5389768
-//   incremental	zroot/test/a@1	zroot/test/a@2	5383936
+//
+//	full	zroot/test/a@1	5389768
+//	incremental	zroot/test/a@1	zroot/test/a@2	5383936
+//
 // => see test cases
 func (s *DrySendInfo) unmarshalInfoLine(l string) (regexMatched bool, err error) {

@@ -1855,7 +1877,6 @@ var ErrBookmarkCloningNotSupported = fmt.Errorf("bookmark cloning feature is not
 // unless a bookmark with the name `bookmark` exists and has the same idenitty (zfs.FilesystemVersionEqualIdentity)
 //
 // v must be validated by the caller
-//
 func ZFSBookmark(ctx context.Context, fs string, v FilesystemVersion, bookmark string) (bm FilesystemVersion, err error) {

 	bm = FilesystemVersion{
@@ -13,7 +13,7 @@ func init() {
 	}
 }

-//nolint[:deadcode,unused]
+//nolint:deadcode,unused
 func debug(format string, args ...interface{}) {
 	if debugEnabled {
 		fmt.Fprintf(os.Stderr, "zfs: %s\n", fmt.Sprintf(format, args...))
@@ -19,6 +19,10 @@ func WithJobID(ctx context.Context, jobID string) context.Context {
 	return context.WithValue(ctx, contextKeyJobID, jobID)
 }

+func GetJobIDOrDefault(ctx context.Context, def string) string {
+	return getJobIDOrDefault(ctx, def)
+}
+
 func getJobIDOrDefault(ctx context.Context, def string) string {
 	ret, ok := ctx.Value(contextKeyJobID).(string)
 	if !ok {
Author	SHA1	Message	Date
Christian Schwarz	beecb4b93d	WIP	2024-05-09 13:25:31 +00:00
Christian Schwarz	c8afaf83ab	generalize trigger kinds	2023-12-22 14:40:53 +00:00
Christian Schwarz	b0caa2d151	WIP: generic activation through + new interval-based replication trigger	2023-12-22 14:01:24 +00:00
Denis Shaposhnikov	ebc46cf1c0	Fix last_n keep rule (#691 ) (#750 ) From https://github.com/zrepl/zrepl/issues/691 The last_n prune rule keeps everything, regardless of if it matches the regex or not, if there are less than count snapshot. The expectation would be to never keep non-regex snapshots, regardless of number.	2023-12-22 13:38:14 +01:00
Denis Shaposhnikov	27012e5623	Allow same root_fs for different jobs: sinks and so on (#752 ) Because some jobs add client identity to root_fs and other jobs don't do that, we can't reliable detect overlapping of filesystems. And and the same time we need an ability to use equal or overlapped root_fs for different jobs. For instance see this config: ``` - name: "zdisk" type: "sink" root_fs: "zdisk/zrepl" serve: type: "local" listener_name: "zdisk" ``` and ``` - name: "remote-to-zdisk" type: "pull" connect: type: "tls" root_fs: "zdisk/zrepl/remote" ``` As you can see, two jobs have overlapped root_fs, but actually datasets are not overlapped, because job `zdisk` save everything under `zdisk/zrepl/localhost`, because it adds client identity. So they actually use two different filesystems: `zdisk/zrepl/localhost` and `zdisk/zrepl/remote`. And we can't detect this situation during config check. So let's just remove this check, because it's admin's duty to configure correct root_fs's. --------- Co-authored-by: Christian Schwarz <me@cschwarz.com>	2023-11-01 00:12:54 +01:00
Christian Schwarz	30faaec26a	build: ci: fix `quickcheck-docs` for external PRs (#763 ) fixes https://github.com/zrepl/zrepl/issues/762	2023-11-01 00:12:23 +01:00
Christian Schwarz	21e0ae63a6	build: fix rpm builds, broken by `ef9a63b`: support package revisions	2023-10-07 18:46:28 +00:00
Christian Schwarz	370f40881d	build: wrap-and-checksum didn't include .deb files fixup of `9d5c892023`	2023-10-07 17:03:27 +00:00
Christian Schwarz	fb71a7e4b0	build: forward ZREPL_VERSION and ZREPL_PACKAGE_RELEASE to docker targets	2023-10-07 16:36:43 +00:00
Christian Schwarz	ef9a63b075	build: support package revisions	2023-10-07 16:36:43 +00:00
Christian Schwarz	faef059edf	build: get rid of bins-all target special case, bring back test vet lint steps of release target	2023-10-07 16:36:43 +00:00
Christian Schwarz	ad9fbf7b6d	build: generic _impl target to run a make target for all GOOS/GOARCH combinations	2023-10-07 16:26:53 +00:00
Christian Schwarz	3bd17b8069	build: remove GO_SUPPORTS_ILLUMOS cruft illumos is supported by all Go versions that can build zrepl	2023-10-07 16:26:53 +00:00
Christian Schwarz	99bf1487ae	build: make `release` only build the binaries	2023-10-07 16:26:53 +00:00
Christian Schwarz	c3b4f01c44	build: CGO_ENABLED=0 for all builds	2023-10-07 16:26:50 +00:00
Christian Schwarz	9d5c892023	build: tooling to use CircleCI artifacts for releasing Also, include RPMs and DEBs in the sha256sum.txt	2023-10-01 15:33:03 +00:00
Christian Schwarz	d8d1d25ec2	docs: changelog for 0.6.1	2023-09-10 11:13:14 +00:00
Christian Schwarz	d02d7e5e1d	address updated golangci-lint errors: S1011 (gosimple) should replace loop with `copy.outs[level] = append(copy.outs[level], os.outs[level]...)` (gosimple)	2023-09-10 11:12:46 +00:00
Christian Schwarz	39f8ff62f0	address updated golangci-lint errors: ineffectual assignment to err (ineffassign)	2023-09-10 11:12:46 +00:00
Christian Schwarz	9a434b0e54	go1.21: update golangci-lint (current version panics on go 1.21) Command used: ``` cd build go get -u github.com/golangci/golangci-lint/cmd/golangci-lint go mod tidy ``` Further, golangci-lint requires go 1.20 to build, so, use that as the lowest version in the CI.	2023-09-10 11:12:46 +00:00
Christian Schwarz	b5053d2659	build: use Go 1.21	2023-09-10 10:19:23 +00:00
Christian Schwarz	0fe2ac6b90	debian packaging: make it work on non-x86_64 hosts (arm64 builder, specifically)	2023-09-10 10:12:20 +00:00
Christian Schwarz	95c924968a	circleci: migrate to scheduled pipelines https://circleci.com/docs/migrate-scheduled-workflows-to-scheduled-pipelines/	2023-09-09 11:55:04 +00:00
Christian Schwarz	523a3bb26b	build: address breakage by golang:1.19 Docker image switching to bookworm Debian bookworm apparently _requires_ pip to be used in venv, at least when we use it inside the build.Dockerfile. So, do that.	2023-09-09 11:55:04 +00:00
Christian Schwarz	96396b2e86	circleci: fixup `bc92660`: docs/publish.sh script -P option didn't work forgot to add it to getopt	2023-09-09 11:08:03 +00:00
Sven Kirmess	8749f0bd3d	docs: talks: add note on keep_bookmarks option (#687 ) Mention that the keep_bookmarks option was removed since the talks were given.	2023-09-09 10:47:28 +00:00
Christian Schwarz	bc92660e09	circleci: ensure docs/publish.sh works as part of pre-merge `ci` workflow (#736 )	2023-09-09 12:41:06 +02:00
Christian Schwarz	8b0637ddcc	docs: switch to sphinx-multiversion for multi-versioned docs (#734 ) The sphinxcontrib-versioning seems unmaintainted and I can't get the fork that we used before this PR working on Python 3.10. The situation wrt maintenance doesn't seem much better for sphinx-multiversion, but, at least I could get it to work with current sphinx versions. The main problem with sphinx-multiversion is that it doesn't render anything at `/`. I.e., `https://zrepl.github.io/configuration.html` will be 404. That's different from `sphinxcontrib-versioning`, and thus switching to sphinx-multiversion would break URLs. We host on GitHub pages and don't control the webserver, so, we can't use webserver-level redirects to keep the URLs working. We could create JS-level redirects, or `http-equiv`, but that's ugly as well. The simplest solution was to fork sphinx-multiversion and hard-code zrepl's specific needs into that fork. The fork is based off v0.2.4 and pinned via requirements.txt. Here are its unique commits: https://github.com/Holzhaus/sphinx-multiversion/compare/master...zrepl:sphinx-multiversion:zrepl We should revisit `sphinx-polyversion` in the future once its docs improve. See https://github.com/Holzhaus/sphinx-multiversion/issues/88#issuecomment-1606221194 This PR updates the various Python packages, as I couldn't get sphinx-multiversion to work with the (very old) versions that were pinned in `requirements.txt` prior to this PR. This PR's `requirements.txt` is from a clean Python 3.10 venv on Ubuntu 22.10 after running ``` pip install sphinx sphinx-rtd-theme pip install 'git+https://github.com/zrepl/sphinx-multiversion/@52c915d7ad898d9641ec48c8bbccb7d4f079db93#egg=sphinx_multiversion' ```	2023-09-09 12:21:25 +02:00
Christian Schwarz	bbdc6f5465	fix handling of tenative cursor presence if protection strategy doesn't use it (#714 ) Before this PR, we would panic in the `check` phase of `endpoint.Send()`'s `TryBatchDestroy` call in the following cases: the current protection strategy does NOT produce a tentative replication cursor AND * `FromVersion` is a tentative cursor bookmark * `FromVersion` is a snapshot, and there exists a tentative cursor bookmark for that snapshot * `FromVersion` is a bookmark != tentative cursor bookmark, but there exists a tentative cursor bookmark for the same snapshot as the `FromVersion` bookmark In those cases, the `check` concluded that we would delete `FromVersion`. It came to that conclusion because the tentative cursor isn't part of `obsoleteAbs` if the protection strategy doesn't produce a tentative replication cursor. The scenarios above can happen if the user changes the protection strategy from "with tentative cursor" to one "without tentative replication cursor", while there is a tentative replication cursor on disk. The workaround was to rename the tentative cursor. In all cases above, `TryBatchDestroy` would have destroyed the tentative cursor. In case 1, that would fail the `Send` step and potentially break replication if the cursor is the last common bookmark. The `check` conclusion was correct. In cases 2 and 3, deleting the tentative cursor would have been fine because `FromVersion` was a different entity than the tentative cursor. So, destroying the tentative cursor would be the right call. The solution in this PR is as follows: * add the `FromVersion` to the `liveAbs` set of live abstractions * rewrite the `check` closure to use the full dataset path (`fullpath`) to identify the concrete ZFS object instead of the `zfs.FilesystemVersionEqualIdentity`, which is only identified by matching GUID. * Holds have no dataset path and are not the `FromVersion` in any case, so disregard them. fixes #666	2023-07-04 20:21:48 +02:00
Goran Mekic	bc5e1ede04	metric to detect filesystems rules that don't match any local dataset (#653 ) This PR adds a Prometheus counter called `zrepl_zfs_list_unmatched_user_specified_dataset_count`. Monitor for increases of the counter to detect filesystem filter rules that have no effect because they don't match any local filesystem. An example use case for this is the following story: 1. Someone sets up zrepl with `filesystems` filter for `zroot/pg14<`. 2. During the upgrade to Postgres 15, they rename the dataset to `zroot/pg15`, but forget to update the zrepl `filesystems` filter. 3. zrepl will not snapshot / replicate the `zroot/pg15<` datasets. Since `filesystems` rules are always evaluated on the side that has the datasets, we can smuggle this functionality into the `zfs` module's `ZFSList` function that is used by all jobs with a `filesystems` filter. Dashboard changes: - histogram with increase in $__interval, one row per job - table with increase in $__range - explainer text box, so, people know what the previous two are about We had to re-arrange some panels, hence the Git diff isn't great. closes https://github.com/zrepl/zrepl/pull/653 Co-authored-by: Christian Schwarz <me@cschwarz.com> Co-authored-by: Goran Mekić <meka@tilda.center>	2023-05-02 22:13:52 +02:00
Tercio Filho	2b3daaf9f1	zrepl status: hide progress bar once all filesystems reach terminal state (#674 ) * Added `IsTerminal` method * Made rendering of progress bar conditional based on IsTerminal	2023-05-02 19:28:56 +02:00
Sebastian Jäger	2b3df7e342	docs: address setup with two or more external disks (#695 )	2023-05-02 18:57:26 +02:00
Christian Schwarz	5e4d4188f4	circleci: use orb circlci/go for module caching	2023-02-26 13:08:05 +01:00
Christian Schwarz	1e8ffe4486	circleci: run platform tests in CircleCI	2023-02-26 13:08:05 +01:00
Christian Schwarz	59389b84a2	platformtest: fix logmockzfs wrapper script / make test-platform for Go 1.19 See the comment in the script. refs https://github.com/golang/go/issues/53962 used by make test-platform breaks the test on Go 1.19	2023-02-26 13:08:05 +01:00
Christian Schwarz	4fae0bb68e	grafana: update dashboard to Grafana 9.3.6 ... by importing the old version of the dashboard JSON into Grafana 9.3.6, then re-exporting it.	2023-02-26 11:28:57 +01:00
Guillermo Ramos	9777a441e9	dist: add openrc service file closes https://github.com/zrepl/zrepl/pull/664	2023-01-27 23:59:45 +01:00
InsanePrawn	1a72edea5d	docs/jobs: add replication- conflict_resolution-options to active job types	2023-01-26 00:09:28 +01:00
Christian Schwarz	96db636582	build: circleci: don't trigger periodic full pipeline build for problame/circleci-build	2023-01-08 12:35:59 +01:00
Christian Schwarz	190ab7c08d	build: circleci: stop using minio for artifact storage CircleCI artifacts are available publicly. And regarding expiration of artifacts, it doesn't really matter because I delete minio artifacts after 30d as well.	2022-12-30 14:24:23 +01:00
Christian Schwarz	6be133f55d	remove unused JobDebugSettings along with docs For this kind of debugging, we switched to env vars a while ago. For example, ZREPL_RPC_DEBUG. I don't think we have a substitute for the RPCLog stuff. However, NetConnLogger is still in the codebase. obsoletes https://github.com/zrepl/zrepl/pull/661	2022-12-22 18:13:45 +01:00
Christian Schwarz	5ffd470596	docs: update comment on overriding mountpoint properties during zfs recv of ZVOLs fixes https://github.com/zrepl/zrepl/issues/430	2022-12-10 12:53:24 +01:00
Christian Schwarz	2119dc40ab	docs: update supporters list	2022-12-10 12:00:57 +01:00
Christian Schwarz	0df1c4cdcc	docs: changelog: move donation banner to 0.6 release	2022-11-01 09:57:24 +01:00
Christian Schwarz	2658695a35	build: bump minimum Go version to 1.18, as a dependency in ./tools requires it https://app.circleci.com/pipelines/github/zrepl/zrepl/6085/workflows/bf5b11f2-8dc4-40a2-bb7a-fcf3cf8205d4/jobs/42340 ... build github.com/golangci/golangci-lint/cmd/golangci-lint: cannot load io/fs: cannot find module providing package io/fs go install github.com/wadey/gocovmerge go: downloading github.com/wadey/gocovmerge v0.0.0-20160331181800-b5bfa59ec0ad go: extracting github.com/wadey/gocovmerge v0.0.0-20160331181800-b5bfa59ec0ad go install golang.org/x/tools/cmd/goimports # golang.org/x/mod/module ../../go/pkg/mod/golang.org/x/mod@v0.6.0/module/module.go:147:5: undefined: errors.As note: module requires Go 1.17 go install golang.org/x/tools/cmd/stringer # golang.org/x/tools/go/internal/gcimporter ../../go/pkg/mod/golang.org/x/tools@v0.2.0/go/internal/gcimporter/iimport.go:520:9: undefined: constant.Make ../../go/pkg/mod/golang.org/x/tools@v0.2.0/go/internal/gcimporter/iimport.go:616:9: undefined: constant.Make note: module requires Go 1.18 go install google.golang.org/grpc/cmd/protoc-gen-go-grpc go: downloading google.golang.org/grpc v1.46.2 go: extracting google.golang.org/grpc v1.46.2 go: downloading google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.1.0 go: extracting google.golang.org/grpc/cmd/protoc-gen-go-grpc v1.1.0 go install google.golang.org/protobuf/cmd/protoc-gen-go Exited with code exit status 123	2022-10-31 20:13:36 +01:00
Christian Schwarz	1ac1635b3d	build: circleci: update CA certs in go 1.12 image	2022-10-31 20:13:26 +01:00
Christian Schwarz	4a2806f6d1	build: fix deb-docker performance on newer Docker See comment in Makefile	2022-10-27 00:47:12 +02:00
Christian Schwarz	0a264b9b41	docs: add announcement for next release	2022-10-27 00:19:06 +02:00
Christian Schwarz	a3379d6785	docs: finalize 0.6 changelog	2022-10-27 00:19:06 +02:00
Christian Schwarz	6260b75031	snapper: fix delayed snapshots caused by system suspend/resume See explainer comment in periodic.go for details. fixes https://github.com/zrepl/zrepl/issues/611	2022-10-27 00:19:06 +02:00
Christian Schwarz	3ffb69bfb0	config: support zrepl's day and week units for snapshotting.interval Originally, I had a patch that would replace all usages of time.Duration in package config with the new config.Duration types, but: 1. these are all timeouts/retry intervals that have default values. Most users don't touch them, and if they do, they don't need day or week units. 2. go-yaml's error reporting for yaml.Unmarshaler is inferior to built-in types (line numbers are missing, so the error would not have sufficient context) fixes https://github.com/zrepl/zrepl/issues/486	2022-10-27 00:19:06 +02:00
Yannick Dylla	1da8f848f2	snapper: support custom timestamp format fixes https://github.com/zrepl/zrepl/issues/465 closes https://github.com/zrepl/zrepl/pull/639	2022-10-27 00:19:06 +02:00
Christian Schwarz	6ed4626df9	grafana dashboard: remove zrepl version number from title fixes https://github.com/zrepl/zrepl/issues/624	2022-10-27 00:19:06 +02:00
Christian Schwarz	c07f9ec62e	build: use go 1.19 for testing & release builds New docker image since the old one was deprecated, according to https://discuss.circleci.com/t/go-lang-docker-image-circleci-golang-1-19-is-missing/44961	2022-10-27 00:19:06 +02:00
Christian Schwarz	fd5b0e6831	build: update golangci-lint The previous commits were done in response to updating to the version that we now pin in this commit. We do the update after the fixes so that each commit builds.	2022-10-27 00:19:06 +02:00
Christian Schwarz	a4cea1b4f3	go1.19: zfs.SendStream.Close() after EOF would return context cancellation error Before upgrading to Go 1.19, these platform tests would sproadically fail due to the reason outlined in the comment github.com/zrepl/zrepl/platformtest/tests.SendStreamMultipleCloseAfterEOF github.com/zrepl/zrepl/platformtest/tests.SendStreamCloseAfterEOFRead	2022-10-27 00:19:06 +02:00
Christian Schwarz	c0b52b92d5	systemd: set GOTRACEBACK=crash so that we have core dumps They are useful, not least to debug issues with debugging SIGSYS caused by overly restrictive settings in the unit file. (See previous commit for an example.)	2022-10-26 22:39:18 +02:00
Christian Schwarz	12018b3685	go1.19: adjust systemd unit to allow setrlimit Go 1.19 uses it during startup. From the Go changelog: > On Unix operating systems, Go programs that import package os now > automatically increase the open file limit (RLIMIT_NOFILE) to the > maximum allowed value; that is, they change the soft limit to match the > hard limit. This corrects artificially low limits set on some systems > for compatibility with very old C programs using the select system call. > Go programs are not helped by that limit, and instead even simple > programs like gofmt often ran out of file descriptors on such systems > when processing many files in parallel. One impact of this change is > that Go programs that in turn execute very old C programs in child > processes may run those programs with too high a limit. This can be > corrected by setting the hard limit before invoking the Go program.	2022-10-26 22:39:18 +02:00
Christian Schwarz	a91fb873e4	fix incorrect use of sort.StringSlice A newer version of staticheck found these: > SA4029: sort.StringSlice is a type, not a function, and > sort.StringSlice(variants) doesn't sort your values; consider using > sort.Strings instead (staticcheck)	2022-10-24 22:22:41 +02:00
Christian Schwarz	a6aa610165	run go1.19 gofmt and make adjustments as needed (Go 1.19 expanded doc comment syntax)	2022-10-24 22:22:41 +02:00
Christian Schwarz	6c87bdb9fb	go1.19: switch to new nolint directive that is compatible with Go 1.19 gofmt	2022-10-24 22:22:11 +02:00
Christian Schwarz	b9250a41a2	go1.18: address net.Error.Temporary() deprecation Go 1.18 deprecated net.Error.Temporary(). This commit cleans up places where we use it incorrectly. Also, the rpc layer defines some errors that implement interface { Temporary() bool } I added comments to all of the implementations to indicate whether they will be required if net.Error.Temporary is ever ever removed in the future. For HandshakeError, the Temporary() return value is actually important. I moved & rewrote a (previously misplaced) comment there. The ReadStreamError changes were 1. necessary to pacify newer staticcheck and 2. technically, an error can implement Temporary() without being net.Err. This applies to some syscall errors in the standard library. Reading list for those interested: - https://github.com/golang/go/issues/45729 - https://groups.google.com/g/golang-nuts/c/-JcZzOkyqYI - https://man7.org/linux/man-pages/man2/accept.2.html Note: This change was prompted by staticheck: > SA1019: neterr.Temporary has been deprecated since Go 1.18 because it > shouldn't be used: Temporary errors are not well-defined. Most > "temporary" errors are timeouts, and the few exceptions are surprising. > Do not use this method. (staticcheck)	2022-10-24 22:21:52 +02:00