v0.8.0

Release v0.8.0

Highlights

• #721
• #672
• #768
• #777
• #778
• #726

Deprecations

• #721 - Drop kube-rbac-proxy from deployment
• #752 - Remove namespace option from component config

Changelog

• Doc updates for 0.7 [main branch] by @danielvegamyhre in #692
• Bump the kubernetes group with 7 updates by @dependabot in #693
• Image support multi-arch by @phuhung273 in #694
• Use meta api native condition status by @tenzen-y in #695
• Introduce Go std slices and maps lib by @tenzen-y in #696
• Remove duplicated condition judgement by @tenzen-y in #697
• Bump github.com/onsi/gomega from 1.34.2 to 1.35.1 by @dependabot in #700
• Bump github.com/onsi/ginkgo/v2 from 2.20.2 to 2.21.0 by @dependabot in #699
• Add Coordinator concept by @avrittrohwer in #702
• Bump github.com/open-policy-agent/cert-controller from 0.11.0 to 0.12.0 by @dependabot in #704
• Propagate schedulingGates set on PodTemplate when resuming JobSet by @mimowo in #705
• Update docs for release v0.7.1 by @ahg-g in #712
• Bump sigs.k8s.io/structured-merge-diff/v4 from 4.4.1 to 4.4.3 in the kubernetes group across 1 directory by @dependabot in #710
• Bump the kubernetes group with 7 updates by @dependabot in #714
• Bump github.com/onsi/ginkgo/v2 from 2.21.0 to 2.22.0 by @dependabot in #716
• fix wget failure to resolve relative paths by @kannon92 in #718
• Bump github.com/stretchr/testify from 1.9.0 to 1.10.0 by @dependabot in #715
• Bump github.com/onsi/gomega from 1.35.1 to 1.36.0 by @dependabot in #717
• Allow for one to install jobset in a different namespace by @kannon92 in #719
• Bump sigs.k8s.io/controller-runtime from 0.19.2 to 0.19.3 in the kubernetes group by @dependabot in #725
• Remove kube-rbac-proxy by @kannon92 in #722
• use component config for default installation by @kannon92 in #724
• Update documents to point to v0.7.2 by @ahg-g in #732
• fix docker warnings by @kannon92 in #727
• upgrade python dependencies for sdk by @kannon92 in #728
• KEP-672: Serial Job Execution with DependsOn API by @andreyvelich in #680
• Bump the kubernetes group with 7 updates by @dependabot in #733
• Bump github.com/onsi/gomega from 1.36.0 to 1.36.1 by @dependabot in #734
• update golang net to 0.33 to satisfy security alerts by @kannon92 in #735
• Bump github.com/onsi/ginkgo/v2 from 2.22.0 to 2.22.1 by @dependabot in #739
• Bump github.com/onsi/gomega from 1.36.1 to 1.36.2 by @dependabot in #741
• Bump github.com/onsi/ginkgo/v2 from 2.22.1 to 2.22.2 by @dependabot in #742
• Bump braces from 3.0.2 to 3.0.3 in /site by @kannon92 in #744
• update k8s to 0.32 apis by @kannon92 in #738
• Propagate job labels and annotations by @imreddy13 in #737
• disable http/2 for metrics server by @kannon92 in #745
• Minimize the number of unnecessary logs that get emitted by @imreddy13 in #746
• inject namespace in case we want to test against non standard deployment by @kannon92 in #749
• fix security warnings in client go code by @kannon92 in #743
• Bump sigs.k8s.io/controller-runtime from 0.19.3 to 0.19.4 in the kubernetes group by @dependabot in #750
• Remove Namespace from the JobSet Config by @andreyvelich in #752
• Turn off internal cert management via config by @ardaguclu in #755
• update gen-sdk.sh to generate sdk using docker container by @epicseven-cup in #681
• Use config metrics binding address if flag is not set by @ardaguclu in #756
• Bump the kubernetes group with 7 updates by @dependabot in #759
• Self nominate Kevin Hannon for approval rights by @kannon92 in #758
• copy all files in project rather than piece by piece by @kannon92 in #765
• add go mod download to Dockerfile by @kannon92 in #769
• update python sdk files with latest changes by @kannon92 in #770
• Bump sigs.k8s.io/controller-runtime from 0.20.0 to 0.20.1 in the kubernetes group by @dependabot in #772
• KEP-672: Implement the DependsOn API by @andreyvelich in #740
• Set user agent for requests coming from the jobset controller to "jobset" by @imreddy13 in #775
• Updating the documentation for Pod DNS and underlying headless service. by @raushan2016 in #779
• Remove the Configuration API as a CRD by @ahg-g in #781
• Set image tag and commit version at build time by @ahg-g in #780
• Increase memory limit and remove cpu limit for the default deployment by @priyanshikhetwani in #783
• Add missing external types to apply configurations by @astefanutti in #782
• Bump the kubernetes group with 7 updates by @dependabot in #784
• feature: add Helm chart for jobset by @ChenYi015 in #785
• helm: disable Promethues metrics exporting by default by @ChenYi015 in #789
• Bump github.com/google/go-cmp from 0.6.0 to 0.7.0 by @dependabot in #794
• add make file targets for helm by @kannon92 in #792
• Bump github.com/prometheus/client_golang from 1.20.5 to 1.21.0 by @dependabot in #795
• Fix helm chart push but disable it from cloud build for testing by @kannon92 in #798
• enable helm chart push for cloudbuild for postsubmit by @kannon92 in #799
• add gotoolchain and update cloudbuild name by @kannon92 in #800

New Contributors

• @phuhung273 made their first contribution in #694
• @avrittrohwer made their first contribution in #702
• @andreyvelich made their first contribution in #680
• @imreddy13 made their first contribution in #737
• @ardaguclu made their first contribution in #755
• @epicseven-cup made their first contribution in #681
• @raushan2016 made their first contribution in #779
• @priyanshikhetwani made their first contribution in #783
• @astefanutti made their first contribution in #782
• @ChenYi015 made their first contribution in #785

Full Changelog: v0.7.0...v0.8.0

v0.7.3

What's Changed

• [Release 0.7]: cherry-pick allow for one to install jobset in a different namespace by @kannon92 in #751
• [release-0.7]: cherry-pick turn off internal cert management via config by @kannon92 in #757

Full Changelog: v0.7.2...v0.7.3

Release v0.7.2

What's Changed

• Update docs for v0.7.0 (release branch) by @danielvegamyhre in #691
• Automated cherry pick of #705: Propagate schedulingGates set on PodTemplate when resuming by @mimowo in #706

Full Changelog: v0.7.0...v0.7.2

Release v0.7.1

What's Changed

• Update docs for v0.7.0 (release branch) by @danielvegamyhre in #691
• Automated cherry pick of #705: Propagate schedulingGates set on PodTemplate when resuming by @mimowo in #706

Full Changelog: v0.7.0...v0.7.1

v0.7.0

Highlights

• Add restart strategy by @nstogner in #686
• Priority-based exclusive placement by @ahg-g in #687
• feat: add component config by @rainfd in #609

What's Changed

• fix: delete active jobs right away when job finishes even when TTLSecondsAfterFinished is set by @CecileRobertMichon in #667
• Bump github.com/onsi/ginkgo/v2 from 2.20.0 to 2.20.1 by @dependabot in #663
• Bump github.com/prometheus/client_golang from 1.20.0 to 1.20.2 by @dependabot in #664
• Bump kubernetes dependencies to v0.31.x. by @mbobrovskyi in #670
• Bump github.com/onsi/ginkgo/v2 from 2.20.1 to 2.20.2 by @dependabot in #668
• Bump github.com/onsi/gomega from 1.34.1 to 1.34.2 by @dependabot in #669
• chore: update README.md e2e test version for v1.31.0 by @googs1025 in #671
• Add test-python-sdk on Makefile test. by @mbobrovskyi in #673
• Bump github.com/prometheus/client_golang from 1.20.2 to 1.20.3 by @dependabot in #674
• feat: add component config by @rainfd in #609
• Bump the kubernetes group with 6 updates by @dependabot in #675
• Add global-job-replicas label/annotation by @GiuseppeTT in #677
• Add examples for three existing failure policy actions. by @jedwins1998 in #601
• Bump github.com/prometheus/client_golang from 1.20.3 to 1.20.4 by @dependabot in #679
• chore: use symbolic link instead of directory by @googs1025 in #630
• Priority-based exclusive placement by @ahg-g in #687
• Bump github.com/prometheus/client_golang from 1.20.4 to 1.20.5 by @dependabot in #688
• Add restart strategy by @nstogner in #686

New Contributors

• @CecileRobertMichon made their first contribution in #667
• @rainfd made their first contribution in #609
• @GiuseppeTT made their first contribution in #677
• @nstogner made their first contribution in #686

Full Changelog: v0.7.0-devel...v0.7.0

v0.6.0

Highlights

• New JobSet Failure Policy API - allows users to configure different behavior for different types of errors, enabling them to use compute resources more efficiently and improve ML training goodput.
• Add Coordinator field to JobSet spec, enabling user to define a global coordinator pod for distributed ML/HPC workloads. The stable network endpoint for this pod will be added as a label and annotation to every Job and Pod in the JobSet for easy use in application code. A common use case for this is TPU Multislice training with multiple different Job templates. See linked issue for details.
• Add global Job index label/annotation to every Job and Pod, which is needed to support TPU Multislice training with multiple different Job templates. See linked issue for details.
• Added new metrics
• Improved test coverage
• Bug fixes
• New examples and documentation

What's Changed

• feat: add e2e test for ttl seconds after finished in jobset by @dejanzele in #511
• add publish not ready headless service to jobset by @kannon92 in #505
• use kube-openapi rather than code generator openapi-gen by @kannon92 in #522
• Allow passing args to ginkgo for integration tests by @danielvegamyhre in #525
• Refactor create jobs by @danielvegamyhre in #516
• Do not default the managedBy field by @mimowo in #528
• feat: add event recorder event by @googs1025 in #507
• use t.Errorf instead of t.Fatalf by @googs1025 in #532
• Fix path for the error when attempting to mutate managedBy by @mimowo in #527
• Fix bug when checking if a JobSet is active during tests. by @jedwins1998 in #531
• Correct typo in configurable failure policy KEP. by @jedwins1998 in #539
• fix: fix ci error caused by typo by @googs1025 in #544
• Bump the kubernetes group with 4 updates by @dependabot in #542
• Bump github.com/onsi/gomega from 1.32.0 to 1.33.0 by @dependabot in #543
• docs: fix site url not found by @googs1025 in #541
• use hugo param to define variables in md language by @googs1025 in #540
• add unit tests for createHeadlessSvcIfNecessary by @dejanzele in #526
• test: add pod controller unit test by @googs1025 in #490
• Add comment explaining why we don't unconditionally compute firstFailedJob by @danielvegamyhre in #549
• Bump github.com/onsi/ginkgo/v2 from 2.17.1 to 2.17.2 by @dependabot in #552
• Track which features in roadmap have been released by @danielvegamyhre in #554
• docs: using kustomize for adjusting resources by @omerap12 in #558
• Bump github.com/onsi/gomega from 1.33.0 to 1.33.1 by @dependabot in #560
• Don't reconcile JobSets with deletion timestamp set by @danielvegamyhre in #562
• Improve the API generated docs for managedBy by @mimowo in #565
• chore: Upgrade e2e local image by @googs1025 in #567
• Bump github.com/onsi/ginkgo/v2 from 2.17.2 to 2.17.3 by @dependabot in #569
• Add support for feature gates by @googs1025 in #557
• Implement configurable failure policy. by @jedwins1998 in #537
• Update the JobSet version to 0.5.1 for installation by @mimowo in #577
• Bump github.com/onsi/ginkgo/v2 from 2.17.3 to 2.19.0 by @dependabot in #581
• Relax validation on ReplicatedJob PodTemplates of suspended JobSets by @danielvegamyhre in #580
• update makefile kind version to v1.30.0 by @googs1025 in #589
• Propagate Job pod template updates to suspended jobs when resuming by @danielvegamyhre in #590
• docs: update to v0.5.2 by @googs1025 in #593
• fix: fix log to avoid panic by @googs1025 in #595
• avoid log panic by @googs1025 in #598
• Add omitempty to annotation of OnJobFailureReasons. by @jedwins1998 in #596
• update readme docs e2e test version to v1.30 by @googs1025 in #602
• Update _index.md MASTER_ADDR by @song-william in #604
• Add client-go example by @danielvegamyhre in #606
• Wait for the webhook service to be listening before advertising the Jobset replica as ready. by @mbobrovskyi in #608
• docs: add simple example for network field by @googs1025 in #550
• feat: add terminalState to jobset status by @googs1025 in #594
• Integration test improvement: rename "update" to "step" by @danielvegamyhre in #610
• docs: add argo workflow example for jobset by @googs1025 in #612
• docs: add JobSet API reference by @googs1025 in #611
• docs: fix typo, Github -> GitHub by @highpon in #615
• Allow mutating schedulingGates when the Jobset is suspended by @mimowo in #623
• Add Coordinator field to JobSet spec by @danielvegamyhre in #618
• Validation for Coordinator field by @danielvegamyhre in #627
• Add example for coordinator by @danielvegamyhre in #628
• docs: add prometheus-operator example for jobset by @googs1025 in #629
• Bump github.com/onsi/gomega from 1.33.1 to 1.34.0 by @dependabot in #631
• Bump github.com/onsi/ginkgo/v2 from 2.19.0 to 2.19.1 by @dependabot in #632
• feat: add metrics for jobset by @googs1025 in #614
• docs: update metrics info for site by @googs1025 in #633
• chore: add github issue, pr template by @googs1025 in #634
• Bump github.com/onsi/gomega from 1.34.0 to 1.34.1 by @dependabot in #638
• fix error output by @googs1025 in #636
• Bump k8s dependencies to 1.30 dependencies and modify update-codegen.sh to be compatible with new code-generator by @danielvegamyhre in #641
• Fix bug in replicatedJobByName by @danielvegamyhre in #645
• Allow to update JobSets on suspend by @mimowo in #644
• Refactor jobset webhook by @danielvegamyhre in #646
• add the unparam linter to golangci and fix those issues flagged by @kannon92 in #643
• drop job-name from labels as it is not used by @kannon92 in #642
• Bump github.com/onsi/ginkgo/v2 from 2.19.1 to 2.20.0 by @dependabot in #647
• Add new job-id annotation to assign globally unique job index to each job by @danielvegamyhre in #650
• Bump github.com/prometheus/client_golang from 1.19.1 to 1.20.0 by @dependabot in #653
• update to k8s 0.30.4 by @kannon92 in #654

New Contributors

• @mimowo made their first contribution in #528
• @omerap12 made their first contribution in #558
• @song-william made their first contribution in #604
• @mbobrovskyi made their first contribution in #608
• @highpon made their first contribution in #615

Full Changelog: v0.6.0-devel...v0.6.0

JobSet v0.5.2

What's Changed

• Automated cherry pick of #580: relax validation on replicated jobs by @danielvegamyhre in #584
• Automated cherry pick of #590: propagate job pod template updates to suspended jobs when by @danielvegamyhre in #591

Full Changelog: v0.5.1...v0.5.2

v0.5.1

Highlights

• Fixed bug causing foreground cascading deletion policy to not work properly on JobSets #562
• Fixed field path in error message in validation for ManagedBy field #527
• Test coverage improvements, refactoring, additional documentation

What's Changed

• Update docs for 0.5.0 by @danielvegamyhre in #517
• [Release-0.5] Do not default the managedBy field by @kannon92 in #533
• Automated cherry pick of #527: Fix path for the error when mutating managedBy by @kannon92 in #534
• Automated cherry pick of #562: don't reconcile jobsets with deletion timestamp set by @danielvegamyhre in #564

Full Changelog: v0.6.0-devel...v0.5.1

v0.5.0

What's Changed

Highlights

• JobSet TTL support added in #443
• Docsite is live at https://jobset.sigs.k8s.io/ with updated documentation and examples.
• Include first failed job name in event emitted when JobSet fails, to speed up the debugging process for large complex workloads #477
• Lower default resource request for JobSet controller manager so it fits on default cloud CPU VMs, but keep high limit to support maximum performance #480
• Perform only 1 JobSet status update per reconcile attempt to reduce pressure on k8s apiserver #494
• Introduced MangedBy field to the JobSet spec to enable Multi-Kueue support

Detailed release notes

• Add info to landing page by @danielvegamyhre in #435
• Validate follower pod owned by same Job as leader pod by @danielvegamyhre in #433
• Bump github.com/stretchr/testify from 1.8.4 to 1.9.0 by @dependabot in #439
• Add descriptions to ReplicatedJobStatus fields by @danielvegamyhre in #442
• Bump github.com/onsi/ginkgo/v2 from 2.15.0 to 2.16.0 by @dependabot in #444
• Add JobSet diagram and other doc updates by @danielvegamyhre in #446
• Update installation version to latest release in public docs by @danielvegamyhre in #450
• add concept image by @moficodes in #454
• Update tasks documentation by @danielvegamyhre in #453
• Emit Job creation failed event by @danielvegamyhre in #448
• Remove Jobset Docs from root by @moficodes in #455
• Fix 404 error when clicking on driver-worker-success-policy.yaml by @kannon92 in #456
• Rename FAQ to troubleshooting on docsite by @danielvegamyhre in #457
• Bump the kubernetes group with 4 updates by @dependabot in #459
• Add features overview to README by @danielvegamyhre in #452
• Update Makefile rules to use more specific paths by @danielvegamyhre in #470
• Fix typo in readme by @danielvegamyhre in #472
• Add jobset roadmap to README by @danielvegamyhre in #468
• Bump github.com/onsi/gomega from 1.31.1 to 1.32.0 by @dependabot in #475
• Bump github.com/onsi/ginkgo/v2 from 2.16.0 to 2.17.1 by @dependabot in #474
• update golang to 1.22 by @kannon92 in #471
• Lower default resource request for controller manager but keep high limit by @danielvegamyhre in #480
• Include first failed job name in event emitted when JobSet fails, as well as the JobSet failure condition by @danielvegamyhre in #477
• Update README.md to correct concepts link by @jtorrex in #486
• Code cleanup and refactoring by @danielvegamyhre in #484
• Move headless service creation outside of createJobs by @danielvegamyhre in #483
• Remove Duplicate Import by @jedwins1998 in #488
• Introduce managedBy field and Remove managed-by label by @jedwins1998 in #487
• fix some typo error by @googs1025 in #489
• Move JobSet webhook into same webhooks package as pod webhook by @danielvegamyhre in #460
• add unit test for jobset webhook updates by @kannon92 in #464
• feat: add support for ttl cleanup for finished jobsets by @dejanzele in #443
• Add unit tests to jobset success policy functions by @zhifei92 in #501
• fix: add IsNotFoundErr when get headlessSvc by @googs1025 in #503
• Update envtest and add back crd generation when updating the api by @kannon92 in #510
• Call Status.Update once in each reconcile attempt by @danielvegamyhre in #494
• Clean up outdated comments by @danielvegamyhre in #512
• Bump sigs.k8s.io/controller-runtime from 0.17.2 to 0.17.3 in the kubernetes group by @dependabot in #513
• Update docs for 0.5.0 by @danielvegamyhre in #517

New Contributors

• @jtorrex made their first contribution in #486
• @jedwins1998 made their first contribution in #488
• @zhifei92 made their first contribution in #501

Full Changelog: v0.5.0-devel...v0.5.0

v0.4.0

What's Changed

• Update main branch installation docs for release v0.3.0 by @danielvegamyhre in #349
• use kind export logs by @kannon92 in #352
• add suspend to replicated job status by @kannon92 in #250
• Update the installation docs to mention the CPU nodes minimum necessary CPU/memory resources by @danielvegamyhre in #354
• Use jobset-system instead of kind-system for jobset by @kannon92 in #358
• A KEP for StartupPolicy by @kannon92 in #244
• Add patches for Kustomize to add objectSelectors to pod webhook configurations by @danielvegamyhre in #362
• Update installation docs for v0.3.1 [main] by @danielvegamyhre in #368
• Bump k8s.io/apimachinery from 0.28.4 to 0.28.5 by @dependabot in #369
• Bump github.com/open-policy-agent/cert-controller from 0.10.0 to 0.10.1 by @dependabot in #373
• Bump k8s.io/api from 0.28.4 to 0.28.5 by @dependabot in #370
• Bump k8s.io/code-generator from 0.28.3 to 0.28.5 by @dependabot in #371
• Bump k8s.io/client-go from 0.28.4 to 0.28.5 by @dependabot in #372
• Bump github.com/onsi/ginkgo/v2 from 2.13.2 to 2.14.0 by @dependabot in #376
• update kind to 0.20.0 by @kannon92 in #359
• Bump k8s.io/code-generator from 0.28.5 to 0.28.6 by @dependabot in #382
• Bump github.com/onsi/gomega from 1.30.0 to 1.31.1 by @dependabot in #383
• Bump k8s.io/client-go from 0.28.5 to 0.28.6 by @dependabot in #384
• upgrade kubernetes apis to 0.29 by @kannon92 in #387
• Move exclusive placement annotation to ReplicatedJob template by @danielvegamyhre in #389
• add dependabot groups for k8s packages by @kannon92 in #391
• add a message to events by @kannon92 in #390
• Migrate from background to foreground cascading deletion policy by @danielvegamyhre in #393
• Default service name in JobSet controller by @danielvegamyhre in #395
• bumping controller tools to see if this fixes ci by @kannon92 in #403
• add suspend field to printcolumn by @kannon92 in #400
• add jobset docsite by @moficodes in #402
• KEP 262: Configurable Failure Policy API by @danielvegamyhre in #381
• Get subdomain via a func instead of defaulting it on the jobset object by @ahg-g in #404
• Bump the kubernetes group with 1 update by @dependabot in #406
• Startup policy implementation by @kannon92 in #246
• Minor cleanup to ensureConditionOpts by @ahg-g in #410
• Validate longest pod name for jobset will not exceed 63 chars by @danielvegamyhre in #409
• Add managed-by label support. by @trasc in #407
• Improve error messages and logging in webhooks by @danielvegamyhre in #421
• Update installation docs for v0.3.2 by @danielvegamyhre in #424
• typo: Fix some comments by @googs1025 in #426
• Bump the kubernetes group with 5 updates by @dependabot in #431
• Update docsite title and subtitle by @danielvegamyhre in #432

New Contributors

• @moficodes made their first contribution in #402
• @trasc made their first contribution in #407
• @googs1025 made their first contribution in #426

Full Changelog: v0.4.0-devel...v0.4.0