Deprecations

• Deprecate apiextensions.k8s.io/v1beta1 CRD (#1330 by @ChenYi015)

Features

• feature(tf-operator): make workers, qps and burst configurable (#1379 by @ChenYi015)

Dependencies

• chore(deps): bump actions/stale from 9 to 10 (#1374 by @dependabot[bot])
• chore(deps): bump actions/setup-go from 5 to 6 (#1375 by @dependabot[bot])
• chore(deps): bump actions/setup-python from 5 to 6 (#1376 by @dependabot[bot])
• chore(deps): bump actions/download-artifact from 5 to 6 (#1383 by @dependabot[bot])
• chore(deps): bump actions/upload-artifact from 4 to 5 (#1382 by @dependabot[bot])
• chore(deps): bump github.com/containerd/containerd from 1.7.27 to 1.7.29 (#1384 by @dependabot[bot])
• chore(deps): bump golang.org/x/crypto from 0.41.0 to 0.45.0 (#1388 by @dependabot[bot])
• chore(deps): bump github.com/spf13/pflag from 1.0.7 to 1.0.10 (#1372 by @dependabot[bot])
• chore(deps): bump google.golang.org/protobuf from 1.36.7 to 1.36.10 (#1381 by @dependabot[bot])

Misc

• Remove vendor directory (#1387 by @ChenYi015)

Full Changelog

Bug Fixes

• Fix: tfjob gets stuck in running state when succeeded pods are garbage collected (#1370 by @ChenYi015)

Dependencies

• Bump helm.sh/helm/v3 from 3.16.3 to 3.18.4 (#1350 by @ChenYi015)
• chore(deps): bump golang.org/x/crypto from 0.39.0 to 0.40.0 (#1351 by @dependabot[bot])
• chore(deps): bump github.com/spf13/pflag from 1.0.6 to 1.0.7 (#1352 by @dependabot[bot])
• chore(deps): bump actions/download-artifact from 4 to 5 (#1356 by @dependabot[bot])
• chore(deps): bump actions/checkout from 4 to 5 (#1359 by @dependabot[bot])
• chore(deps): bump actions/setup-java from 4 to 5 (#1366 by @dependabot[bot])
• chore(deps): bump helm.sh/helm/v3 from 3.18.4 to 3.18.6 (#1364 by @dependabot[bot])
• chore(deps): bump github.com/onsi/ginkgo/v2 from 2.22.0 to 2.25.2 (#1369 by @dependabot[bot])

Full Changelog

⚠️ Announcements

We will deprecate CRDs in the apiextensions.k8s.io/v1beta1 API version and support only CRDs in the apiextensions.k8s.io/v1 API version within two months.
Additionally, we will no longer support Kubernetes clusters running versions older than v1.16.0 during the same period.

✨ Features

• Add support for configuring tolerations (#1337 by @ChenYi015)

🌱 Misc

• Remove kubernetes artifacts (#1329 by @ChenYi015)
• [CI] Add CI workflow for releasing Arena images (#1340 by @ChenYi015)
• Update uninstall bash script (#1335 by @ChenYi015)
• Fix golangci-lint issues (#1341 by @ChenYi015)
• Bump golang version from 1.22.7 to 1.23.10 (#1345 by @ChenYi015)
• chore(deps): bump github.com/prometheus/common from 0.60.1 to 0.65.0 (#1343 by @dependabot[bot])
• chore(deps): bump golang.org/x/crypto from 0.38.0 to 0.39.0 (#1334 by @dependabot[bot])

Full Changelog

⚠️ Announcements

• We will deprecate CRDs in the apiextensions.k8s.io/v1beta1 API version and support only CRDs in the apiextensions.k8s.io/v1 API version within three months.
• Additionally, we will no longer support Kubernetes clusters running versions older than v1.16.0 during the same period.

New Features

• refactor: use helm lib instead of helm binary (#1207 by @ChenYi015)
• feat: add new value for using localtime in cron-operator (#1296 by @ChenYi015)
• Delete all services when the TFJob is terminated (#1316 by @ChenYi015)
• Make number of replicas of cron-operator deployment configurable (#1325 by @ChenYi015)
• Make number of replicas of tf-operator deployment configurable (#1323 by @ChenYi015)
• Add custom device support for kserve and kserving. (#1315 by @Leoyzen)
• Feat: support affinity policy for kserve and tfjob (#1319 by @Syspretor)
• Feat: support separate affinity policy configuration for PS and worke… (#1331 by @Syspretor)

Bug Fixes

• fix: job status displays incorrectly (#1289 by @ChenYi015)
• fix: service account should use release namespace (#1308 by @ChenYi015)

Misc

• Add basic e2e tests (#1225 by @ChenYi015)
• Bump github.com/containerd/containerd from 1.7.23 to 1.7.27 (#1290 by @dependabot[bot])
• Add stale bot to mark stale issues and PRs (#1141 by @ChenYi015)
• Fix typos in multiple files (#1304 by @co63oc)
• Fix typos in multiple files (#1310 by @co63oc)

Full Changelog

Misc

• Fix typos (#1276 by @co63oc)
• Update pytorch operator image (#1281 by @ChenYi015)

Full Changelog

Bug Fixes

• fix: device value does not support k8s resource quantity (#1267 by @ChenYi015)
• fix: pytorchjob does not support backoff limit (#1272 by @ChenYi015)
• unset env NVIDIA_VISIBLE_DEVICES when gpushare is enabled (#1273 by @ChenYi015)

Misc

• docs: fixed typo (#1257 by @DBMxrco)
• Bump github.com/golang/glog from 1.2.3 to 1.2.4 (#1263 by @dependabot[bot])
• fix: format of tensorflow standalone training docs is messed up (#1265 by @ChenYi015)

Full Changelog

Features

• rename parameter (#1262 by @gujingit)

Misc

• Add changelog for v0.13.1 (#1248 by @ChenYi015)
• Bump github.com/go-resty/resty/v2 from 2.16.0 to 2.16.5 (#1254 by @dependabot[bot])

Full Changelog

Misc

• feat: add linux/arm64 support for tf-operator image (#1238 by @ChenYi015)
• feat: add linux/arm64 support for mpi-operator image (#1239 by @ChenYi015)
• feat: add linux/arm64 support for cron-operator image (#1240 by @ChenYi015)
• feat: add linux/arm64 support for et-operator image (#1241 by @ChenYi015)
• Add PyTorch mnist example (#1237 by @ChenYi015)
• Update the version of elastic-job-supervisor in arena-artifacts (#1247 by @AlanFokCo)

Full Changelog

New Features

• feat: add support for torchrun (#1228 by @ChenYi015)
• Update pytorch-operator image (#1234 by @ChenYi015)

Bug Fix

• Avoid listing jobs and statefulsets when get pytorchjob (#1229 by @ChenYi015)

Misc

• Update tfjob standalone training job doc (#1222 by @ChenYi015)
• Remove archived docs (#1208 by @ChenYi015)
• Add changelog for v0.12.1 (#1224 by @ChenYi015)
• Bump golang.org/x/crypto from 0.29.0 to 0.31.0 (#1231 by @dependabot[bot])
• Bump google.golang.org/protobuf from 1.35.1 to 1.36.0 (#1227 by @dependabot[bot])

Full Changelog

New Features

• Support MPI Job with generic devices (#1209 by @cheyang)

Bug Fix

• Update tf-operator image to fix clean pod policy issues (#1200 by @ChenYi015)
• Fix etjob rendering error when using local logging dir (#1203 by @TrafalgarZZZ)
• Fix the functionality of generating kubeconfig (#1204) (#1205 by @wqlparallel)
• Update cron operator image (#1214 by @ChenYi015)

Misc

• Add changelog for v0.12.0 (#1199 by @ChenYi015)
• Add go mod vendor check to integration test (#1198 by @ChenYi015)
• bump github.com/go-resty/resty/v2 from 2.15.3 to 2.16.0 (#1202 by @dependabot[bot])
• Publish releases only on master branch (#1210 by @ChenYi015)
• Add docs for releasing arena (#1201 by @ChenYi015)
• Bump golang.org/x/crypto from 0.28.0 to 0.29.0 (#1206 by @dependabot[bot])
• Release v0.12.1 (#1215 by @ChenYi015)

Full Changelog