Skip to content

Releases: nebius/soperator

1.14.14

29 Oct 13:27
0574392
Compare
Choose a tag to compare

Changes made since version 1.14.13 prior to version 1.14.14:

  • no changes
📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
5 29 25

1.14.13

25 Oct 13:52
b6064cc
Compare
Choose a tag to compare

Changes made since version 1.14.12 prior to version 1.14.13:

🐛 Fixes

  • Fix getting uniq node for all slurm partitions in nccl test

📦 Dependencies

  • build(deps): bump actions/setup-go from 5.0.2 to 5.1.0
  • build(deps): bump docker/login-action from 1f36f5b7a2d2f7bfd524795fc966e6d88c37baa9 to 5d8785b43a795ee002a17dbf1a2235dc1997224b

Contributors:
@dependabot[bot], @asteny

📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
3 0 7 25 25

1.14.12

24 Oct 10:12
8c72c1a
Compare
Choose a tag to compare

Changes made since version 1.14.11 prior to version 1.14.12:

📦 Dependencies

  • Bump go.opentelemetry.io/otel/sdk from 1.30.0 to 1.31.0 in /images/jail/gpubench
  • Bump go.opentelemetry.io/otel/exporters/otlp/otlpmetric/otlpmetrichttp from 1.28.0 to 1.31.0 in /images/jail/gpubench
  • build(deps): bump actions/checkout from 4.2.1 to 4.2.2

Other

  • other: bump soperator

Contributors:
@dependabot[bot], @asteny

📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
3 1 7 68 68

1.14.11

22 Oct 19:22
8467124
Compare
Choose a tag to compare

Changes made since version 1.14.10 prior to version 1.14.11:

🚀 Features

  • feature: slurm partition configuration
  • feature: Enable Scrontab

🐛 Fixes

  • CPU-only workers w/o toolkit-validation
  • MSP-3091: fix reconcile condition accounting

Contributors:
@asteny, @dstaroff, @Uburro

📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
4 0 14 232 46

1.14.10

17 Oct 17:34
acef4c4
Compare
Choose a tag to compare

Changes made since version 1.14.9 prior to version 1.14.10:

🐛 Fixes

  • Use LoadBalancer as default service type & do not require specifying IP for it

Contributors:
@dstaroff

📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
1 0 17 27 26

1.14.9

17 Oct 16:22
e61b438
Compare
Choose a tag to compare

Changes made since version 1.14.8 prior to version 1.14.9:

🐛 Fixes

  • FIX: Use empty string as default for sshdServiceLoadBalancerIP

Contributors:
@dstaroff

📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
1 0 2 27 23

1.14.8

17 Oct 16:12
e61b438
Compare
Choose a tag to compare

Changes made since version 1.14.7 prior to version 1.14.8:

🧪 Tests

  • Fix versions and add versions check test

🐛 Fixes

  • fix: remove fromTag
  • Fix versions and add versions check test
  • Accept no IP for LB Service type
  • Fixes for version check and yq installation
  • create release only for new tags

Contributors:
@asteny, @dstaroff

📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
5 0 17 77 40

1.14.7

12 Oct 17:35
e7ecd40
Compare
Choose a tag to compare

🚀 Features

  • feature: create release with changelog

🐛 Fixes

  • MSP-3088: Fix configurable ulimits

📦 Dependencies

  • Bump actions/checkout from 4.1.7 to 4.2.0
  • Bump actions/checkout from 4.1.7 to 4.2.0
  • Bump actions/checkout from 4.2.0 to 4.2.1
  • Bump docker/login-action from 3b8fed7e4b60203b2aa0ecc6c6d6d91d12c06760 to 1f36f5b7a2d2f7bfd524795fc966e6d88c37baa9
  • Bump docker/setup-buildx-action from 3.6.1 to 3.7.1

Other

  • NOTIC: Update layers diagram
  • Fix missing conditions
  • MSP-2791: added support mariadb for accounting
  • NOTASK: add accounting to readme
  • fix: in react to -> in reaction to in architecture.md
  • MSP-2852: add values slurmdbd.conf
  • MSP-2979: add slurm.conf for accounting
  • MSP-2983: add dependency slurmdbd.conf for login, controller and worker controller
  • MSP-2983: fix bug with reconcile state accounting
  • Run dependabot for dev branch
  • [FIX] Cluster name altering
  • HOTFIX: moving accounting steps
  • HOTFIX: moving accounting
  • [MSP-2905] Support clusters without GPUs
  • remove CPU-only future plan from readme
  • Bump k8s.io/client-go, go.opentelemetry.io/otel
  • Add Nebius docker registry
  • Change default docker registry from github to nebius
  • Fix cpu only clusters

Contributors:
@rdjjke, @dstaroff, @Uburro, @dependabot[bot], @dvolk, @asteny

📁 Categorized PRs 📂 Uncategorized PRs 📥 Commits Lines added Lines deleted
7 19 57 22959 711

v1.14.2

24 Sep 14:38
5be8480
Compare
Choose a tag to compare

What's Changed

Images

Docker

Operator

Cluster

Helm

v1.14.1

23 Sep 14:39
bbf690f
Compare
Choose a tag to compare

What's Changed

  • [ENH] Move cgroup v2 creation from worker's init container to entrypoint by @Uburro in #13
  • [ADD] CEL validation for job telemetry and shared memory by @Uburro in #14
  • [ADD] slurmdbd image by @Uburro in #15
  • [ENH] Rename slurmcontroller to common name by @Uburro in #18
  • [ADD] ulimits by @asteny in #19
  • [ENH] CRD check for prometheus in Helm operator by @Uburro in #31
  • [FIX] Exporter refactoring && batch bug fixes by @asteny in #30
  • [ENH] Push Docker images into GitHub by @asteny in #56
  • [ENH] Using lightweight CUDA images by @asteny
  • [ENH] Better support for metrics exporting by @Uburro and @asteny
  • [ADD] Support accounting with external DB by @Uburro

Additional changes

Images

Docker

Operator

Cluster

Helm