Skip to content

v1.14.3

Choose a tag to compare

View all tags
@JesseStutler JesseStutler released this 27 Jun 03:49
· 342 commits to master since this release
v1.14.3
This tag was signed with the committer’s verified signature.
JesseStutler Zicong Chen(Jesse Stutler)
GPG key ID: A5D248F299A37AF7
Verified
Learn about vigilant mode.
27fec3b
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's Changed

Bug fixes

  • [release-1.14] Fix Network-Topology-Aware scheduling soft mode for jobs and subjobs by @3th4novo in #5513
  • [release-1.14] Fix scalar in-queue resource accounting to use milli-units by @WHOIM1205 in #5503
  • [release-1.14] Clear device-related annotations when releasing pods by @archlitchi in #5486
  • [release-1.14] Fix maxFloat and maxInt display in scheduler resource strings by @r0hansaxena in #5477
  • [release-1.14] Fall back minAvailable to replicas when partition minPartitions is omitted by @togettoyou in #5459
  • [release-1.14] Check Ascend vNPU health using Allocatable resources by @DSFans2014 in #5437
  • [release-1.14] Prevent unbounded growth of job.Status.Conditions by @avinxshKD in #5448
  • [release-1.14] Fix HAMi vGPU scheduling failures in medium and large clusters by @Copilot in #5434
  • [release-1.14] Restore ImageStates in scheduler node snapshots by @kitianFresh in #5356
  • [release-1.14] Reprieve higher-priority pods first during preemption by @dengaosong in #5307

Full Changelog: v1.14.2...v1.14.3

Contributors

kitianFresh, togettoyou, and 7 other contributors
Assets 2
Loading