AI-Hypercomputer/xpk

Issues

• upgrading a cluster with xpk also upgrades kueue but does so improperly

  #296 opened 13 days ago by logicalhan
  3

• xpk-large-scale-guide should use Cloud DNS instead of core-dns

  #287 opened 17 days ago by avrittrohwer
  0

• update version on pypi?

  #285 opened 19 days ago by GallagherCommaJack
  0

• installation broken after updating main branch

  #284 opened 21 days ago by samos123
  0

• Error on `cluster describe` when creating cluster with `--device-type`

  #282 opened 22 days ago by IrvingMg
  0

• Create integration test using Kind

  #267 opened a month ago by IrvingMg
  0

• Consider configuring kueue waitForPodsReady

  #191 opened 3 months ago by avrittrohwer
  3

• Support setting node auto-provisioning cpu and memory parameters

  #190 opened 3 months ago by avrittrohwer
  0

• Support multiple clusterqueues / localqueues with xpk

  #51 opened a year ago by Obliviour
  1

• Create cluster from several reservations

  #161 opened 5 months ago by DwarKapex
  1

• xpk Cluster Queue resource group "cpu" resource quota incorrect for a CPU-only cluster

  #158 opened 5 months ago by bernardhan33
  11

• Set cluster versions by default to be based on the recommended rapid channel version

  #40 opened 8 months ago by Obliviour
  1

• Clear up xpk backward compatibility story

  #19 opened 9 months ago by Obliviour
  1

• Create Github runner to execute xpk on vm.

  #23 opened 10 months ago by Obliviour
  1

• Make XPK Handle multiple slice sizes

  #17 opened a year ago by rwitten
  3

• Bump JobSet to v0.3.1 after testing integration

  #52 opened a year ago by danielvegamyhre
  1

• Make sure the user doesn't have the region set using `gcloud config set region` without having a zone set or a zone flag.

  #24 opened a year ago by Obliviour
  1

• Add xpk workload label in jobset api

  #43 opened a year ago by Obliviour
  0

• Wrong output value of TPUVMs when the cluster have the string tpu in its name

  #14 opened a year ago by williampispico
  1

• Create github pip release workflow

  #42 opened a year ago by Obliviour
  0

• MaxText/XPK story is kind of intertwined

  #10 opened a year ago by rwitten
  0

• Graceful Error Message on Failing To Apply Kueue

  #12 opened a year ago by rwitten
  0

• Improve CPU Node Pool Defaults

  #30 opened a year ago by Obliviour
  0

• support positional arguments?

  #31 opened a year ago by GallagherCommaJack
  0

• Set reservation to default capacity type and add Reservation-affinity flag to help users set their reservation type.

  #22 opened a year ago by Obliviour
  0

• Improve Jobset error messaging to clearly list solution around Kubernetes Permissions

  #20 opened a year ago by Obliviour
  0

• Make XPK pip3 installable

  #16 opened a year ago by rwitten
  1

• List of Nits From An Early User

  #15 opened a year ago by rwitten
  0

• Marrying User Metadata Into Cluster

  #13 opened a year ago by rwitten
  0

• Error applying Kueue CRDs on MacOS

  #9 opened a year ago by danielvegamyhre
  1