All pull requests

Pull requests

• [CELEBORN-2473] Do not relocate zstd-jni in the Tez shaded jar

  kind:buildmodule:clientmodule:tez
  #3865·senthh opened Oct 9, 2026Contributor·

  ·
  2

• [CELEBORN-2472] Unregister Celeborn application from MR AppMaster on shutdown

  module:clientmodule:mr
  #3864·senthh opened Oct 9, 2026Contributor·

  ·

• [CELEBORN-2474] Send WWW-Authenticate challenge when no auth header present

  kind:documentationmodule:commonmodule:service
  #3863·senthh opened Oct 9, 2026Contributor·

  ·

• [MINOR] Update pnpm/action-setup to specific version v6.1.0 in web_lint.yml

  kind:infra
  #3862·cxzl25 opened Oct 9, 2026Contributor·2 tasks·

  ·

• [CELEBORN-XXXX] Fix memory leak of committedFileInfos when graceful shutdown is disabled

  correctnessmodule:worker
  #3860·SoulKyu opened Oct 5, 2026·1 of 2 tasks·

  ·
  3

• [CELEBORN-XXXX] Skip memory file storage for replica partitions

  kind:documentationmodule:commonmodule:worker
  #3859·venkata91 opened Sep 29, 2026·1 of 2 tasks·

  ·

• [CELEBORN-2469] Add worker stats collection and reporting for autoscaling

  kind:documentationmodule:commonmodule:mastermodule:worker
  #3858·Kalvin2077 opened Sep 24, 2026Contributor·1 of 2 tasks·

  ·
  4

• [CELEBORN-2468] Support pluggable transport authentication and authorization

  correctnesskind:documentationmodule:clientmodule:commonmodule:flinkmodule:mastermodule:worker
  #3857·Kalvin2077 opened Sep 23, 2026Contributor·2 tasks done·

  ·
  15

• [CELEBORN-2470] Add backoff with jitter to push retries and log hard split once per file

  kind:documentationmodule:clientmodule:commonmodule:worker
  #3856·venkata91 opened Sep 20, 2026·

  ·
  7

• [CELEBORN-2466] Support per-zone worker StatefulSets and autoscaling in the Helm chart

  kind:deploy
  #3855·strelok89 opened Sep 17, 2026Contributor·1 of 2 tasks·

  ·
  17

• [CELEBORN-2464] Wire Kubernetes health probes into the Helm chart

  kind:deploystale
  #3854·strelok89 opened Sep 17, 2026Contributor·1 of 2 tasks·

  ·
  3

• [CELEBORN-2465][SPARK] Proactively unregister shuffles on SQL query completion

  kind:documentationmodule:clientmodule:commonmodule:sparkmodule:tests
  #3853·yew1eb opened Sep 17, 2026Contributor·

  ·
  9

• [CELEBORN-2359] Deduplicate # HELP/# TYPE lines in PrometheusServlet output

  module:servicestale
  #3852·afterincomparableyum opened Sep 17, 2026Contributor·1 of 2 tasks·

  ·
  7

• [CELEBORN-2357] Add Metrics Prefix

  kind:documentationmodule:common
  #3851·afterincomparableyum opened Sep 17, 2026Contributor·2 tasks·

  ·
  1

• [CELEBORN-2356] Fix NPE before null-check in handleChunkFetchRequest

  correctnessmodule:worker
  #3850·afterincomparableyum opened Sep 17, 2026Contributor·1 of 2 tasks·

  ·
  1

• [CELEBORN-2463] Add ManuallyExcludedWorkerCount metric to distinguish manually excluded workers

  kind:documentationmodule:master
  #3848·yew1eb opened Sep 15, 2026Contributor·1 of 2 tasks·

  ·
  1

• [CELEBORN-2462][WORKER] Reduce misleading WARN/ERROR logs during decommission shutdown

  module:worker
  #3847·yew1eb opened Sep 14, 2026Contributor·2 tasks·

  ·
  2

• [CELEBORN-2460] Reply SHUFFLE_EXPIRED for a released shuffle instead of an empty file group

  correctnessmodule:clientmodule:commonmodule:tests
  #3845·shlomitubul opened Sep 10, 2026Contributor·1 of 2 tasks·

  ·
  11

• [CELEBORN-2459] Fix race causing DataPusher to accept data during termination and drop it

  correctnessmodule:clientstale
  #3844·rjvkr2021 opened Sep 9, 2026·1 task done·

  ·
  18

• [CELEBORN-2457][CIP-24] Adaptive partition write parallelism

  kind:documentationmodule:clientmodule:commonmodule:testsmodule:worker
  #3843·yew1eb opened Sep 8, 2026Contributor·1 of 2 tasks·

  ·
  1

• [CELEBORN-2458] Invalidate transport client after push timeout

  module:commonmodule:testsmodule:worker
  #3842·venkata91 opened Sep 7, 2026·

  ·
  10

• [CELEBORN-2452] Prevent device error broadcast from deleting committed shuffle files

  correctnessmodule:workerstale
  #3837·wang-haihua opened Sep 3, 2026Contributor·1 task done·

  ·
  5

• [CELEBORN-2451] Make DRA safe under AUTO fallback by tying supportsReliableStorage to fallback policy

  correctnesskind:documentationmodule:clientmodule:spark
  #3834·venkata91 opened Aug 31, 2026·2 tasks done·

  ·
  14

• [CELEBORN-2361] Unblock stage-end when CommitFilesResponse carries an empty map-id bitmap

  correctnessmodule:clientmodule:common
  #3833·strelok89 opened Aug 31, 2026Contributor·1 of 2 tasks·

  ·
  5

• [CELEBORN-2446] Support topologySpreadConstraints for master and worker pods in Helm chart

  kind:deploystale
  #3831·strelok89 opened Aug 29, 2026Contributor·1 of 2 tasks·

  ·
  3