dev  

Messages by Thread

• [jira] [Created] (YUNIKORN-3458) Premature victim loop termination in calculateVictimsByNode starves preemption Hedger Lai (Jira)
• [jira] [Created] (YUNIKORN-3457) Add SECURITY.md to document vulnerability reporting process and security model for yunikorn-core Mit Desai (Jira)
• [jira] [Resolved] (YUNIKORN-3428) Application.taskMap read without the lock from String() and AreAllTasksTerminated Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3456) K8s 1.37 API changes in Shim Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3455) Add K8s 1.37 to the e2e tests in CI Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3454) [Umbrella] Kubernetes 1.37 support Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3449) [Core] Preemption falsely aborts and starves large asks by ignoring node available capacity in shortfall check Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3453) Send the plugin errors back to core for preemption ask Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3452) Node Preemption: Fix reversed penalty scores between originator and opted-out pods weichen lai (Jira)
• [jira] [Created] (YUNIKORN-3451) replace subprocess call with run Wilfred Spiegelenburg (Jira)
• [ANNOUNCE] YuniKorn 1.10.0 preview release Wilfred Spiegelenburg
• [jira] [Created] (YUNIKORN-3449) [Core] Preemption falsely aborts and starves large asks by ignoring node available capacity in shortfall check Hedger Lai (Jira)
• [jira] [Created] (YUNIKORN-3450) [Core] False preemption shortfall aborts queue preemption and starves large tasks despite sufficient guaranteed quota Hedger Lai (Jira)
• [ANNOUNCE] Branch cut for YuniKorn 1.10.0 Wilfred Spiegelenburg
• [jira] [Resolved] (YUNIKORN-3445) SortAllocationsBasedOnAsk prioritizes originator and opted-out pods for preemption instead of protecting them Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3439) Ensure task.checkPodPVCs() correctness Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3371) DRA resource-slice tracker goroutine leaks in tests: created with an uncancellable context and no shutdown handle Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3431) Add regression test coverage for KubernetesShim scheduling loop shutdown Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3447) Drop managed fields in the informers Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3137) Fails to preempt more than 2 victims for a larger ask. Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3448) Revert autoscaling state when a scale-up-triggered ask loses queue headroom Kishan Ghetia (Jira)
• [jira] [Created] (YUNIKORN-3447) Drop managed fields in the informers Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3300) Use feasible nodes returned from K8s PreFilter plugin in core itself Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3365) Shutdown ordering: scheduler allocation notify can block on an RM reply the proxy never drains TingYao Huang (Jira)
• [jira] [Resolved] (YUNIKORN-3379) Scheduler panics on a malformed PLACEHOLDER_REPLACED release with no linked replacement allocation TingYao Huang (Jira)
• [jira] [Resolved] (YUNIKORN-3358) Remove the unreachable replace-existing-ask branch from Application.AddAllocationAsk Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3420) String() methods that take their own lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3446) [UGM] Race condition in ensureGroupTrackerForApp creates duplicate GroupTrackers and bypasses group quota Hedger Lai (Jira)
• [jira] [Resolved] (YUNIKORN-3381) Foreign pod that terminates while orphaned is later adopted as a live allocation Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3445) [Core] SortAllocationsBasedOnAsk prioritizes originator and opted-out pods for preemption instead of protecting them weichen lai (Jira)
• [jira] [Resolved] (YUNIKORN-3444) merge_pr.sh: tail +2 fails on GNU tail Chia-Ping Tsai (Jira)
• [jira] [Created] (YUNIKORN-3444) merge_pr.sh: tail +2 fails on GNU tail Chia-Ping Tsai (Jira)
• [jira] [Resolved] (YUNIKORN-3441) Support podman for building images and running e2e tests Chia-Ping Tsai (Jira)
• [jira] [Created] (YUNIKORN-3443) Support Podman in release tools for building images and validation weichen lai (Jira)
• [jira] [Created] (YUNIKORN-3442) [Core] Downstream TryPreemption drops K8s predicate victims and falsely rejects preemption when node has available capacity Hedger Lai (Jira)
• [jira] [Resolved] (YUNIKORN-3409) AddRejectedApplication writes the rejected-application map without the partition lock Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3380) RemoveAllAllocations leaks user/group tracked usage when a release-all arrives with pending asks Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3435) upgrade go dependencies for CVEs Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3364) Event stream history replay blocks forever if the client disconnects before the stream opens with a non-zero history request TingYao Huang (Jira)
• [jira] [Resolved] (YUNIKORN-3436) Event stream forwarder leaks on slow-consumer eviction: in-loop consumer send is not selectable TingYao Huang (Jira)
• [jira] [Resolved] (YUNIKORN-3440) Use `https` instead of `http` in the README files TingYao Huang (Jira)
• [jira] [Created] (YUNIKORN-3441) Support podman for building images and running e2e tests Chia-Ping Tsai (Jira)
• [jira] [Created] (YUNIKORN-3440) Use `https` instead of `http` in the README files Chia-Ping Tsai (Jira)
• [jira] [Created] (YUNIKORN-3439) Ensure task.checkPodPVCs() correctness Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3438) Identify Informers not being used Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3437) Handle shutdown signal during K8s CRUD operations Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3369) AsyncRMCallback.UpdateAllocation's AssumePod retry blocks the RM proxy event loop and cannot be interrupted by shutdown Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3436) Event stream forwarder leaks on slow-consumer eviction: in-loop consumer send is not selectable Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3372) Event-stream tests create a stream via CreateEventStream and never call RemoveStream, leaking the forwarder goroutine Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3434) upgrade protoc in SI Makefile Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3416) UserGroupCache cleanup locks the singleton instead of the receiver; nil dereference after Stop() Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3432) Cleanup LDAP resolver Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3435) upgrade go dependencies for CVEs Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3434) upgrade protoc in SI Makefile Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3433) Send the plugin errors back to core Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3432) Cleanup LDAP resolver Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3393) Update go_repro_vesion for CVE fixes Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3315) Remove old placeholder flags and processing Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3320) tryReservedAllocate can release required node reservations Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3431) Add regression test coverage for KubernetesShim scheduling loop shutdown Andy Huang (Jira)
• [jira] [Created] (YUNIKORN-3423) postAppAccepted reads taskGroups and taskMap without the application lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3420) String() methods that take their own lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3424) createAppPlaceholders walks the application task map without the application lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3428) Application.taskMap read without the lock from String() and AreAllTasksTerminated Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3429) onReserving's goroutine reads originatingTask after the lock is released Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3430) registerNodesInternal releases and retakes the context lock around the wait group Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3419) Queue fields read without the queue lock in parent-first paths and the constructors Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3421) shouldAppRelease drops the task lock inside an FSM callback: deadlock with a concurrent task event Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3422) SchedulerCache node-list getters populate their cache under the read lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3426) Pod bind retries with backoff run under the task write lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3418) User/group resolution runs under the PartitionContext read lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3417) RM event replies are unbuffered sends made under the ClusterContext write lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3427) flushReleaseableTasks removes from the context map and reads task state under the application lock only Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3425) Admission controller serving goroutine reads the server field Shutdown nils Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3413) Wildcard limit config read without the manager lock on scheduling paths Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3409) AddRejectedApplication writes the rejected-application map without the partition lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3411) Application write lock held across the RM release round trip Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3416) UserGroupCache cleanup locks the singleton instead of the receiver; nil dereference after Stop() Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3414) Manager reaches into user/group tracker internals without the tracker lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3412) Event system handler reads the fields Stop() rewrites; Stop() blocks under the write lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3415) Node.String() reads guarded fields without the lock and cannot take it Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3408) Adopt the vet-lock static lock analyser in core and k8shim (lock annotations + build wiring) Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3410) Partition removal from a config reload or RM re-registration self-deadlocks on the ClusterContext lock Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3398) Fix unit test failures due to RollbackAllocation behavior conflict Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3374) Phantom allocation when a node with an assumed pod is removed and re-added Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3352) Replace the O(n) ask-priority rescan with a pending histogram Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3400) Cleanup outdated quota.preemption.delay inheritance note in docs Yu-Lin Chen (Jira)
• [jira] [Created] (YUNIKORN-3407) [k8shim] Add DEBUG_BUILDS=1 make image target support KAI-JAY TU (Jira)
• [jira] [Created] (YUNIKORN-3406) Support queue-headroom-based limits for application unschedulable backoff Kishan Ghetia (Jira)
• [jira] [Resolved] (YUNIKORN-3266) Document remote container debugging steps for scheduler Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3367) KubernetesShim.Stop() sends to the unbuffered stopChan instead of closing it, leaving one doScheduling loop running after stop Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3355) Failed bind leaves a stale node assignment in the shim cache Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3399) Align RollbackAllocation tests with the ask-presence guard from YUNIKORN-3360 Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3404) PlaceholderManager Start and Stop operations have race conditions Andy Huang (Jira)
• [jira] [Created] (YUNIKORN-3405) Dispatcher Start and Stop operations have race conditions Andy Huang (Jira)
• [jira] [Resolved] (YUNIKORN-3368) KubernetesShim.Stop() after a failed Run() is a no-op, leaking the dispatcher and placeholder-manager goroutines Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3378) TestSchedulerRecoveryQuotaPreemption is flaky Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3354) Recovery pod ordering is not reproducible despite intending to be Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3403) Use cycle state in scheduler cache Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3402) Use cycle state to Get and Find Pod volumes Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3401) Fix nonamedreturns lint error in application_property_test.go Yu-Lin Chen (Jira)
• [jira] [Created] (YUNIKORN-3401) Fix nonamedreturns lint error in application_property_test.go weichen lai (Jira)
• [jira] [Created] (YUNIKORN-3400) Cleanup outdated quota.preemption.delay inheritance note in docs Yu-Lin Chen (Jira)
• [jira] [Created] (YUNIKORN-3399) Align RollbackAllocation tests with the ask-presence guard from YUNIKORN-3360 Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3398) Fix unit test failures due to RollbackAllocation behavior conflict weichen lai (Jira)
• [jira] [Resolved] (YUNIKORN-3293) Update perf-tools dependencies for yunikorn-release TingYao Huang (Jira)
• [jira] [Resolved] (YUNIKORN-3359) run e2e tests with deadlock detection enabled Yu-Lin Chen (Jira)
• [jira] [Created] (YUNIKORN-3397) release artifact push Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3394) [UMBRELLA] YuniKorn 1.10.0 release efforts Wilfred Spiegelenburg (Jira)
• [ANNOUNCE] Branching for YuniKorn 1.10 Wilfred Spiegelenburg
• [jira] [Created] (YUNIKORN-3396) Update website for 1.10.0 Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-2804) [Umbrella] Rethink general retry policy for post allocation failed task Aditya Maheshwari (Jira)
• [jira] [Created] (YUNIKORN-3395) Release notes for 1.10.0 Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3128) Yunikorn ignores pending pods after apiserver errors Aditya Maheshwari (Jira)
• [jira] [Created] (YUNIKORN-3393) Update gfo_repro_vesion for CVE fixes Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3392) DRA Bindings using appropriate K8s Plugins Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3391) Enhancing Scheduling decisions for DRA Pods Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3390) Assess DRA Informers Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3347) Improve scheduler config checksum generation and checks Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3332) Prevent timed out Placeholder creation Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-2884) Task fail with post allocated but the pod will keep pending Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3389) Fix pre-existing broken anchors reported by pnpm build in versioned docs KAI-JAY TU (Jira)
• [jira] [Created] (YUNIKORN-3388) Update quota_preemptor design doc to match current implementation KAI-JAY TU (Jira)
• [jira] [Resolved] (YUNIKORN-3345) Turn on nonamedreturns linter Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3360) RollbackAllocation should safety check request existence Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3387) Deterministic synctest reproduction pilot Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3386) Chaos scenarios for preemption under fault injection. Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3385) Informer event-ordering property fuzzer Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3384) Partition accounting property fuzzer Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3383) Gang / placeholder chaos scenarios + full state-diff oracle Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3382) Failed victim pod delete permanently wedges a preemption; release-path DeleteTaskPod is never retried Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3381) Foreign pod that terminates while orphaned is later adopted as a live allocation Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3380) RemoveAllAllocations leaks user/group tracked usage when a release-all arrives with pending asks Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3379) Scheduler panics on a malformed PLACEHOLDER_REPLACED release with no linked replacement allocation Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3378) TestSchedulerRecoveryQuotaPreemption is flaky Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3361) Improve performance tests in yunikorn-core Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3349) Enhance Developer concurrency documentation Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3376) Data race on Task.pod: checkPodMetadataBeforeScheduling reads without the task lock Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3377) Add fault-injection tests with phantom-allocation oracles for the shim Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3376) Data race on Task.pod: checkPodMetadataBeforeScheduling reads without the task lock Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3375) Node delete/re-add silently corrupts queue accounting: adopted pods are never re-reported to the core Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3374) Phantom allocation when a node with an assumed pod is removed and re-added Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3373) Add an invariant property fuzzer for the shim scheduler cache Dale Richardson (Jira)
• A nice performance win to share Dale Richardson
  • Re: A nice performance win to share Dale Richardson
  • A nice performance win to share Dale Richardson
• [jira] [Resolved] (YUNIKORN-3366) Instrumented-package tests leak core-service goroutines by not stopping the services they start Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3372) Event-stream tests create a stream via CreateEventStream and never call RemoveStream, leaking the forwarder goroutine Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3363) Event system restarts after Stop(): configmap callback never deregistered, leaking handler and publisher goroutines Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3369) AsyncRMCallback.UpdateAllocation's AssumePod retry blocks the RM proxy event loop and cannot be interrupted by shutdown Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3371) DRA resource-slice tracker goroutine leaks in tests: created with an uncancellable context and no shutdown handle Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3370) yunikorn-core services are not restartable in-process, so the shim cannot stop the core it starts without intermittent failures Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3368) KubernetesShim.Stop() after a failed Run() is a no-op, leaking the dispatcher and placeholder-manager goroutines Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3367) KubernetesShim.Stop() sends to the unbuffered stopChan instead of closing it, leaving one doScheduling loop running after stop Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3366) Instrumented-package tests leak core-service goroutines by not stopping the services they start Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3365) Shutdown race: scheduler allocation notify blocks forever because the RM proxy stops without draining pending events Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3364) Event stream forwarder leaks permanently when a consumer stalls; slow-consumer eviction cannot release it Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3363) Event system restarts after Stop(): configmap callback never deregistered, leaking handler and publisher goroutines Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3362) Remove asks that have been scheduled from sortedRequests Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3331) Run preemption preconditions before instantiating preemptor Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3348) Develop doc fixups Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3216) Document unschedulable backoff settings Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3343) Flaky Event publisher test Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3362) Remove asks that have been scheduled from sortedRequests Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3361) Improve performance tests in yunikorn-core Aditya Maheshwari (Jira)
• [jira] [Created] (YUNIKORN-3360) RollbackAllocation should safety check request existence Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3359) run e2e tests with deadlock detection enabled Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3358) Remove the unreachable replace-existing-ask branch from Application.AddAllocationAsk Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3357) Add goroutine leak detection to core tests via uber-go/goleak Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3316) Retry pod scheduling in case of failure during AssumePod Aditya Maheshwari (Jira)
• [jira] [Resolved] (YUNIKORN-3351) Hoist the per-ask backoff getter out of tryAllocate Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3323) Replace google btree for node structures Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3319) Reservations released by preemption not tracked Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3356) Split the shim's Kubernetes clients by concern (writes / informers / events) with per-concern rate limits and User-Agents Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3355) Failed bind leaves a stale node assignment in the shim cache Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3354) Recovery pod ordering is not reproducible despite intending to be Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3353) Keep only pending asks in sortedRequests Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3352) Replace the O(n) ask-priority rescan with a pending histogram Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3351) Hoist the per-ask backoff getter out of tryAllocate Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3350) Fix quadratic performance of ask scans in the application allocation cycle Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3349) Enhance Developer concurrency documentation Dale Richardson (Jira)
• [jira] [Created] (YUNIKORN-3348) Develop doc fixups Dale Richardson (Jira)
• [jira] [Resolved] (YUNIKORN-3335) flaky test TestQuotaChangeTryPreemptionForParentQueue Wilfred Spiegelenburg (Jira)
• [DISCUSS] YuniKorn 1.9.1 release Wilfred Spiegelenburg
• [jira] [Created] (YUNIKORN-3347) Improve scheduler config checksum generation and checks Siddharth Krishna (Jira)
• [jira] [Created] (YUNIKORN-3346) AssumePod does not return error if node or pod is not found Aditya Maheshwari (Jira)
• [jira] [Created] (YUNIKORN-3345) Turn on nonamedreturns linter Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3338) Golden test trace generation and validation Wilfred Spiegelenburg (Jira)
• [jira] [Created] (YUNIKORN-3344) Remove deprecated user label references from site documentation KAI-JAY TU (Jira)
• [jira] [Created] (YUNIKORN-3343) Flaky Event publisher test Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3334) flaky unit test TestQuotaChangeTryPreemptionWithDifferentResTypes Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3325) Deduplicate cancel reservations code flow Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3342) Fix TestQuotaChangeTryPreemptionWithDifferentResTypes flakiness Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3341) gzip middleware double-compresses responses that handlers already encoded Wilfred Spiegelenburg (Jira)
• [jira] [Resolved] (YUNIKORN-3321) removeAsksInternal return ignored Manikandan R (Jira)
• [jira] [Created] (YUNIKORN-3342) Fix TestQuotaChangeTryPreemptionWithDifferentResTypes flakiness Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3311) Remove user label processing code Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3220) Test setpreemptiontime with different resource types Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3333) flaky unit test TestQuotaChangeTryPreemption Manikandan R (Jira)
• [jira] [Resolved] (YUNIKORN-3277) expose ask backoff and quota preemption in REST Manikandan R (Jira)

• Earlier messages