carlory
aab7a079fa
make each scheduler test independent
...
Signed-off-by: carlory <baofa.fan@daocloud.io >
2025-03-13 14:39:50 +08:00
dom4ha
4deb4f2b5f
Trigger rescheduling on delete event also when unscheduled pod is removed
2025-03-10 15:03:50 +00:00
Kubernetes Prow Robot
4806519423
Merge pull request #130622 from pohly/DRAPrioritizedList
...
DRA: Prioritized Alternatives in Device Requests, II
2025-03-10 06:51:51 -07:00
Kubernetes Prow Robot
c7088e6437
Merge pull request #130260 from Bowser1704/fix/skip-best-effort-pods
...
fix(scheduler): skip best-effort pods in BalancedAllocation PreScore
2025-03-07 01:11:45 -08:00
Patrick Ohly
dfb8ab6521
DRA scheduler: fail in PreFilter when DRAPrioritizedList is disabled and used
...
This was previously caught during Filter by the allocator check. Doing it
sooner avoids wasting resources on a pod which ultimately cannot get scheduled.
While at it, be a bit more clear about which feature is disabled. The user
might not know that.
2025-03-07 08:45:32 +01:00
Hongqi Yu
d76f40d2f3
fix(scheduler): skip best-effort pods in BalancedAllocation PreScore
...
- Refactored `PreScore` method in `balanced_allocation.go` to skip
best-effort pods.
- Updated unit tests in `balanced_allocation_test.go` to check for
the new status codes.
2025-03-07 13:13:02 +08:00
Kubernetes Prow Robot
9d45ea8b9d
Merge pull request #128586 from mortent/DRAPrioritizedList
...
Prioritized Alternatives in Device Requests
2025-03-06 21:01:44 -08:00
Kubernetes Prow Robot
0f7becbc44
Merge pull request #130537 from saintube/scheduler-expose-nodeinfo-in-score
...
Expose NodeInfo to the ScorePlugin
2025-03-06 06:53:55 -08:00
Maciej Skoczeń
2fc3cd90b1
Store pod backoff expiration time in QueuedPodInfo
2025-03-06 10:45:38 +00:00
saintube
afb4e96510
Expose NodeInfo to Score plugins
...
Co-authored-by: shenxin <rougang.hrg@alibaba-inc.com >
Signed-off-by: saintube <saintube@foxmail.com >
2025-03-04 17:57:14 +08:00
Kubernetes Prow Robot
58704903c5
Merge pull request #129887 from rhrmo/STOR-2270-make-failedscheduling-event-more-intuitive
...
Change volume affinity scheduling error message to be more intuitive
2025-03-03 22:51:48 -08:00
Richard Hrmo
526960f988
Change FailedScheduling Event message to be more intuitive
2025-03-03 16:56:18 +01:00
Morten Torkildsen
2229a78dfe
DRA: Update allocator for Prioritized Alternatives in Device Requests
2025-02-28 19:30:10 +00:00
Kubernetes Prow Robot
b8c95e1954
Merge pull request #129688 from cpanato/update-main-go124
...
[go] Bump images, dependencies and versions to go 1.24.0
2025-02-27 09:10:30 -08:00
Kubernetes Prow Robot
facb1a8c55
Merge pull request #129905 from ania-borowiec/129778_replace_equal
...
Replace reflect.DeepEqual with cmp.Diff in pkg/scheduler tests
2025-02-26 08:24:30 -08:00
googs1025
239aad8e4b
chore(scheduler): use framework.Features in scheduler plugins
2025-02-26 19:16:07 +08:00
Jordan Liggitt
8090db5dcf
Switch to private instances of rand for seeding for tests
2025-02-26 11:27:10 +01:00
Ania Borowiec
4205f04ce3
Replace uses of reflect.DeepEqual with cmp.Diff in pkg/scheduler tests
2025-02-26 09:27:51 +00:00
Kubernetes Prow Robot
4032177faf
Merge pull request #129557 from googs1025/feature/add_QueueingHint_for_VolumeAttachment_deletion_events
...
feature(scheduler): add queueinghint for volumeattachment deletion
2025-02-22 00:10:26 -08:00
googs1025
86f504284c
feature(scheduler): add queueinghint for volumeattachment deletion
2025-02-22 14:57:41 +08:00
googs1025
004c5f5a39
chore: remove unnecessary check for node is zero
2025-02-18 10:24:26 +08:00
googs1025
938b5f8262
chore(scheduler): add more log info when no preemption candidates is found in preemption
2025-02-11 13:53:32 +08:00
Kensei Nakada
3701e39327
fix: remove the mention that DRA uses Pending
2025-02-04 06:45:05 +09:00
Kubernetes Prow Robot
1b7a059187
Merge pull request #128999 from macsko/improve_goroutines_metric_writes_in_parallelizer_until
...
Improve Goroutines metric calls in parallelizer.Until
2025-02-03 07:30:57 -08:00
Kubernetes Prow Robot
fc268ecd09
Merge pull request #129823 from googs1025/chore/log_improve
...
fix(dra plugin): when there is no resourceclaim, return directly
2025-02-02 16:28:56 -08:00
googs1025
ed826dddfe
fix(dra plugin): when there is no resourceclaim, return directly
2025-01-29 08:47:52 +08:00
Maciej Skoczeń
bd8dee9637
Improve Goroutines metric calls in parallelizer.Until
2025-01-27 12:01:17 +00:00
Davanum Srinivas
4e05bc20db
Linter to ensure go-cmp/cmp is used ONLY in tests
...
Signed-off-by: Davanum Srinivas <davanum@gmail.com >
2025-01-24 20:49:14 -05:00
Kubernetes Prow Robot
2056dbca18
Merge pull request #129697 from saza-ku/remove-pvc-not-found
...
chore: remove duplicated test case of `pvc not found`
2025-01-24 09:35:21 -08:00
googs1025
27ec5de863
chore(scheduler): improve log output for podtopologyspread filter plugin
2025-01-24 19:29:23 +08:00
Kubernetes Prow Robot
a716ea756d
Merge pull request #129635 from macsko/cache_calculateresource_result_when_adding_pod_to_nodeinfo
...
Use cached calculateResource result when removing pod from NodeInfo in preemption
2025-01-21 04:16:37 -08:00
Maciej Skoczeń
0452ae402a
Use cached calculateResource result when removing pod from NodeInfo in preemption
2025-01-21 10:02:57 +00:00
saza-ku
e26fb1c393
remove duplicated test case of pvc not found
2025-01-20 01:52:35 +09:00
Kubernetes Prow Robot
0b789d7cca
Merge pull request #129427 from macsko/improve_map_in_interpodaffinity_prefilter
...
Improve topologyToMatchedTermCount map in InterPodAffinity PreFilter
2025-01-10 10:40:33 -08:00
Maciej Skoczeń
2d82687114
Improve topologyToMatchedTermCount map in InterPodAffinity PreFilter
2025-01-10 10:55:49 +00:00
Paco Xu
2653caa248
fix dra test lint
2025-01-09 10:42:40 +08:00
googs1025
77eae7c34f
feature(scheduler): remove dra plugin resourceslice QueueingHintFn
2025-01-08 16:24:28 +08:00
Kubernetes Prow Robot
1c2b2cce10
Merge pull request #129119 from macsko/fix_podtopologyspread_for_multiple_constraints_with_the_same_key
...
Fix PodTopologySpread matching pods counts for constraints with the same topologyKey
2025-01-01 11:04:14 +01:00
Maciej Skoczeń
c3a54926a4
Fix PodTopologySpread matching pods counts for constraints with the same topologyKey
2024-12-30 09:35:24 +00:00
Kubernetes Prow Robot
078664b424
Merge pull request #129023 from zhifei92/cleanup-actiontype
...
scheduler: Rename UpdatePodTolerations for code style consistency
2024-12-12 05:28:52 +00:00
zhifei92
27608fa25d
refactor(scheduler): Rename UpdatePodTolerations for code style consistency.
2024-11-29 13:13:09 +08:00
googs1025
c725e18e07
feature(scheduler): more fine-grained QHints for interpodaffinity plugin
2024-11-14 20:00:38 +08:00
Kubernetes Prow Robot
8fe10dc378
Merge pull request #128691 from sanposhiho/flaky-candidate
...
fix: flaky TestPrepareCandidate
2024-11-09 19:38:44 +00:00
Kensei Nakada
00f7b95b7b
fix: flaky TestPrepareCandidate
2024-11-09 12:07:16 +09:00
ndixita
6db40446de
Scheduler changes:
...
1. Use pod-level resource when feature is enabled and resources are set at pod-level
2. Edge case handling: When a pod defines only CPU or memory limits at pod-level (but not both), and container-level requests/limits are unset, the pod-level requests stay empty for the resource without a pod-limit. The container's request for that resource is then set to the default request value from schedutil.
2024-11-08 03:00:54 +00:00
Kensei Nakada
e601eb7c5a
fix: run activate() only when fail
2024-11-07 14:09:35 +09:00
Kensei Nakada
9b5928f764
fix: refactor TestPrepareCandidate
2024-11-07 14:09:35 +09:00
Kensei Nakada
d4d91d4ace
fix: use set methods
2024-11-07 14:09:35 +09:00
Kensei Nakada
a95b8b5085
fix: use Activate always
2024-11-07 14:09:35 +09:00
Kensei Nakada
8be51d3f32
chore: fix based on reviews
2024-11-07 14:09:35 +09:00
