Releases: kubernetes-sigs/descheduler
descheduler-helm-chart-0.25.2
Descheduler for Kubernetes is used to rebalance clusters by evicting pods that can potentially be scheduled on better nodes. In the current implementation, descheduler does not schedule replacement of evicted pods but relies on the default scheduler for that.
Descheduler v0.25.1
Image for this release is available at k8s.gcr.io/descheduler/descheduler:v0.25.1
What's Changed
- Update helm chart version to v1.25.0 by @pravarag in #936
- backport 7349b39 (issue 960) into release-1.25 by @knelasevero in #962
- [release-1.25] Doc updates for v0.25.1 by @damemi in #963
Full Changelog: v0.25.0...v0.25.1
Contributors
Assets 2
descheduler-helm-chart-0.25.1
Descheduler for Kubernetes is used to rebalance clusters by evicting pods that can potentially be scheduled on better nodes. In the current implementation, descheduler does not schedule replacement of evicted pods but relies on the default scheduler for that.
descheduler-helm-chart-0.25.0
Descheduler for Kubernetes is used to rebalance clusters by evicting pods that can potentially be scheduled on better nodes. In the current implementation, descheduler does not schedule replacement of evicted pods but relies on the default scheduler for that.
Descheduler v0.25.0
This release includes bug fixes, improvements, and many updates toward the refactoring effort involved in the descheduler framework.
Image for this release is available at k8s.gcr.io/descheduler/descheduler:v0.25.0
What's Changed
- Bump To Go 1.18.2 by @seanmalloy in #801
- bump: golang.org/x/crypto by @damemi in #809
- Add podAnnotations and podLabels to values and docs by @stephan2012 in #814
- Arguments must be strings, not bool or number by @stephan2012 in #813
- Update release guide docs by @damemi in #818
- Update helm tests util versions and release guide by @damemi in #821
- Fix version command to parse helm chart tags by @damemi in #823
- Increase cloudbuild timeout to 25 minutes by @damemi in #825
- cleanup lownodeutilization code by @xiaoanyunfei in #805
- Revert "cleanup lownodeutilization code" by @a7i in #830
- PodLifeTime: sort pods by creation timestamp by @a7i in #833
- nodeFit: fix docs spacing for placement of the field by @a7i in #840
- Update docs, manifests for 0.24.1 on master by @damemi in #843
- Check existing constraints before assigning by @knelasevero in #852
- remove random creation timestamp from pod sort by age test by @a7i in #856
- Split pod evictor and evictor filter by @ingvagabund in #847
- PodLifeTime: support container states PodInitializing and ContainerCreating by @a7i in #834
- Improving helm setup by @knelasevero in #854
- TopologySpreadConstraint: only evaluate nodes below ideal avg when balancing domains by @a7i in #836
- fix log-file and log-dir issue by @JaneLiuL in #872
- fix: chart path can't be relative by @knelasevero in #871
- Use lease resource name from Helm value by @iijimakazuyuki in #876
- code approvers: add a7i by @a7i in #883
- Prepare pod evictor for the descheduling framework plugin by @ingvagabund in #846
- Add default lease resource name in Helm chart's ClusterRole by @iijimakazuyuki in #882
- Add EvictOptions struct to EvictPod() by @damemi in #885
- Migrate RemovePodsViolatingNodeTaints to a plugin by @ingvagabund in #857
- Remove RemovePodsViolatingNodeTaints strategy already migrated into a plugin by @ingvagabund in #894
- fix: events memory leak. Using new events implementation and take recorder out of EvictPod by @knelasevero in #888
- Remove emeritus_approvers from reviewers by @ingvagabund in #896
- Migrate RemoveFailedPods to plugin by @a7i in #861
- separate args validation for better reuse by @a7i in #899
- add helm ct install. by @knelasevero in #904
- Migrate RemovePodsViolatingNodeAffinity to plugin by @knelasevero in #860
- Migrate RemovePodsHavingTooManyRestarts to plugin by @BinacsLee in #902
- NodeAffinity plugin to use the existing validation methods by @a7i in #906
- bring removeduplicates to plugin by @JaneLiuL in #892
- introduce ct for local helm install test by @knelasevero in #911
- CONTAINER_ENGINE to override the default docker engine by @ingvagabund in #912
- Migrate PodLifeTime to plugin by @a7i in #913
- RemovePodsViolatingTopologySpreadConstraint Plugin by @jklaw90 in #908
- add security scan into gh-actions by @JaneLiuL in #898
- NodeFit: do not check whether node fitsRequest when a pod is already assigned to the node by @ingvagabund in #916
- Migrate RemovePodsViolatingInterPodAntiAffinity into a plugin by @knelasevero in #903
- bring lownodeutilization and highnodeutilization to plugin by @JaneLiuL in #910
- removing dupe plugin interface check by @jklaw90 in #922
- Bump to k8s 1.25 by @a7i in #933
- remove dependency on kubectl by @a7i in #938
- Add RemovePodsViolatingTopologySpreadConstraint to values.yaml by @gallowaystorm in #943
- clarify which version PodLifeTime introduced states parameter and deprecated podStatusPhases by @a7i in #928
- feat: add RemovePodsHavingTooManyRestarts to values.yaml by @gallowaystorm in #945
- Update docs & manifests for v0.25.0 by @pravarag in #935
- remove TODO comments for cronjob v1beta1 support by @a7i in #931
- Add new DefaultEvictor plugin with args by @knelasevero in #929
- feat: change DeschedulerVersion and GitVersion labels by @eminaktas in #947
New Contributors
- @stephan2012 made their first contribution in #814
- @xiaoanyunfei made their first contribution in #805
- @iijimakazuyuki made their first contribution in #876
- @gallowaystorm made their first contribution in #943
Full Changelog: v0.24.1...v0.25.0
Contributors
Assets 2
descheduler-helm-chart-0.24.1
Descheduler for Kubernetes is used to rebalance clusters by evicting pods that can potentially be scheduled on better nodes. In the current implementation, descheduler does not schedule replacement of evicted pods but relies on the default scheduler for that.
Descheduler v0.24.1
Image for this release is available at docker run k8s.gcr.io/descheduler/descheduler:v0.24.1
What's Changed
- Update helm chart version to v0.24 by @JaneLiuL in #796
- [release-1.24] Bump To Go 1.18.2 by @damemi in #808
- [release-1.24] bump: golang.org/x/crypto by @damemi in #810
- [release-1.24] helm chart fixes by @damemi in #817
- [release-1.24] Fix version command to parse helm chart tags by @damemi in #824
Full Changelog: v0.24.0...v0.24.1
Contributors
Assets 2
descheduler-helm-chart-0.24.0
Descheduler for Kubernetes is used to rebalance clusters by evicting pods that can potentially be scheduled on better nodes. In the current implementation, descheduler does not schedule replacement of evicted pods but relies on the default scheduler for that.
Descheduler v0.24.0
Container image for this release is available at: k8s.gcr.io/descheduler/descheduler:v0.24.0
What's Changed
- fix helmchart fail to watch namespace issue by @JaneLiuL in #712
- [release-1.23.1] Update helm chart version to v0.23.1 by @JaneLiuL in #717
- OWNERS: add a7i as a reviewer by @a7i in #697
- Update charts README to reflect the new parameters by @ingvagabund in #704
- fix: add imagePullSecrets for deployment resource by @eminaktas in #716
- Eviction Logs by @jklaw90 in #726
- fix: Remove deprecated parameters from cmdOptions and add the parameters under policy by @eminaktas in #715
- Clarify resource calculations in NodeUtilization strategy Readmes by @damemi in #708
- Bugfix: Cronjob ctx cancel by @jklaw90 in #731
- remove MostRequestedPriority from doc since already deprecated by @JaneLiuL in #733
- Update golang image by @antonio-te in #744
- Update manifests and doc for v0.23.1 by @damemi in #748
- fix doc about NPD description by @JaneLiuL in #740
- Share links to all descheduler ehnacements proposals in the project repo by @JaneLiuL in #739
- add conflicting taint to the logs by @prune998 in #757
- Update docs for maxNoOfPodsToEvictPerNamespace by @hiroyaonoe in #758
- [e2e] TestTooManyRestarts: check err and len before accessing pod items by @ingvagabund in #760
- [e2e] TestTooManyRestarts: check if container status is set before accessing by @ingvagabund in #761
- OWNERS: add janeliul as a reviewer by @JaneLiuL in #759
- RemovePodsViolatingNodeTaints: optionally include PreferNoSchedule taint by @ingvagabund in #764
- Promote NodeUsage to NodeInfo, evaluate thresholds separately by @ingvagabund in #762
- Add RemovePodsViolatingNodeTaints taint exclusion by @JacobHenner in #777
- feat: Add DeviationThreshold Paramter for LowNodeUtilization, (Previous attempt - #473 ) by @HelmutLety in #751
- feat(leaderelection): impl leader election for HA Deployment by @Dentrax in #722
- Sort nodes in ascending order for HighNodeUtilization by @dineshbhor in #771
- feat: Add metric scrape configs in Helm Chart by @eminaktas in #732
- Fix missing param in user-guide for PodLifeTime strategy by @pravarag in #779
- bump to k8s 1.24-rc.0 by @JaneLiuL in #787
- fix: incorrect yaml indentation in readme by @ryan4yin in #788
- Added request considerations to NodeFit Feature [#636 follow up] by @ingvagabund in #790
- Update e2e test versions by @damemi in #795
- Update Docs and Manifests for v0.24.0 by @pravarag in #792
- Bump to k8s 1.24.0 by @JaneLiuL in #791
New Contributors
- @eminaktas made their first contribution in #716
- @antonio-te made their first contribution in #744
- @prune998 made their first contribution in #757
- @hiroyaonoe made their first contribution in #758
- @JacobHenner made their first contribution in #777
- @HelmutLety made their first contribution in #751
- @Dentrax made their first contribution in #722
- @dineshbhor made their first contribution in #771
- @ryan4yin made their first contribution in #788
Full Changelog: v0.23.0...v0.24.0
Contributors
Assets 2
Descheduler v0.23.1
Container image for this release is available at: k8s.gcr.io/descheduler/descheduler:v0.23.1
This release contains a fix for a panic when running Descheduler v0.23 as a cronjob (reported in #728 #741 #743), fixed by @jklaw90 in #731
It also bumps the Go version to 1.17.7 to address CVE-2021-44716 as identified by @antonio-te in #744