CKS: use --delete-emptydir-data instead of deprecated --delete-local-data#10234
CKS: use --delete-emptydir-data instead of deprecated --delete-local-data#10234DaanHoogland merged 1 commit intoapache:4.19from
Conversation
|
@blueorangutan package |
|
@weizhouapache a [SL] Jenkins job has been kicked to build packages. It will be bundled with KVM, XenServer and VMware SystemVM templates. I'll keep you posted as I make progress. |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## 4.19 #10234 +/- ##
============================================
- Coverage 15.13% 15.13% -0.01%
+ Complexity 11279 11276 -3
============================================
Files 5408 5408
Lines 474007 474007
Branches 57822 57822
============================================
- Hits 71745 71734 -11
- Misses 394241 394254 +13
+ Partials 8021 8019 -2
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
|
Packaging result [SF]: ✔️ el8 ✔️ el9 ✔️ debian ✔️ suse15. SL-JID 12160 |
|
@blueorangutan test keepEnv |
|
@weizhouapache a [SL] Trillian-Jenkins test job (ol8 mgmt + kvm-ol8) has been kicked to run smoke tests |
|
[SF] Trillian test result (tid-12163)
|
DaanHoogland
left a comment
There was a problem hiding this comment.
clgtm
needs regressions testing and maybe deprecation of some older kubctl versions as supported.
kiranchavala
left a comment
There was a problem hiding this comment.
LGTM
Tested manually
- Generate a cks iso based on 1.31.1 version of kubernetes
./create-kubernetes-binaries-iso.sh ./ 1.31.1 1.3.0 1.27.0 https://raw.githubusercontent.com/weaveworks/weave/master/prog/weave-kube/weave-daemonset-k8s-1.11.yaml https://raw.githubusercontent.com/kubernetes/dashboard/v2.7.0/aio/deploy/recommended.yaml setup-1.31.1
-
Register it as a kubernetes iso and deploy a cks cluster
-
Enable autoscaling on the cluster ( specify min and max size)
-
Deploy a test deployment say nginx on the cks cluster
-
Trigger cks autoscaling by executing the command
kubectl scale --replicas=250 deployment/nginx-deployment -
The node count of the cks cluster will increase
kubectl get nodes -
Trigger cks autodescaling
kubectl scale --replicas=25 deployment/nginx-deployment
Check the logs
Before fix
2025-01-28 11:10:51,991 INFO [c.c.k.c.a.KubernetesClusterScaleWorker] (API-Job-Executor-50:[ctx-ddc3650b, job-74, ctx-e8087ea8]) (logid:a9c772dc) Scaling Kubernetes cluster : test
2025-01-28 11:10:51,999 INFO [c.c.k.c.a.KubernetesClusterScaleWorker] (API-Job-Executor-50:[ctx-ddc3650b, job-74, ctx-e8087ea8]) (logid:a9c772dc) Removing vm : test-node-194ac81c54e from cluster test
2025-01-28 11:10:52,719 ERROR [c.c.u.s.SshHelper] (API-Job-Executor-50:[ctx-ddc3650b, job-74, ctx-e8087ea8]) (logid:a9c772dc) SSH execution of command sudo /opt/bin/kubectl drain test-node-194ac81c54e --ignore-daemonsets --delete-local-data has an error status code in return. Result output: error: unknown flag: --delete-local-data
See 'kubectl drain --help' for usage.
After fix
No error message observed , and cks cluster scaled downed successfully
|
thanks @kiranchavala for the testing |
Description
This PR fixes #10233
--delete-local-datais dropped since Kubernetes 1.31https://github.com/kubernetes/kubernetes/blob/a78983906f4cf6f94029a3e6cf3171ef42b395d7/CHANGELOG/CHANGELOG-1.31.md?plain=1#L1338
--delete-emptydir-datahas been supported since 2020kubernetes/kubernetes@625e47a
Types of changes
Feature/Enhancement Scale or Bug Severity
Feature/Enhancement Scale
Bug Severity
Screenshots (if appropriate):
How Has This Been Tested?
How did you try to break this feature and the system with this change?