kubernetes: "kubectl rollout status" returns error message before rollout finish

When we execute kubectl rollout status, we’re receiving this error:

kubectl rollout status deployment/app --namespace=app
Waiting for rollout to finish: 2 out of 6 new replicas have been updated...
Waiting for rollout to finish: 2 out of 6 new replicas have been updated...

error: timed out waiting for the condition

I also tried to put the option -w, but it did’t change anything.

Since we upgraded to Kubernetes 1.5.2 we’re receiving this. In 1.4.6 it works with no problems. I’ve tried to update kubectl and use previous versions too, but seems to be a problem with the cluster, not with kubectl command.

This is a problem for us because Jenkins is using this command to monitor the deployment status, so our builds are crashing in the end.

We’re running the cluster on AWS with Kops provisioning.

About this issue

Original URL
State: closed
Created 7 years ago
Reactions: 34
Comments: 42 (23 by maintainers)

Commits related to this issue

Output extra logs to debug K8s issue See more https://github.com/kubernetes/kubernetes/issues/40224 Signed-off-by: Oleksandr Slynko <oslynko@pivotal.io> — committed to cloudfoundry-incubator/kubo-ci by mordebites 7 years ago
Use helm to deploy istio in tests The normal deploy fails waiting for `kubectl rollout status` due to this bug: kubernetes/kubernetes#40224 Using helm takes 15 minutes extra to run the tests, which ... — committed to cloudfoundry-incubator/kubo-ci by deleted user 6 years ago
cakephp-mysql-persistent: wait before rollout status There is an issue with 'oc rollout status', sometimes it fails with: error: watch closed before Until timeout See https://github.com/kubern... — committed to redhat-cop/agnosticd by fridim 6 years ago
Merge pull request #67817 from tnozicka/fix-rollout-status-wait Automatic merge from submit-queue (batch tested with PRs 67986, 68210, 67817). If you want to cherry-pick this change to another branch... — committed to kubernetes/kubernetes by deleted user 6 years ago

Most upvoted comments

mooperd on Oct 20, 2017

Fix is here https://github.com/kubernetes/kubernetes/pull/50102

/sig apps

tnozicka on Apr 5, 2018

@guineveresaenger there is a PR, https://github.com/kubernetes/kubernetes/pull/67817 linked just above your comment, fixing the issue; in review process, still targeting v1.12

tnozicka on Aug 27, 2018

@kubernetes/sig-cli-bugs @kubernetes/sig-api-machinery-bugs either fix watch.Until or remove

0xmichalis on Apr 19, 2017

Meet the same problem：

# kubectl version
Client Version: version.Info{Major:"1", Minor:"7", GitVersion:"v1.7.9", GitCommit:"19fe91923d584c30bd6db5c5a21e9f0d5f742de8", GitTreeState:"clean", BuildDate:"2017-10-19T17:09:02Z", GoVersion:"go1.8.3", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"7", GitVersion:"v1.7.9", GitCommit:"19fe91923d584c30bd6db5c5a21e9f0d5f742de8", GitTreeState:"clean", BuildDate:"2017-10-19T16:55:06Z", GoVersion:"go1.8.3", Compiler:"gc", Platform:"linux/amd64"}

# kubectl rollout status deploy test
Waiting for rollout to finish: 0 out of 2 new replicas have been updated...
Waiting for rollout to finish: 0 out of 2 new replicas have been updated...
error: watch closed before Until timeout

root cause：

apiVersion: extensions/v1beta1
kind: Deployment
metadata:
  name: test
  namespace: default
spec:
spec:
  paused: true    # Should be false.
  ...

Solution：

Set .spec.paused as false.

.spec.paused is an optional boolean field for pausing and resuming a Deployment. It defaults to false (a Deployment is not paused).

xref: http://kubernetes.kansea.com/docs/user-guide/deployments/#paused

supereagle on May 17, 2018