Skip to content
This repository has been archived by the owner on Jan 11, 2023. It is now read-only.

heapster issues in v0.12.0 #2083

Closed
jackfrancis opened this issue Jan 18, 2018 · 2 comments · Fixed by #2091
Closed

heapster issues in v0.12.0 #2083

jackfrancis opened this issue Jan 18, 2018 · 2 comments · Fixed by #2091

Comments

@jackfrancis
Copy link
Member

Is this an ISSUE or FEATURE REQUEST? (choose one): ISSUE


What version of acs-engine?: v0.12.0


@jackfrancis
Copy link
Member Author

Example heapster pod logs from a newly deployed cluster:

$ kubectl logs heapster-84bb879958-kmfsq heapster -n kube-system
I0118 15:39:53.875470       1 heapster.go:78] /heapster --source=kubernetes.summary_api:""
I0118 15:39:53.875504       1 heapster.go:79] Heapster version v1.5.0
I0118 15:39:53.875672       1 configs.go:61] Using Kubernetes client with master "https://10.0.0.1:443" and version v1
I0118 15:39:53.875687       1 configs.go:62] Using kubelet port 10255
I0118 15:39:53.876332       1 heapster.go:202] Starting with Metric Sink
I0118 15:39:54.077461       1 heapster.go:112] Starting heapster on port 8082
E0118 15:40:05.076012       1 manager.go:101] Error in scraping containers from kubelet_summary:10.240.0.66:10255: Get http://10.240.0.66:10255/stats/summary/: dial tcp 10.240.0.66:10255: getsockopt: connection refused
E0118 15:40:05.076092       1 manager.go:101] Error in scraping containers from kubelet_summary:10.240.0.5:10255: Get http://10.240.0.5:10255/stats/summary/: dial tcp 10.240.0.5:10255: getsockopt: connection refused
E0118 15:40:05.077037       1 manager.go:101] Error in scraping containers from kubelet_summary:10.240.0.4:10255: Get http://10.240.0.4:10255/stats/summary/: dial tcp 10.240.0.4:10255: getsockopt: connection refused
E0118 15:40:05.078374       1 manager.go:101] Error in scraping containers from kubelet_summary:10.240.0.97:10255: Get http://10.240.0.97:10255/stats/summary/: dial tcp 10.240.0.97:10255: getsockopt: connection refused
W0118 15:40:25.000433       1 manager.go:152] Failed to get all responses in time (got 0/4)

@jackfrancis
Copy link
Member Author

#2091 fixed this bug, which was subsequently included in the v0.12.1 patch release here:

To manually fix this if you've built a cluster from v0.12.0 and don't want to re-build, update kubelet config on each node to remove the --read-only-port=0 runtime config setting. This is set in /etc/default/kubelet. sudo systemctl restart kubelet on each node host to restart kubelet with the modified configuration.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

Successfully merging a pull request may close this issue.

1 participant