-
Notifications
You must be signed in to change notification settings - Fork 39.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Restore cAdvisor prometheus metrics to the main port #49079
Restore cAdvisor prometheus metrics to the main port #49079
Conversation
@kubernetes/sig-instrumentation-pr-reviews |
While I do agree with the general idea that these two sets of metrics should be separate, this does not restore the previous state. Is that acceptable or considered a breaking change? |
We don't guarantee backward compatibility for metrics, although we strive. With this change to gather metrics from a new cluster is just a new scrape job, which while annoying... is not the end of the world. An admin who hasn't upgraded to 1.7 could reasonably anticipate and set a scrape job ahead of time. |
But under a new path - `/metrics/cadvisor`. This ensures a secure port still exists for metrics while getting the benefit of separating out container metrics from the kubelet's metrics.
029bdd9
to
f5f9893
Compare
@kubernetes/sig-instrumentation-bugs @piosz |
/retest |
/lgtm |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: brancz, smarterclayton Associated issue: 48483 The full list of commands accepted by this bot can be found here.
Needs approval from an approver in each of these OWNERS Files:
You can indicate your approval by writing |
Removing label |
This is a cherrypick candidate for v1.7.3 -- v1.7.2 is already being built today. cc @wojtek-t |
Automatic merge from submit-queue |
Kubernetes 1.7+ no longer exposes cAdvisor metrics on the Kubelet metrics endpoint. Update the example configuration to scrape cAdvisor in addition to Kubelet. The provided configuration works for 1.7.3+ and commented notes are given for 1.7.2 and earlier versions. Also remove the comment about node (Kubelet) CA not matching the master CA. Since the example no longer connects directly to the nodes, it doesn't matter what CA they're using. References: - kubernetes/kubernetes#48483 - kubernetes/kubernetes#49079
Kubernetes 1.7+ no longer exposes cAdvisor metrics on the Kubelet metrics endpoint. Update the example configuration to scrape cAdvisor in addition to Kubelet. The provided configuration works for 1.7.3+ and commented notes are given for 1.7.2 and earlier versions. Also remove the comment about node (Kubelet) CA not matching the master CA. Since the example no longer connects directly to the nodes, it doesn't matter what CA they're using. References: - kubernetes/kubernetes#48483 - kubernetes/kubernetes#49079
Commit found in the "release-1.7" branch appears to be this PR. Removing the "cherrypick-candidate" label. If this is an error find help to get your PR picked. |
@wojtek-t @luxas @smarterclayton ^ is this being taken care of? |
@brancz I don't understand your question. What do you mean? If you're asking about removing "cherrypick-candiate" label, this is no longer needed, because this has already been cherrypicked to 1.7 rbanch and will be part of 1.7.3 patch release. |
Sorry about that, I misread it. All good. |
But under a new path -
/metrics/cadvisor
. This ensures a secure port still exists for metrics while getting the benefit of separating out container metrics from the kubelet's metrics as recommended in the linked issue.Fixes #48483