Fix scaling dashboard to work on multi-zone ingesters #365
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Marco Pracucci <[email protected]>
tomwilkie
reviewed
Jul 28, 2021
| label_replace(kube_statefulset_replicas, "deployment", "$1", "statefulset", "(.*)") | ||
| label_replace( | ||
| kube_deployment_spec_replicas, | ||
| "deployment", "$1", "deployment", "(.*?)(?:-zone-[a-z])?" |
There was a problem hiding this comment.
Why is the first question mark necessary? Shouldn't it be:
Suggested change
| "deployment", "$1", "deployment", "(.*?)(?:-zone-[a-z])?" | |
| "deployment", "$1", "deployment", "(.*)(?:-zone-[a-z])?" |
There was a problem hiding this comment.
The first question mark is to make it non-greedy. Since the (?:-zone-[a-z])? is optional (ending ?), if the first (.*) is greedy then it always match everything and never removes the zone. Adding (.*?) we make the first .* non greedy.
There was a problem hiding this comment.
TIL! Can you add a comment to this effect please?
aknuds1
approved these changes
Jul 28, 2021
tomwilkie
reviewed
Jul 28, 2021
Comment on lines
83
to
86
| label_replace( | ||
| label_replace(kube_statefulset_replicas, "deployment", "$1", "statefulset", "(.*)"), | ||
| "deployment", "$1", "deployment", "(.*?)(?:-zone-[a-z])?" | ||
| ) |
There was a problem hiding this comment.
The inner label replace is just moving the statefuleset label to the deployment label, so could be done with this I believe:
Suggested change
| label_replace( | |
| label_replace(kube_statefulset_replicas, "deployment", "$1", "statefulset", "(.*)"), | |
| "deployment", "$1", "deployment", "(.*?)(?:-zone-[a-z])?" | |
| ) | |
| label_replace(kube_statefulset_replicas, "deployment", "$1", "statefulset", "(.*?)(?:-zone-[a-z])?"), |
There was a problem hiding this comment.
That's right. I've applied the suggested change and manually tested it.
tomwilkie
approved these changes
Jul 28, 2021
…ng rule Signed-off-by: Marco Pracucci <[email protected]>
Signed-off-by: Marco Pracucci <[email protected]>
1 task
simonswine
pushed a commit
to grafana/mimir
that referenced
this pull request
Oct 18, 2021
…g-dashboard-for-multi-zone-deployments Fix scaling dashboard to work on multi-zone ingesters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What this PR does:
We have some clusters running Cortex ingesters in multi-zone. Each zone is a StatefulSet whose name matches this pattern
ingester-zone-[a-z], so their pod names are likeingester-zone-a-0oringester-zone-b-1. All dashboards work correctly except for the scaling dashboard and this PR proposes a fix for that.Reason why the scaling dashboard doesn't work is that some recording rule computes the expected scaling value summing up all ingesters (eg.
cluster_namespace_deployment_reason:required_replicas:count) while the actual usage metrics are grouped by deployment/statefulset name, so they're splitted by zone.I think the desired behaviour is summing up all ingesters, regardless their zone, so in this PR I'm proposing to remove the
-zone-[a-z]suffix (if found) when we compute thedeploymentname.A couple of notes:
label_replace()due to conflicts with regex greedinessWhich issue(s) this PR fixes:
N/A
Checklist
CHANGELOG.mdupdated - the order of entries should be[CHANGE],[FEATURE],[ENHANCEMENT],[BUGFIX]