Add ingester.autoforget-unhealthy-timeout opt-in feature #3919
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
34e05cf to
00ca779
Compare
|
I have tested it, because I was also looking for a feature like this. It works well. Thank you. |
Co-authored-by: Karen Miller <84039272+KMiller-Grafana@users.noreply.github.com>
3 tasks
jmcarp
added a commit
to jmcarp/cortex
that referenced
this pull request
Feb 4, 2022
Implementation adapted from grafana/loki#3919. Related to cortexproject#1521. Signed-off-by: Josh Carp <jm.carp@gmail.com>
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Outdated suggestions cannot be applied.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.
Follow up of #3661, I was too long to push an update and the PR was automatically closed :(
What this PR does / why we need it:
It adds a new parameter that allows to automatically forget unhealthy ingesters after heartbeat timeout.
Which issue(s) this PR fixes:
Partially fixes #3360
Special notes for your reviewer:
Not everyone is using stateful sets. In my case, I'm running on AWS ECS in a stateless configuration. If one of my ingester get killed for whatever reason, this is not a drama because I have RF=3. A new task is spawned.
At the moment the failed ingester will remain in the ring forever and requires to push
forgetbutton on ingester/ring. If not, another task failing will make the cluster unavailable even if 2 out of 3 are healthy because Loki does not use a sloppy quorum.Checklist