Skip to content
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

Block builder: update alerts and add critical alerts #10270

Merged
merged 5 commits into from
Jan 24, 2025
Merged

Conversation

codesome
Copy link
Member

@codesome codesome commented Dec 17, 2024

What this PR does

Updated some warnings and add critical alerts to block builders.

  • BlockBuilderNoCycleProcessing is a little sensitive. Meaning if a consumption ran a little longer than 5m compared to the earlier one (for duration is 5m), this alert will fire. Made the for period to be 20 mins
  • BlockBuilderLagging current warning alert fires when the lag is over 4M for 75mins (basically lag went above 4M, but it stayed above 4M in the next cycle as well). We have a critical alert with for duration of 2h20m. Meaning once it went above 4M in a cycle, it remained about 4M in the next two cycles as well. If the cycle run was not successful, then BlockBuilderNoCycleProcessing will already fire much the lagging alert will fire. BlockBuilderLagging will catch other issues that is not just "not able to consume", like block builders are not scaled enough for example.
  • If BlockBuilderCompactAndUploadFailed fired, it should be a critical alert immediately. Removed the for period.

Checklist

  • Tests updated.
  • Documentation added.
  • CHANGELOG.md updated - the order of entries should be [CHANGE], [FEATURE], [ENHANCEMENT], [BUGFIX].
  • about-versioning.md updated with experimental features.

Copy link
Contributor

@narqo narqo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure about having two of each kind of alerts.

E.g. the "no cycles" two: to have only 10 minutes gap between a warning and a critical alert is a fairly small window for a person, who'll be paged, to do any meaningful investigations.

I suggest we only keep one but make sure the for is long enough to be critical.

@codesome
Copy link
Member Author

Fair point! We can probably just keep a critical alert for 20 mins and remove the warning for 10 mins.

Signed-off-by: Ganesh Vernekar <[email protected]>
Signed-off-by: Ganesh Vernekar <[email protected]>
@codesome codesome marked this pull request as ready for review January 23, 2025 21:35
@codesome codesome requested a review from a team as a code owner January 23, 2025 21:35
@codesome codesome requested a review from seizethedave January 23, 2025 21:35
@codesome
Copy link
Member Author

I updated the PR and the description.

I removed the BlockBuilderNoCycleProcessing warning and made for period to be 20m for the critical alert.

Remove for period for BlockBuilderCompactAndUploadFailed to fire immediately. If we have for period of 5m, it will never fire because in the query we lookback only 1 min.

Copy link
Contributor

@seizethedave seizethedave left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LG2M.

Signed-off-by: Ganesh Vernekar <[email protected]>
@codesome codesome merged commit 0c923bb into main Jan 24, 2025
30 checks passed
@codesome codesome deleted the codesome/bb-alerts branch January 24, 2025 18:04
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants