ts: fix tenant metrics scan in tsdb#99511
Merged
craig[bot] merged 1 commit intocockroachdb:masterfrom Mar 25, 2023
Merged
Conversation
In cockroachdb#98077, we introduced tenant scoped metrics into tsdb. The TenantIDs were part of the source field used for NodeIDs previously. Since the intended purpose for the system tenant was to be able to aggregate across all tenant metrics, we introduced a new scan over all the sources for the given Node source. This introduced a bug: When we prefix match the source to scan over its keyspace, it includes all sources that match that prefix. For a 10 node cluster with 3 tenants, this meant fetching timeseries for Node 1 would include: ``` 1 1-2 1-3 10 10-2 10-3 ``` This would then aggregate data across Node 1 and Node 10. If there were no secondary tenants, it would include the soruces: ``` 1 10 ``` This fix separates the get of the initial system tenant metric and scan across all tenants (with a tenant prefix) to only scan the desired source. Now in the same 10 Node cluster with 3 tenants, it will include: ``` 1 1-2 1-3 ``` The scan is now from the beginning of `1-` to its end, ensuring only Node 1's tenants are included and not Node 10 or 100 or 1000. Fixes: cockroachdb#99486 Release note: None
Member
|
This change is |
dhartunian
approved these changes
Mar 24, 2023
Collaborator
dhartunian
left a comment
There was a problem hiding this comment.
Reviewed 3 of 3 files at r1, all commit messages.
Reviewable status:complete! 1 of 0 LGTMs obtained (waiting on @aadityasondhi)
Contributor
Author
|
TFTR! bors r+ |
Contributor
|
Build failed (retrying...): |
Contributor
|
Build failed (retrying...): |
Contributor
|
Build succeeded: |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
In #98077, we introduced tenant scoped metrics into tsdb. These metric were part of the source field used for NodeIDs previously. Since the intended purpose for the system tenant was to be able to aggregate across all tenant metrics, we introduced a new scan over the entire keyspace for the given Node source. This introduced a bug:
When we prefix match the source to scan over its keyspace, it includes all sources that match that prefix. For a 10 node cluster with 3 tenants, this meant fetching timeseries for Node 1 would include:
This would then aggregate data across Node 1 and Node 10. If there were no secondary tenants, it would include the sources:
This fix separates the get of the initial system tenant metric and scan across all tenants (with a tenant prefix) to only scan the desired source. Now in the same 10 Node cluster with 3 tenants, it will include:
The scan is now from the beginning of
1-to its end, ensuring only Node 1's tenants are included and not Node 10 or 100 or 1000.Fixes: #99486
Release note: None