ksonnet: Reduce querier parallelism to a more sane default value and remove the default setting for storage_backend #2452
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
…nts explaining how to set it. Remove the default setting for storage_backend which was very misleading for anyone not using the defaults.
Codecov Report
@@ Coverage Diff @@
## master #2452 +/- ##
==========================================
- Coverage 62.79% 62.75% -0.04%
==========================================
Files 162 162
Lines 13963 13963
==========================================
- Hits 8768 8763 -5
- Misses 4513 4516 +3
- Partials 682 684 +2
|
owen-d
approved these changes
Jul 30, 2020
cyriltovena
pushed a commit
to cyriltovena/loki
that referenced
this pull request
Jun 11, 2021
* Refactor: export NextDelay() so we can call it from other packages Signed-off-by: Bryan Boreham <bryan@weave.works> * Refactor: use newer AWS API for paginated queries This is less code, and more robust when retrying requests. We don't need an indirection on the request object for testing now. Signed-off-by: Bryan Boreham <bryan@weave.works> * Don't need DynamoDB request wrapper to do so much now Now we are calling QueryPagesWithContext directly we don't need the paging interface and we never re-use request objects. Signed-off-by: Bryan Boreham <bryan@weave.works> * Stop AWS Retryer if context is cancelled Ingester.flushUserSeries() puts a timeout on the context, so don't retry for longer than that. Signed-off-by: Bryan Boreham <bryan@weave.works>
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Outdated suggestions cannot be applied.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.
The default of 32 for querier parallelism would only make sense for machines with 32 cores which is unlikely.
Setting a new default of 4 which is probably a bit low but safer, including comments on what the appropriate value should be.
Unrelated to this change but while I was here,
I removed the default setting for storage_backend which was very misleading for anyone not using the defaults and instead made it an error not to define this directly.
Anyone not using the defaults would previously get a confusing error about not defining bigtable names, now they should get a more helpful error on how to set the
storage_backendvariable.