chore(block-builder): update block builder to use kafka clients directly #15433
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
ashwanthgoli
changed the title
benclive
reviewed
Dec 17, 2024
| // No more records available | ||
| break | ||
| continue |
There was a problem hiding this comment.
Why is this continue instead of break? And do we want to reset the backoff after a successful poll first?
There was a problem hiding this comment.
this is unlikely, but if we are already caught-up to the latest produced offset this should allow a few retries before failing. I don't have a strong opinion about this change
| } | ||
|
|
||
| // Reset backoff on successful poll | ||
| boff.Reset() | ||
|
|
||
| converted := make([]AppendInput, 0, len(records)) | ||
| for _, record := range records { | ||
| converted := make([]AppendInput, 0, fs.NumRecords()) |
There was a problem hiding this comment.
Comparing to the existing partition reader, would it be useful to add some metrics on number of records processed etc.?
There was a problem hiding this comment.
definitely! i'll take that in a follow-up :)
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Outdated suggestions cannot be applied.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 / why we need it:
Currently block builders create ephemeral partition reader for processing a new job, but this fails to work as expected because of duplicate kprom metric registration that is triggered on new client creation. We could get around this by wrapping the registerer with
partition_idandworker_idlabels and carefully deregistering them when the job processing is complete.But at this point of time, I prefer directly using the kafka clients (one per each block builder worker) which offers better control over addition/removal of partitions to consume. We can look into partition reader refactor down the line if it's neccessary.
This pr updates block builder to use direct kafka clients, one client per worker. We add partition consume offset when we start a job and remove it at the end of the job to allow reusing the same client.
Which issue(s) this PR fixes:
Fixes #
Special notes for your reviewer:
Checklist
CONTRIBUTING.mdguide (required)featPRs are unlikely to be accepted unless a case can be made for the feature actually being a bug fix to existing behavior.docs/sources/setup/upgrade/_index.mddeprecated-config.yamlanddeleted-config.yamlfiles respectively in thetools/deprecated-config-checkerdirectory. Example PR