[server] Add the leader of bucket to metadata cache #386
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
luoyuxia
force-pushed
the
add-metadata-cache-final
branch
3 times, most recently
from
20fbae3 to
c6aca6e
Compare
luoyuxia
force-pushed
the
add-metadata-cache-final
branch
4 times, most recently
from
96ba9ea to
8432333
Compare
luoyuxia
force-pushed
the
add-metadata-cache-final
branch
from
8432333 to
07bfb61
Compare
luoyuxia
changed the title
|
@wuchong Add the leader of bucket to metadata cache to avoid get table metadata cost too much time.. Pr ready |
wuchong
reviewed
Mar 3, 2025
| @@ -130,7 +129,7 @@ void beforeEach() throws Exception { | |||
| } catch (CatalogException e) { | |||
| // the auto partitioned manager may create the db zk node | |||
| // in an another thread, so if exception is NodeExistsException, just ignore | |||
| if (!ExceptionUtils.findThrowable(e, KeeperException.NodeExistsException.class) | |||
| if (!ExceptionUtils.findThrowableWithMessage(e, "KeeperException$NodeExistsException") | |||
There was a problem hiding this comment.
Could you extract this into a separate hotfix pull request? This seems quite unstable in these days. https://github.com/alibaba/fluss/actions/runs/13626469089/job/38084662575
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Outdated suggestions cannot be applied.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.
Purpose
Linked issue: close #386
Currnetly, to know the leader of a TableBucket, we always need to get from zk..When the buckets increase, it will cost too many time which will cause the job fail to startup just like the issue describes
In this pr, we only add the leader of a TableBucket to metadata for simplicity, maybe in the future, we can add more to metadata cache.
ServerMetadataCache, cache the mapping fromtableBuckettoleaderUpdateMetadataRequestin rpc proto to addPbTableBucketMetadata,deleted_table_id, `deleted_partition_id.UpdateMetadataRequestwith the updated leader inPbTableBucketMetadatato tablet servers, then tablet servers can update their own cache.UpdateMetadataRequestwithdeleted_table_id/deleted_partition_idto tablet servers, then tablet servers can remove these tables from their own cacheTests
ServerMetadataCacheImplTestto verify theServerMetadataCacheworks as expected.MetadataUpdateITCaseto verify when table created, bucket leader changed, server down/up, the cache is as expected.API and Format
N/A
Documentation
N/A