Zhuoxi2000 opened a new pull request, #29308:
URL: https://github.com/apache/flink/pull/29308
<!--
*Thank you very much for contributing to Apache Flink - we are happy that
you want to help us improve Flink. To help the community review your
contribution in the best possible way, please go through the checklist below,
which will get the contribution into a shape in which it can be best reviewed.*
*Please understand that we do not do this to make contributions to Flink a
hassle. In order to uphold a high standard of quality for code contributions,
while at the same time managing a large number of contributions, we need
contributors to prepare the contributions well, and give reviewers enough
contextual information for the review. Please also understand that
contributions that do not follow this guide will take longer to review and thus
typically be picked up with lower priority by the community.*
## Contribution Checklist
- Make sure that the pull request corresponds to a [JIRA
issue](https://issues.apache.org/jira/projects/FLINK/issues). Exceptions are
made for typos in JavaDoc or documentation files, which need no JIRA issue.
- Name the pull request in the form "[FLINK-XXXX] [component] Title of the
pull request", where *FLINK-XXXX* should be replaced by the actual issue
number. Skip *component* if you are unsure about which is the best component.
Typo fixes that have no associated JIRA issue should be named following
this pattern: `[hotfix] [docs] Fix typo in event time introduction` or
`[hotfix] [javadocs] Expand JavaDoc for PuncuatedWatermarkGenerator`.
- Fill out the template below to describe the changes contributed by the
pull request. That will give reviewers the context they need to do the review.
- Make sure that the change passes the automated tests, i.e., `mvn clean
verify` passes. You can set up Azure Pipelines CI to do that following [this
guide](https://cwiki.apache.org/confluence/display/FLINK/Azure+Pipelines#AzurePipelines-Tutorial:SettingupAzurePipelinesforaforkoftheFlinkrepository).
- Each pull request should address only one issue, not mix up code from
multiple issues.
- Each commit in the pull request has a meaningful commit message
(including the JIRA id)
- Once all items of the checklist are addressed, remove the above text and
this checklist, leaving only the filled out template below.
**(The sections below can be removed for hotfixes of typos)**
-->
## What is the purpose of the change
`SplitFetcherManager` calls the split-reader supplier on the thread that
creates the `SplitFetcher`, which is the source reader's task thread, not the
fetcher thread that then uses the reader. Clients that a `SplitReader` builds
in its constructor therefore capture the wrong thread. For Kafka this meant the
`KafkaConsumer` was created on a thread it never ran on, which FLINK-36434
worked around inside `KafkaPartitionSplitReader` by creating the consumer
lazily
([flink-connector-kafka#290](https://github.com/apache/flink-connector-kafka/pull/290)).
This pull request moves the supplier call into the fetcher itself: a
`SplitFetcher` creates its reader on its own thread when it starts, so no
connector needs its own laziness workaround.
## Brief change log
- `SplitFetcherManager#createSplitFetcher` passes the supplier to the
`SplitFetcher` instead of calling it.
- `SplitFetcher#run` creates the reader first thing on the fetcher thread.
A supplier failure goes through the fetcher's error handler, like a failing
`fetch()`.
- `SplitFetcher#getSplitReader` keeps working before the fetcher starts:
an early call creates the reader on the calling thread, and the reader is still
created only once. Connectors that reach for the reader before starting the
fetcher, such as the Kafka offset-commit path, behave as before; nothing inside
apache/flink itself calls it.
- The internal fetcher tasks (`AddSplitsTask`, `RemoveSplitsTask`,
`PauseOrResumeSplitsTask`, `FetchTask`) resolve the reader when they run
instead of when they are enqueued.
- A fetcher that exits before its reader was created skips closing it.
Behavior change for connector authors: the supplier no longer runs
synchronously inside `addSplits`, so a supplier exception surfaces through
`SplitFetcherManager#checkErrors` on the next poll instead of being thrown from
`addSplits`, and the supplier runs on the fetcher thread (which inherits the
task thread's context class loader, as before, because the pool creates its
threads from the task thread).
## Verifying this change
Please make sure both new and modified tests in this PR follow [the
conventions for tests defined in our code quality
guide](https://flink.apache.org/how-to-contribute/code-style-and-quality-common/#7-testing).
This change added tests and can be verified as follows:
- `SplitFetcherManagerTest#testSplitReaderIsCreatedOnFetcherThread`: the
supplier runs on the fetcher thread, not on the thread adding the split.
-
`SplitFetcherManagerTest#testSplitReaderCreationFailureIsReportedThroughCheckErrors`:
adding a split does not throw when the supplier fails; `checkErrors` reports
the failure, and the manager still closes.
- `SplitFetcherTest#testSplitReaderRequestedBeforeStartIsCreatedOnce`:
`getSplitReader()` before the fetcher runs creates the reader once, and the
fetcher closes that same reader.
- `SplitFetcherPauseResumeSplitReaderTest`: the reader-count assertions
now run after the fetchers first run, since readers are no longer created while
splits are added.
- The rest of `flink-connector-base` passes unchanged (`mvn verify` on the
module: unit tests, ITs such as `CoordinatedSourceITCase` and
`HybridSourceITCase`, and japicmp).
## Does this pull request potentially affect one of the following parts:
- Dependencies (does it add or upgrade a dependency): no
- The public API, i.e., is any changed class annotated with
`@Public(Evolving)`: yes. `SplitFetcher` and `SplitFetcherManager` are
`@PublicEvolving`; no signature changes, but the supplier now runs on the
fetcher thread, as described above.
- The serializers: no
- The runtime per-record code paths (performance sensitive): no. The fetch
task reads the reader through a volatile field once per `fetch()` call, not per
record.
- Anything that affects deployment or recovery: JobManager (and its
components), Checkpointing, Kubernetes/Yarn, ZooKeeper: no
- The S3 file system connector: no
## Documentation
- Does this pull request introduce a new feature? no
- If yes, how is the feature documented? JavaDocs (the threading of the
supplier is described on `SplitFetcherManager` and
`SplitFetcher#getSplitReader`)
---
##### Was generative AI tooling used to co-author this PR?
<!--
If generative AI tooling has been used in the process of authoring this PR,
please
change the checkbox below to `[X]` and replace the placeholder in the
"Generated-by"
line with the tool name and version. Otherwise remove the "Generated-by"
line.
See the ASF Generative Tooling Guidance for details:
https://www.apache.org/legal/generative-tooling.html
You are responsible for the quality and correctness of every change in this
PR
regardless of the tooling used. Low-effort AI-generated PRs will be closed.
See
AGENTS.md for the full guidance.
-->
- [X] Yes (please specify the tool below)
Generated-by: Claude Code 2.1.259 (Claude Opus 5.5)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
