oneby-wang opened a new pull request, #25913:
URL: https://github.com/apache/pulsar/pull/25913
### Motivation
The async ZooKeeper addWatch wrapper in PulsarZooKeeperClient defined a
retry callback but delegated the operation using the original callback and
context. If addWatch returned a recoverable error, the caller could observe the
transient failure directly instead of retrying. The wrapper's completion branch
also recursively invoked itself, which would be incorrect if the wrapper
callback were used.
This affects the SessionReestablished path in ZKMetadataStore, where Pulsar
recreates the persistent recursive watch before notifying session listeners.
### Modifications
- Route async addWatch calls through the retry wrapper callback and
ZooWorker context.
- Complete the caller-provided callback with the final result after retry
handling.
- Fix the addWatch retry operation description.
- Add a regression test that verifies a recoverable CONNECTIONLOSS is
retried and the caller receives the final OK result.
### Verifying this change
- [x] Make sure that the change passes the CI checks.
This change added tests and can be verified as follows:
- Added MetadataStoreTest.testAsyncAddWatchRetriesWithWrapperCallback to
cover addWatch retry callback behavior.
### Does this pull request potentially affect one of the following parts:
- [ ] Dependencies (add or upgrade a dependency)
- [ ] The public API
- [ ] The schema
- [ ] The default values of configurations
- [ ] The threading model
- [ ] The binary protocol
- [ ] The REST endpoints
- [ ] The admin CLI options
- [ ] The metrics
- [ ] Anything that affects deployment
--
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]
