Zakelly opened a new pull request #12312:
URL: https://github.com/apache/flink/pull/12312
# What is the purpose of the change
Since local recovery for RocksDB can greatly help the recovery, and no
overhead is introduced when incremental checkpoints are used. It should be
activated by default.
So as not to add understanding costs for this setting to users, I simply
activate the local recovery for all state backends. For FsStateBackend, it may
helps the recovery when using distributed filesystem.
## Brief change log
- set the default value of CheckpointingOptions.LOCAL_RECOVERY into *true*.
- change the relative docs.
- change unit test *TaskExecutorLocalStateStoresManagerTest*.
## Verifying this change
This change is already covered by existing tests, such as the
*TaskExecutorLocalStateStoresManagerTest*.
I also try the WordCount streaming job with default settings and make sure
the local recovery is activated by checking the local state directory.
## 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)`: (no)
- The serializers: (no)
- The runtime per-record code paths (performance sensitive): (no)
- Anything that affects deployment or recovery: JobManager (and its
components), Checkpointing, Kubernetes/Yarn/Mesos, ZooKeeper: (yes)
- The S3 file system connector: (no)
## Documentation
- Does this pull request introduce a new feature? (no)
- If yes, how is the feature documented? (not applicable)
CC @StephanEwen @carp84
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
