[Bug] [chunjun-core] failedConsumedCounter达到阈值后不能停止任务
- Dominant language
- Java
- Stars
- 4.1k
- Forks
- 1.7k
- PR merge metrics
- No merged PRs in 30d
Description
### Search before asking
- [X] I had searched in the [issues](https://github.com/DTStack/chunjun/issues) and found no similar issues.
### What happened

如图当failedConsumedCounter达到阈值时会抛出异常NoRestartException,但是该异常在子线程中抛出,无法影响到主线程,导致即使达到阈值,flink job也无法因为异常结束
### What you expected to happen
failedConsumedCounter达到阈值后 flink任务可以正常结束
### How to reproduce
使用kafka connector 设置dirty-data.max-rows=100,dirty-data.max-collect-failed-rows=5,在源端造10条脏数据,任务无法正常结束
### Anything else
_No response_
### Version
master
### Are you willing to submit PR?
- [ ] Yes I am willing to submit a PR!
### Code of Conduct
- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)
Contributor guide
Assessment
This issue has not been assessed yet.