You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@paimon.apache.org by "A-little-bit-of-data (via GitHub)" <gi...@apache.org> on 2025/09/13 10:02:19 UTC

[I] [Bug] Sink multi-parallelism error hive lock [paimon]

A-little-bit-of-data opened a new issue, #6251:
URL: https://github.com/apache/paimon/issues/6251

   ### Search before asking
   
   - [x] I searched in the [issues](https://github.com/apache/paimon/issues) and found nothing similar.
   
   
   ### Paimon version
   
   1.1.1
   
   ### Compute Engine
   
   flink 1.20.1
   
   ### Minimal reproduce step
   
   CREATE TABLE t_changelog_input (
         age BIGINT,
         money BIGINT,
         name STRING,
         PRIMARY KEY (name) NOT ENFORCED
   )WITH (
   "bucket" = "4",
   	'file.compression' = 'snappy',
       'merge-engine' = 'deduplicate',
       'changelog-producer' = 'input',
       "sink.parallelism" = "4"
   );
   
   The table is created as above. When a large amount of data is written or a large amount of data is changed, the following error will appear. However, when sink.parallelism is 1, it will not appear. My data is stored on S3 and uses hivemetastore 3.1.2.Is there something wrong with my usage? Is there any way to increase the write parallelism?
   
   <img width="1391" height="237" alt="Image" src="https://github.com/user-attachments/assets/da2aea69-64a3-4b7e-9d9c-2c7262897498" />
   
   ### What doesn't meet your expectations?
   
   I hope that even when the data is stored on S3, which does not have atomic storage, it can still support multiple concurrent writes.
   
   ### Anything else?
   
   _No response_
   
   ### Are you willing to submit a PR?
   
   - [ ] I'm willing to submit a PR!


-- 
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: issues-unsubscribe@paimon.apache.org.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [I] [Bug] Sink multi-parallelism error hive lock [paimon]

Posted by "SagarJadhav30 (via GitHub)" <gi...@apache.org>.
SagarJadhav30 commented on issue #6251:
URL: https://github.com/apache/paimon/issues/6251#issuecomment-3707768332

   i understand the Bug i can solve 
   Use commit.parallelism = 1
   WITH (
     'bucket' = '4',
     'file.compression' = 'snappy',
     'merge-engine' = 'deduplicate',
     'changelog-producer' = 'input',
     'sink.parallelism' = '4',
     'commit.parallelism' = '1'
   );
   Increase Hive lock timeout this is the mitigation, not fix
   hive.metastore.client.socket.timeout=600
   hive.metastore.lock.numretries=10
   hive.metastore.lock.retry.sleep.between.retries=10
   
   For production on S3 + Hive:
   sink.parallelism > 1
   commit.parallelism = 1  Or use FilesystemCatalog


-- 
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: issues-unsubscribe@paimon.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [I] [Bug] Sink multi-parallelism error hive lock [paimon]

Posted by "A-little-bit-of-data (via GitHub)" <gi...@apache.org>.
A-little-bit-of-data commented on issue #6251:
URL: https://github.com/apache/paimon/issues/6251#issuecomment-4426693170

   > i understand the Bug i can solve Use commit.parallelism = 1 WITH ( 'bucket' = '4', 'file.compression' = 'snappy', 'merge-engine' = 'deduplicate', 'changelog-producer' = 'input', 'sink.parallelism' = '4', 'commit.parallelism' = '1' ); Increase Hive lock timeout this is the mitigation, not fix hive.metastore.client.socket.timeout=600 hive.metastore.lock.numretries=10 hive.metastore.lock.retry.sleep.between.retries=10
   > 
   > For production on S3 + Hive: sink.parallelism > 1 commit.parallelism = 1 Or use FilesystemCatalog
   
   I searched the documentation and source code for version 1.4.1 for the `commit.parallelism` configuration, but couldn't find its source, although the documentation does mention commit-related aspects. I also tried FilesystemCatalog, but it had other issues, and given my preference for unified metadata management, I abandoned using FilesystemCatalog.
   <img width="2790" height="1227" alt="Image" src="https://github.com/user-attachments/assets/12293c0a-eeb7-4399-9440-20cb8fdc2adf" />


-- 
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: issues-unsubscribe@paimon.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org