[GitHub] [spark] HeartSaVioR commented on issue #23840: [WIP][DISCUSSION_NEEDED][SPARK-24295][SS] Add option to retain only last batch in file stream sink metadata

[GitBox <gi...@apache.org>]

HeartSaVioR commented on issue #23840: [WIP][DISCUSSION_NEEDED][SPARK-24295][SS] Add option to retain only last batch in file stream sink metadata
URL: https://github.com/apache/spark/pull/23840#issuecomment-469479092
 
 
   > if we ignore sink metadata then we may face correct files/duplicate file issue as mentioned by @tdas in this mail chain
   
   Yes. Actually there's no way to achieve both. Looks like the only way is let sink checks and purges file entries when files are removed outside of query via some retention policy, but that will only work when query is running (it might be OK since metadata is growing only when query is running).
   
   > will CompactibleFileStreamLog.compact fail because there will not be any batchid to compact?
   
   I guess so, so the proposed patch is not a complete solution as of now. We may need to focus alternatives I've suggested as well or raise a new idea.

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org