swamirishi opened a new pull request, #7567:
URL: https://github.com/apache/ozone/pull/7567

   ## What changes were proposed in this pull request?
   Snapshot diff DAG traversal should not skip node based on prefix presence. 
   
   Consider the case 
   
   Snap1 has 1, 2, 3 files
   
   Say 1,4 compacted to => 5
   
   And 5, 6 => 7
   
   Snap 2 has 7, 2, 3 file.
   
   Now diff b/w Snap1 & Snap2 would be in 4 & 6 sst files.
   
   Consider 1 had +k & 4 had -k
   
   Now 5 would have neither of these files.
   
   But as per this condition:
   
https://github.com/apache/ozone/blob/b11b80707d4e6721c06966936ad87063ec107da6/hadoop-hdds/rocksdb-checkpoint-differ/src/main/java/org/apache/ozone/rocksdiff/RocksDBCheckpointDiffer.java#L1009-L1016
 traversal on would skip node 5 thus traversal would never reach 1,4. We should 
not have this optmization check. This would end up not using the compaction 
even though it is not there.
   
   ## What is the link to the Apache JIRA
   https://issues.apache.org/jira/browse/HDDS-11908
   
   ## How was this patch tested?
   Unit tests
   


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to