F64116045 commented on code in PR #11199:
URL: https://github.com/apache/ozone/pull/11199#discussion_r3971182765
##########
hadoop-ozone/recon/src/main/java/org/apache/hadoop/ozone/recon/fsck/ReconReplicationManager.java:
##########
@@ -313,8 +279,11 @@ public synchronized void processAll() {
// readOnly=true ensures no commands are generated
processContainer(container, replicas, pendingOps, nullQueue, report,
true);
- // ADDITIONAL CHECK: Detect REPLICA_MISMATCH (Recon-specific, not in
SCM)
- if (hasDataChecksumMismatch(replicas)) {
+ // Persist checksum mismatches in Recon's REPLICA_MISMATCH state.
+ if (container.getState() == CLOSED &&
+ container.getReplicationType() == RATIS &&
+ hasMismatch(replicas, ContainerReplica::getSequenceId,
Review Comment:
Thanks.
I removed the debounce after [the
review](https://github.com/apache/ozone/pull/11199#issuecomment-5583750640), so
Recon no longer needs its own two scan state.
For RATIS, SCM and Recon now use the same helper and report a mismatch on
the first scan.
I also merged your EC fix from #11215. Resolved the conflict by keeping the
same-index grouping and using the same BCSID and checksum comparison within
each group.
--
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]
