smengcl commented on code in PR #12093:
URL: https://github.com/apache/ozone/pull/12093#discussion_r4638998325


##########
hadoop-ozone/recon/src/main/java/org/apache/hadoop/ozone/recon/tasks/ReconTaskControllerImpl.java:
##########
@@ -553,6 +553,16 @@ private void processOMUpdateBatch(OMUpdateEventBatch 
events) {
       return;
     }
 
+    // A batch that failed both attempts only raises tasksFailed. The next 
sync drains this buffer and
+    // rebuilds from a checkpoint that already contains every queued batch. 
Applying the batches still
+    // waiting would move the task cursor up to the OmDeltaRequest cursor, so 
a restart before that
+    // rebuild would find nothing to redo and the failed batch would be lost.
+    if (hasTasksFailed()) {
+      LOG.warn("Skipping buffered OM update batch ending at sequence {} 
because a task has failed and " +

Review Comment:
   nit: since this runs once per buffered batch, a backlog in the buffer (tasks 
slower than the OM sync) would print one WARN per skipped batch until the next 
sync drains it. The failure is already logged at WARN by the retry path and by 
the sync thread when it queues the rebuild, so maybe DEBUG is enough here?
   
   ```suggestion
         LOG.debug("Skipping buffered OM update batch ending at sequence {} 
because a task has failed and " +
   ```



-- 
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