keith-turner commented on code in PR #5543:
URL: https://github.com/apache/accumulo/pull/5543#discussion_r2084817453
##########
server/compaction-coordinator/src/main/java/org/apache/accumulo/coordinator/CompactionFinalizer.java:
##########
@@ -99,15 +108,18 @@ public void commitCompaction(ExternalCompactionId ecid,
KeyExtent extent, long f
var ecfs =
new ExternalCompactionFinalState(ecid, extent, FinalState.FINISHED,
fileSize, fileEntries);
- LOG.debug("Initiating commit for external compaction: {}", ecfs);
+ LOG.trace("Initiating commit for external compaction: {} {}", ecid, ecfs);
// write metadata entry
- context.getAmple().putExternalCompactionFinalStates(List.of(ecfs));
Review Comment:
Removed the method in 22e5eec. In that change also avoided creating an
intermediate list of objects. Would have needed to add more complex logic to
the SharedBatchWriter to support a batch of mutations and wanted to keep it
simple. So only used the shared batch writer when there was a single mutation,
otherwise create a batch writer to handle multiple mutations. This will make
RPCs from compactors that report a failed compaction use the shared batch
writer (this change makes all the RPCs use the shared batch writer, which is
really good). If dead compaction detection finds a lot of stuff to process,
then it will run in the backgound w/ its own batch writer and not interfere w/
RPC related batch writes.
--
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]
