dsmiley commented on code in PR #3702:
URL: https://github.com/apache/solr/pull/3702#discussion_r2380641962
##########
solr/CHANGES.txt:
##########
@@ -72,6 +72,8 @@ Optimizations
* SOLR-17568: The CLI bin/solr export tool now contacts the appropriate nodes
directly for data instead of proxying through one.
(David Smiley)
+* SOLR-17929: Remove obsolete overseer internal work queue. This queue was
read-only since Solr 8. (Pierre Salagnac)
Review Comment:
IMO this should be an "Other" category. It's a deprecation cleanup. Nobody
will really experience this as an optimization.
##########
solr/core/src/java/org/apache/solr/cloud/Overseer.java:
##########
@@ -268,50 +253,50 @@ public void run() {
new ZkStateWriter(reader, stats,
minStateByteLenForCompression, compressor);
refreshClusterState = false;
- // if there were any errors while processing
- // the state queue, items would have been left in the
- // work queue so let's process those first
- byte[] data = fallbackQueue.peek();
+ // if there were any errors while processing the queue, items
would have been left in
+ // the queue with a fallback size greater than 0, so let's
process those first
+ byte[] data = stateUpdateQueue.peek();
while (fallbackQueueSize > 0 && data != null) {
Review Comment:
If the queue being removed is always empty, then this loop here is dead
code, and thus there's really no such thing as a "fallback" to even process.
That's my reading of the code as it is. So I'm confused what you've been
changing here.
--
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]
