[GitHub] [phoenix] stoty commented on a change in pull request #1074: PHOENIX-6305 Throttling decision does not take offheap memstore size …

[GitBox <gi...@apache.org>]

stoty commented on a change in pull request #1074:
URL: https://github.com/apache/phoenix/pull/1074#discussion_r554083743



##########
File path: phoenix-core/src/main/java/org/apache/phoenix/coprocessor/UngroupedAggregateRegionObserver.java
##########
@@ -260,7 +260,9 @@ void commitBatch(Region region, List<Mutation> mutations, long blockingMemstoreS
        Mutation[] mutationArray = new Mutation[mutations.size()];
       // When memstore size reaches blockingMemstoreSize we are waiting 3 seconds for the
       // flush happen which decrease the memstore size and then writes allowed on the region.
-      for (int i = 0; blockingMemstoreSize > 0 && region.getMemStoreHeapSize() > blockingMemstoreSize && i < 30; i++) {
+      for (int i = 0; blockingMemstoreSize > 0
+              && region.getMemStoreHeapSize() + region.getMemStoreOffHeapSize() > blockingMemstoreSize

Review comment:
       I originally thought that this bug was introduced when we adopted to the API change in HBASE-18294, but on closer examination it it was fine. 
   It was broken recently by the PHOENIX-6067 indexing rebase. 
   
   We are really just restoring the HBase 2 semantics that were broken by PHOENIX-6067.




----------------------------------------------------------------
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.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org