ChenSammi commented on code in PR #9272:
URL: https://github.com/apache/ozone/pull/9272#discussion_r2525776629
##########
hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/block/DeletedBlockLogImpl.java:
##########
@@ -147,13 +156,23 @@ public void incrementCount(List<Long> txIDs)
}
private DeletedBlocksTransaction constructNewTransaction(
- long txID, long containerID, List<Long> blocks) {
- return DeletedBlocksTransaction.newBuilder()
+ long txID, long containerID, List<DeletedBlock> blocks) {
+ List<Long> localIdList = blocks.stream().map(b ->
b.getBlockID().getLocalID()).collect(Collectors.toList());
+ DeletedBlocksTransaction.Builder builder =
DeletedBlocksTransaction.newBuilder()
.setTxID(txID)
.setContainerID(containerID)
- .addAllLocalID(blocks)
- .setCount(0)
- .build();
+ .addAllLocalID(localIdList)
+ .setCount(0);
+
+ if
(VersionedDatanodeFeatures.isFinalized(HDDSLayoutFeature.STORAGE_SPACE_DISTRIBUTION))
{
Review Comment:
The transactions counted into summary have two conditions, one is tx should
be no less than firstTxID, second is tx has size and replicated size set.
This check here is for case that SCM is finalized while there is still old
OM sending old requests, those requests should not be counted into summary.
Since their tx ID is greater than firstTxID, then their size and replicated
size should not be set, otherwise these txs will be counted int the summary,
which is should not.
--
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]
