sk0x50 commented on code in PR #6977:
URL: https://github.com/apache/ignite-3/pull/6977#discussion_r2538269412
##########
modules/catalog-compaction/src/main/java/org/apache/ignite/internal/catalog/compaction/CatalogCompactionRunner.java:
##########
@@ -580,18 +571,17 @@ private CompletableFuture<Void> collectRequiredNodes(
int partitions = zone.partitions();
- List<ReplicationGroupId> replicationGroupIds = new
ArrayList<>(partitions);
+ List<ZonePartitionId> replicationGroupIds = new
ArrayList<>(partitions);
for (int p = 0; p < partitions; p++) {
- replicationGroupIds.add(nodeProperties.colocationEnabled() ? new
ZonePartitionId(table.zoneId(), p)
- : new TablePartitionId(table.id(), p));
+ replicationGroupIds.add(new ZonePartitionId(table.zoneId(), p));
}
return placementDriver.getAssignments(replicationGroupIds, nowTs)
.thenAccept(tokenizedAssignments -> {
assert tokenizedAssignments.size() ==
replicationGroupIds.size();
- if (nodeProperties.colocationEnabled() &&
currentCatalog.table(table.id()) == null) {
+ if (currentCatalog.table(table.id()) == null) {
// Table no longer exists
deletedTables.put(table.id(), true);
Review Comment:
Yes, it is required. Otherwise, the catalog compaction will not be able to
make progress since the assignments for a zone will never be `null` while the
zone exists. See `ItCatalogCompactionTest.testCompactionRun`.
--
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]
