rpuch commented on code in PR #6211:
URL: https://github.com/apache/ignite-3/pull/6211#discussion_r2192208547
##########
modules/storage-api/src/main/java/org/apache/ignite/internal/storage/engine/StorageEngine.java:
##########
@@ -78,4 +80,12 @@ static long defaultDataRegionSize() {
//noinspection NumericCastThatLosesPrecision
return Math.max(256 * MiB, (long) (0.2 * getTotalMemoryAvailable()));
}
+
+ /**
+ * Returns IDs of tables for which there are MV partition storages that
were created, but they were not yet destroyed
+ * (or the destruction did not finish).
+ */
+ default Set<Integer> nonDestroyedTableIds() {
Review Comment:
It's subtle: there are tables contained in the storage engines (for which
createMvTable() was called in this Ignite instance), and there are data
structures on disk that represent tables in the storage engine. Those are
different. This difference is especially visible for volatile storage engines:
they can have some tables, but they never have anything on disk, so the new
method I added will always return an empty set for them.
How about `tableIdsOnDisk()`?
--
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]
