[
https://issues.apache.org/jira/browse/CASSANDRA-17619?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17534412#comment-17534412
]
Paulo Motta commented on CASSANDRA-17619:
-----------------------------------------
This was working correctly in the initial implementation of CASSANDRA-16843,
but ended up getting removed when I simplified the patch in the final review
pass.
I added a few regression tests [on this
commit|https://github.com/pauloricardomg/cassandra/commit/dd10697ecdd234ddbbbbe032fc076f36787c9197]
reproducing the issue with both auto_snapshots and manual snapshots.
The fix is straightforward: just use the new {{SnapshotLoader}} implementation
when doing the initial snapshot load on {{{}SnapshotManager{}}}, which loads
snapshots of dropped tables correctly
([commit|https://github.com/pauloricardomg/cassandra/commit/6731ee21b79b2345274307e2ba8800f50eb36a98]).
> Expired snapshots of dropped tables are not removed after node restart
> ----------------------------------------------------------------------
>
> Key: CASSANDRA-17619
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17619
> Project: Cassandra
> Issue Type: Bug
> Reporter: Paulo Motta
> Priority: Normal
>
> Expired TTL snapshots from a dropped table are only removed if the node is
> not restarted after the table is dropped.
> The reason for this is because {{SnapshotManager}} is using the old
> implementation of {{listsnapshots}} to load snapshots in memory during node
> startup, before CASSANDRA-16843 is fixed so it doesn't keep track of
> snapshots of dropped tables.
> We should make {{SnapshotManager}} use the new implementation to load
> snapshots of dropped tables in memory, what ensures they will be removed when
> expired.
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
