sadanand48 commented on code in PR #9129:
URL: https://github.com/apache/ozone/pull/9129#discussion_r2427962579
##########
hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/snapshot/SnapshotCache.java:
##########
@@ -151,7 +151,6 @@ public void invalidate(UUID key) {
LOG.warn("SnapshotId: '{}' does not exist in snapshot cache.", k);
} else {
try {
- v.get().getMetadataManager().getStore().flushDB();
Review Comment:
no that won't explicitly flush it and this is reproed by the test , we also
transfer the wal file too because of which there is no issue in the read. Now
that I don't flush it as part of this, I removed the deleteWalFiles method from
the test. I can keep it too. I removed it because of the following conversation
with @swamirishi
> q. do you think we should also flush DB as part of cleanup in write lock.
> i.e when snapshot ref count becomes 0 it only closes it today, doesn't
flush it. so if there are last minute updates it would be only in the wal.)
> a. There would be no updates since we have taken a bootstrap lock
Let me keep it anyway (keep the flush call) it would be better to flush the
changes into an sst rather than relying on a WAL file.
@swamirishi @jojochuang hope you agree with this
--
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]
