[
https://issues.apache.org/jira/browse/ZOOKEEPER-1558?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Flavio Junqueira updated ZOOKEEPER-1558:
----------------------------------------
Attachment: ZOOKEEPER-1558.patch
I have removed the call to takeSnapshot(). When I did it,
ZkDatabaseCorruptionTest started failing and I realized that there was a bug in
the test case. We set snapCount to have servers taking a number of snapshots
during the test. However, by the time we set snapCount to 100, randRoll has
already been set and the values are typically much larger than the 2000 ops we
are running. Consequently, we never have a snapshot.
To fix it, I have changed SyncRequestProcessor to change the value of randRoll
when we set snapCount. For the purposes of this test, I think this is
sufficient.
> Leader should not snapshot uncommitted state
> --------------------------------------------
>
> Key: ZOOKEEPER-1558
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-1558
> Project: ZooKeeper
> Issue Type: Sub-task
> Components: quorum
> Affects Versions: 3.4.6
> Reporter: Flavio Junqueira
> Assignee: Flavio Junqueira
> Priority: Blocker
> Fix For: 3.4.6
>
> Attachments: ZOOKEEPER-1558.patch, ZOOKEEPER-1558.patch,
> ZOOKEEPER-1558.patch, ZOOKEEPER-1558.patch, ZOOKEEPER-1558.patch,
> ZOOKEEPER-1558.patch
>
>
> Leader currently takes a snapshot when it calls loadData in the beginning of
> the lead() method. The loaded data, however, may contain uncommitted state.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
