[
https://issues.apache.org/jira/browse/CASSANDRA-6510?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13889940#comment-13889940
]
Robert Coli commented on CASSANDRA-6510:
----------------------------------------
This bug seems to have the implication that no ConsistencyLevel has had its
supposed meaning for the duration of the bug, because there is no guarantee
that the acknowledged-to-the-client local write actually succeeds? Is that
correct?
If so, this issue seems quite fundamental and serious; why did automated
testing not surface it? Is there now a test which covers this case?
What is the "since" for this issue? Looks like at least 1.2.0?
> Don't drop local mutations without a trace
> ------------------------------------------
>
> Key: CASSANDRA-6510
> URL: https://issues.apache.org/jira/browse/CASSANDRA-6510
> Project: Cassandra
> Issue Type: Bug
> Reporter: Aleksey Yeschenko
> Assignee: Aleksey Yeschenko
> Fix For: 1.2.14, 2.0.4
>
> Attachments: 6510.txt
>
>
> SP.insertLocal() uses a regular DroppableRunnable, thus timed out local
> mutations get dropped without leaving a trace. SP.insertLocal() should be
> using LocalMutationRunnable instead.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
