[
https://issues.apache.org/jira/browse/CASSANDRA-5442?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13658429#comment-13658429
]
Sylvain Lebresne commented on CASSANDRA-5442:
---------------------------------------------
Not sure if the patch was pushed for review or not. In any case, fixing the
build is pretty trivial, the changes made to ModificationStatement should just
be reverted (and we should just pass the CL in argument to
executeWithCondition() to SP.cas()).
Other than that:
* I believe we now refuse CL.SERIAL for the cas() method (because CL.blockfor()
throw an UnsupportedOperationException for SERIAL). Maybe we can make SERIAL
mean "wait for nothing on commit, not even a local one" (which probably amount
to making CL.blockfor() return 0 for SERIAL and adding a 'if' in
WriteResponseHandler ctor to signal right away if blockfor is 0).
* I don't think CL.ANY will really work as for normal writes since we don't
generate hints for commit. Maybe we could (generate hints) ultimately btw, but
in the meantime, it might make more sense to refuse it (and thus make the
default for CAS be either SERIAL if we do the preceding point or ONE).
* We don't call CL.validateForWrite() when calling CAS (and we shouldn't as it
complains with SERIAL), but maybe we should add a validateForConditionalWrite
that would at least check we have the right replication strategy when using the
multi-DC specific CLs.
> Add support for specifying CAS commit CL
> ----------------------------------------
>
> Key: CASSANDRA-5442
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5442
> Project: Cassandra
> Issue Type: Sub-task
> Components: API, Core
> Reporter: Jonathan Ellis
> Assignee: Sylvain Lebresne
> Fix For: 2.0
>
>
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
