[
https://issues.apache.org/jira/browse/CASSANDRA-4487?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13430668#comment-13430668
]
Pavel Yaskevich commented on CASSANDRA-4487:
--------------------------------------------
bq. Hum, I could have swear we did that for thrift but apparently not. However
we do do it for CQL (that is both CQL2 and for CQL3). Maybe that just a relic
of old time, I don't know.
It should be, because it only makes sense to check it for schema migrations.
bq. I think the waiting for schema settled shouldn't be part of the request
path, but should be done separately by clients if they want to. But if we were
doing the wait in thrift, I would have suggested to keep it just for the sake
of not changing the behavior. But anyway, if we don't do it, I guess there is
no problem.
I'm not sure if that hurts that much to leave such validation on schema
migration patch, definitely would avoid problems related to change contention
or errors on the read/write patch with CL > ONE if things weren't settled
properly. But I think if we are to remove checks we probably should remove all
of them to keep behavior consistent.
> remove uses of SchemaDisagreementException
> ------------------------------------------
>
> Key: CASSANDRA-4487
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4487
> Project: Cassandra
> Issue Type: Bug
> Components: API
> Affects Versions: 1.2
> Reporter: Jonathan Ellis
> Assignee: Pavel Yaskevich
> Priority: Minor
> Fix For: 1.2
>
> Attachments: 0001-code-changes.patch, 0002-re-generated-thrift.patch
>
>
> Since we can handle concurrent schema changes now, there's no need to
> validateSchemaAgreement before modification now.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
