[
https://issues.apache.org/jira/browse/CASSANDRA-2388?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14025213#comment-14025213
]
Paulo Motta commented on CASSANDRA-2388:
----------------------------------------
Attached
[2.0-CASSANDRA-2388.patch|https://issues.apache.org/jira/secure/attachment/12649379/2.0-CASSANDRA-2388.patch]
(against 2.0), where CASSANDRA-6302 is ported to ColumnFamilyRecordReader, so
the next replicas are tried when the first one fails.
Reading from a non-local-DC replica is not a problem anymore due to the
introduction of describe_local_ring (CASSANDRA-6268), that limits the input
splits to the local DC.
I would like to acknowledge my colleague Danilo Penna Queiroz who paired with
me on this patch.
> ColumnFamilyRecordReader fails for a given split because a host is down, even
> if records could reasonably be read from other replica.
> -------------------------------------------------------------------------------------------------------------------------------------
>
> Key: CASSANDRA-2388
> URL: https://issues.apache.org/jira/browse/CASSANDRA-2388
> Project: Cassandra
> Issue Type: Bug
> Components: Hadoop
> Affects Versions: 0.6
> Reporter: Eldon Stegall
> Assignee: Mck SembWever
> Priority: Minor
> Labels: hadoop, inputformat
> Fix For: 2.0.9
>
> Attachments: 0002_On_TException_try_next_split.patch,
> 2.0-CASSANDRA-2388.patch, CASSANDRA-2388-addition1.patch,
> CASSANDRA-2388-extended.patch, CASSANDRA-2388.patch, CASSANDRA-2388.patch,
> CASSANDRA-2388.patch, CASSANDRA-2388.patch
>
>
> ColumnFamilyRecordReader only tries the first location for a given split. We
> should try multiple locations for a given split.
--
This message was sent by Atlassian JIRA
(v6.2#6252)