cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <>
Subject [jira] [Resolved] (CASSANDRA-4838) ColumnFamilyRecordWriter does not allow adjustment of Thrift/socket timeout
Date Mon, 22 Oct 2012 15:50:11 GMT


Jonathan Ellis resolved CASSANDRA-4838.

    Resolution: Not A Problem

I think you're confusing the socket timeout (from client to coordinator) with the RPC timeout
(from coordinator to replicas).  Increasing the former will not help when you are hitting
the latter.

Instead, you should lower mapreduce.output.columnfamilyoutputformat.batch.threshold.
> ColumnFamilyRecordWriter does not allow adjustment of Thrift/socket timeout
> ---------------------------------------------------------------------------
>                 Key: CASSANDRA-4838
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Hadoop
>    Affects Versions: 1.0.10
>         Environment: Linux/Ubuntu, DSE 2.1 / Cassandra 1.0.10
>            Reporter: Evan Chan
>         Attachments: cassandra-thrift-timeout.patch
> I'm using ColumnFamilyRecordWriter with an M/R job to dump data into a cluster, it was
running Cassandra 1.0.10, and is now running DSE 2.1.   Either way, I hit Thrift timeout errors
sometimes.  Looking at the code, it seems that it does not allow for setting of the Thrift
timeout, and this code has not been updated in the latest trunk.  
> I have a patch that adds a configuration parameter to allow adjustment of the Thrift
timeout in CFRecordWriter.   Let me know if you guys would be interested.

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:

View raw message