Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Flume >> mail # user >> doubt on hdfs sink


Copy link to this message
-
Re: doubt on hdfs sink
I got this working by setting all the properties
rollIinterval
rollCount
rollSize

Also realized, rollSize is after decompression

Thanks,
Nitin

On Tue, Sep 18, 2012 at 2:02 PM, Nitin Pawar <[EMAIL PROTECTED]> wrote:
> hello,
>
> I have a working setup of flume which writes into hdfs.
> I am using flum 1.2.0
>
> I have following settings.
>
> agent1.sinks.HDFS.hdfs.file.Type = SequenceFile
> agent1.sinks.HDFS.hdfs.writeFormat = Text
> agent1.sinks.hdfs.rollInterval = 86400
> agent1.channels.MemoryChannel-2.type = memory
>
> My doubt is, why does flume creates one hdfs file for one event. I
> want it to write a single hdfs file per day for the log.
> Can someone please help me find out what I have done wrong?
>
> Thanks
>
> --
> Nitin Pawar

--
Nitin Pawar
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB