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
HBase >> mail # user >> Extending HBase Metrics Framework


Copy link to this message
-
Re: Extending HBase Metrics Framework
Thanks for pointing out the way. I've looked at the Git code as well
however looking back at the apidocs <http://hbase.apache.org/apidocs/> for
HBase-0.97.0-Snapshot I can only see org.apache.hadoop.metrics2 package
there.

I am using HBase 0.94.7, could you kindly help directing to the right code
repo and package?

Thanks again...
Regards,
Joarder Kamal
On 7 June 2013 03:50, Jeff Kolesky <[EMAIL PROTECTED]> wrote:

> If you want to add metrics, you can implement a
> org.apache.hadoop.metrics.Updater and register it with a MetricsContext.
>  There is an example at:
>
>
> https://github.com/jeffkole/odoop/blob/master/hbase-filters/src/main/java/com/opower/hadoop/hbase/filter/DeployedFilterMetrics.java
>
> This exposes metrics for a filter that I wrote, but if you want to pull out
> metrics for HBase internals, you likely need to do modify HBase source code
> to get at the data or hook up to your metric updater.
>
> Jeff
>
>
> On Thu, Jun 6, 2013 at 10:35 AM, Joarder KAMAL <[EMAIL PROTECTED]> wrote:
>
> > Hi Bing,
> >
> > Thanks for answering.
> >
> > 1. Probably my question was unclear. I was asking whether it is possible
> to
> > generate custom metrics from HBase by modifying the source code.
> > 2. Thanks for letting me know that it is possible to log the metrics in
> any
> > kind of file. But I guess it would be still in XML format !! Can I save
> the
> > logs in raw text format?? I'll try to do this tomorrow.
> >
> > Thanks again.
> >
> >
> > Regards,
> > Joarder Kamal
> >
> >
> >
> > On 6 June 2013 17:42, Bing Jiang <[EMAIL PROTECTED]> wrote:
> >
> > > 1) Metrics infos have been collected from HMaster\RS, and covering rpc,
> > > jvm, hbase functional module (like compaction\flush\request
> > Count\DataBlock
> > > read &&write etc).
> > >
> > > 2) You can set metrics info handler in conf/hadoop-metrics.properties
> > > for example, you can set hbase.class > > > org.apache.hadoop.metrics.ganglia.GangliaContext31, it will tell
> Metrics
> > > System where the metrics will be sent to. Of course, this is Ganglia.
> > > if set
> > >
> hbase.class=org.apache.hadoop.hbase.metrics.file.TimeStampingFileContext,
> > > it will write the metrics to file.
> > >
> > > # hbase.class=org.apache.hadoop.metrics.ganglia.GangliaContext
> > > # hbase.class=org.apache.hadoop.metrics.ganglia.GangliaContext31
> > > # hbase.period=10
> > > # hbase.servers=10.100.4.151:8650
> > >
> #hbase.class=org.apache.hadoop.hbase.metrics.file.TimeStampingFileContext
> > > #hbase.period=10
> > > #hbase.fileName=/opt/hbase/logs/metrics_hbase.log
> > >
> > >
> > >
> > >
> > > 2013/6/6 Joarder KAMAL <[EMAIL PROTECTED]>
> > >
> > > > Two Questions:
> > > >
> > > > 1. Is it possible to extend HBase metrics/reporting framework to
> > generate
> > > > custom metrics? I was looking at the api docs
> > > > http://hbase.apache.org/apidocs/
> > > > 2. Is there any way to write my own client (instead of using Ganglia
> or
> > > > else) to retrieve the currently available metrics?
> > > >
> > > > HBase version: 0.94.7
> > > >
> > > > Any suggestions will be very helpful.
> > > >
> > > > Many thanks ...
> > > >
> > > > Regards,
> > > > Joarder Kamal
> > > >
> > >
> > >
> > >
> > > --
> > > Bing Jiang
> > > Tel:(86)134-2619-1361
> > > weibo: http://weibo.com/jiangbinglover
> > > BLOG: http://blog.sina.com.cn/jiangbinglover
> > > National Research Center for Intelligent Computing Systems
> > > Institute of Computing technology
> > > Graduate University of Chinese Academy of Science
> > >
> >
>
>
>
> --
> *Jeff Kolesky*
> Chief Software Architect
> *Opower*
>
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