hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Esteban Gutierrez <este...@cloudera.com>
Subject Re: Region Server Thread with a Single High Idle CPU
Date Wed, 17 Dec 2014 20:46:31 GMT
Can you confirm if the RegionServer that is getting hammered by those
requests is serving META? How many connections per unique client do you
see? You can get that number if you ssh into that RegionServer host and
 run something like this as root: "lsof -n | grep :60020- | sed
's/.*->//;s/:.*//' | sort | uniq -c | sort -nr" If those requests are only
coming from the Thrift server, can you repeat the same process but instead
of using 60020 use the thrift server port number (e.g. 9090)?

please keep us posted,
esteban.


--
Cloudera, Inc.


On Tue, Dec 16, 2014 at 1:59 PM, uamadman <uamadman1@gmail.com> wrote:
>
> I posted these stats in the first post/ I updated to include the Thrift
> Server:
>
> Cluster Stats:
> 1 Master/14 Nodes/1 Thrift Server
> Quad\Dual-Core 2.8-3.0Ghz 6-8gb ram each with a single 500Gb Drive
> Table Stats: ~150 Tables Total Split into 590 Regions
> 1 table w/ 2.5Billion Rows 500 Columns
> 4 tables w/ 60-250 Million Rows 1000 Columns
> ~145 tables w/ 100K Rows 25 Columns
>
> Standard-Key-Template: abcd123456
>
> There is only one Thrift Server, No Rest Servers or Gateways.
> The Thrift server is allowed 200 Threads and 1.5GB of Heap Memory
>
> I've posted some images in the following drop box.
> I've attempted to best describe the images with their names.
> https://www.dropbox.com/sh/eb5xo3bmuifuapo/AABw1b9pgL6NYYCqadcG9cO8a?dl=0
>
>
>
>
> --
> View this message in context:
> http://apache-hbase.679495.n3.nabble.com/Region-Server-Thread-with-a-Single-High-Idle-CPU-tp4066876p4066938.html
> Sent from the HBase User mailing list archive at Nabble.com.
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message