[rabbitmq-discuss] High cpu usage on CentOS boxes

Tim Watson watson.timothy at gmail.com
Wed Nov 27 10:21:10 GMT 2013


As Simon said, this is highly unlikely o be anything to do with the OS distro, however it could be due to differences in the Linux kernel version the two distro run over and/different versions of erlang/OTP. Are both the servers runnin the same Linux kernel and the same erlang version? Are both 64/32bit architectures and is the erlang runtime they're both running compiled for the same?

Cheers,
Tim

On 26 Nov 2013, at 05:24, shridharan muthu <shridharan.m at gmail.com> wrote:

> Hello Simon,
>    Cent OS takes much lesser traffic than I have on Ubuntu. I migrated 5% cpu traffic from ubuntu and It caused 15% cpu increase in Cent OS. I goggled and noticed quite similar questions but didn't find any solutions. If there are a few knobs that I can tune,  I would be happy to test it out. 
>  
>     BTW, I don't use Mandatory/Immediate flags or Acks. All my publisher & consumer calls are asynchronous. 
> 
> Thanks,
> Shri
> 
> 
> On Mon, Nov 25, 2013 at 9:28 AM, Simon MacMullen <simon at rabbitmq.com> wrote:
> I'm afraid there isn't a good way to see which parts of the broker are busy at the moment (we hope to add something along these lines in a future release).
> 
> I'm suspicious of the idea that it's distro-related though. Is the CentOS server really performing the same workload as the Ubuntu server?
> 
> To get to the bottom of this you'd have to gradually take bits away until you see the CPU usage drop - stop connections, disable plugins and so on.
> 
> Cheers, Simon
> 
> 
> On 25/11/2013 16:53, shridharan muthu wrote:
> Hello There,
> 
>     Reg my previous message, is there anyway to dig into the high cpu
> usage issue? With similar configs in Ubuntu & Cent OS, Cent OS takes lot
> more cpu for beam.smp process. I switched log level to "info" but
> noticed nothing unusual.
> 
> Any help would be appreciated.
> 
> Thanks,
> Shri
> 
> 
> On Fri, Nov 22, 2013 at 3:56 PM, shridharan muthu
> <shridharan.m at gmail.com <mailto:shridharan.m at gmail.com>> wrote:
> 
>     Hi There,
>          I am experiencing high cpu usage (above 400%) on Cent OS rabbit
>     mq cluster (with 2 nodes). We have a similar setup in Ubuntu and cpu
>     usage seems fine. Pls fine the report, status and top below.
> 
>     OS: /CentOS release 6.4/
> 
>     Report:
>     /{pid,2148},/
>     / {running_applications,/
>     /     [{rabbitmq_management,"RabbitMQ Management Console","3.0.2"},/
>     /      {rabbitmq_management_agent,"RabbitMQ Management Agent","3.0.2"},/
>     /      {rabbit,"RabbitMQ","3.0.2"},/
>     /      {os_mon,"CPO  CXC 138 46","2.2.7"},/
>     /      {rabbitmq_mochiweb,"RabbitMQ Mochiweb Embedding","3.0.2"},/
>     /      {webmachine,"webmachine","1.9.1-rmq3.0.2-git52e62bc"},/
>     /      {mochiweb,"MochiMedia Web Server","2.3.1-rmq3.0.2-gitd541e9a"},/
>     /      {xmerl,"XML parser","1.2.10"},/
>     /      {inets,"INETS  CXC 138 49","5.7.1"},/
>     /      {mnesia,"MNESIA  CXC 138 12","4.5"},/
>     /      {amqp_client,"RabbitMQ AMQP Client","3.0.2"},/
>     /      {sasl,"SASL  CXC 138 11","2.1.10"},/
>     /      {stdlib,"ERTS  CXC 138 10","1.17.5"},/
>     /      {kernel,"ERTS  CXC 138 10","2.14.5"}]},/
>     / {os,{unix,linux}},/
>     / {erlang_version,/
>     /     "Erlang R14B04 (erts-5.8.5) [source] [64-bit] [smp:12:12]
>     [rq:12] [async-threads:30] [kernel-poll:true]\n"},/
>     / {memory,/
>     /     [{total,285898072},/
>     /      {connection_procs,16592856},/
>     /      {queue_procs,1064888},/
>     /      {plugins,269872},/
>     /      {other_proc,13173224},/
>     /      {mnesia,100632},/
>     /      {mgmt_db,2167672},/
>     /      {msg_index,35352},/
>     /      {other_ets,72058416},/
>     /      {binary,7326576},/
>     /      {code,17434264},/
>     /      {atom,3356065},/
>     /      {other_system,152318255}]},/
>     / {vm_memory_high_watermark,0.4},/
>     / {vm_memory_limit,4991811584},/
>     / {disk_free_limit,1000000000},/
>     / {disk_free,17927184384},/
>     / {file_descriptors,/
>     /     [{total_limit,199900},/
>     /      {total_used,314},/
>     /      {sockets_limit,179908},/
>     /      {sockets_used,312}]},/
>     / {processes,[{limit,1048576},{used,2703}]},/
>     / {run_queue,0},/
>     / {uptime,3207797}]/
> 
>      From Top:
>     /   2249 rabbitmq  20   0 2989m 319m 2540 S 371.5  2.7  25109:43
>     beam.smp /
> 
> 
> 
>       Pls advise me how to proceed further.
> 
>     Thanks,
>     Shri
> 
> 
> 
> 
> _______________________________________________
> rabbitmq-discuss mailing list
> rabbitmq-discuss at lists.rabbitmq.com
> https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss
> 
> 
> 
> _______________________________________________
> rabbitmq-discuss mailing list
> rabbitmq-discuss at lists.rabbitmq.com
> https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20131127/f15b71ff/attachment.htm>


More information about the rabbitmq-discuss mailing list