[Gluster-users] Fwd: Problem in Benchmarking Glusterfs with Bonnie++ on CentOS 6 machines

gauri Desai gauridesai719 at gmail.com
Sun Nov 16 06:41:29 UTC 2014


---------- Forwarded message ----------
From: gauri Desai <gauridesai719 at gmail.com>
Date: Sun, Nov 16, 2014 at 12:40 AM
Subject: Re: [Gluster-users] Problem in Benchmarking Glusterfs with
Bonnie++ on CentOS 6 machines
To: Humble Devassy Chirammal <humble.devassy at gmail.com>


Hello Chirammal,

The following are the details

What protocol ( fuse/nfs..etc) has been used here to mount the gluster
volume in your client?
Fuse

Are these bricks formatted with XFS?
no they are not formatted with XFS

version on Bonnie++?
bonnie++-1.03e.tgz <http://www.coker.com.au/bonnie++/bonnie++-1.03e.tgz>

Latest when i tried running Bonnie++ again it created several files
Bonnie.(numbered extension) at the destination (/mnt/glusterfs) but it is
not possible to view the content of these files.

Thanks,
Gauri

On Sun, Nov 16, 2014 at 12:29 AM, Humble Devassy Chirammal <
humble.devassy at gmail.com> wrote:

> Can you provide some more information about the setup .
>
> What protocol ( fuse/nfs..etc) has been used here to mount the gluster
> volume in your client ?
>
> Are these bricks formatted with XFS ?
>
> version of bonnie++ binary ?
>
>
>
> --Humble
>
>
> On Sun, Nov 16, 2014 at 9:59 AM, gauri Desai <gauridesai719 at gmail.com>
> wrote:
>
>> Hello Vijay,
>>
>>  The following are the details:
>>
>> 1) glusterfs version:
>>  3.6.1
>>
>> 2) Glusterfs volume info
>>
>> i have attached a screen shot
>>
>> 3) command used :
>>
>>  Bonnie++ -d /mnt/glusterfs/ -r 490 -u root
>>
>> 4)possible errors in glusterfs log files
>>
>> nothing noticeable in the log files but after running the bonnie command
>> the following error message appears
>>
>> Stat files in sequential order.... Expected 16384 files but only got
>> 163811 files.
>> Cleaning up test directory after error.
>>
>>
>>
>>
>>
>> On Sat, Nov 15, 2014 at 11:54 AM, Vijay Bellur <vbellur at redhat.com>
>> wrote:
>>
>>> On 11/14/2014 11:13 PM, gauri Desai wrote:
>>>
>>>> Hello List,
>>>>
>>>>        I have set up a 4 node replicated distributed Filesystem using
>>>> Glusterfs on centos 6 machines.
>>>>
>>>> I am trying to benchmark the glusterfs filesystem using Bonnie++, but as
>>>> soon as i run bonnie++ on  the client after some time i get an error
>>>> saying:
>>>>
>>>> Stat files in sequential order.... Expected 16384 files but only got
>>>> 163811 files.
>>>> Cleaning up test directory after error.
>>>>
>>>>
>>>> The RAM size on my client is 490M and the mounted GlusterFS filessystem
>>>> mounted on the client is 11G.
>>>>
>>>> Why might I be facing this issue? What can be a solution to this?
>>>>
>>>>
>>> More details on your volume configuration (o/p of gluster volume info),
>>> possible errors in glusterfs log files, glusterfs version, the exact
>>> bonnie++ command line used etc. will help in narrowing down the problem.
>>>
>>> Regards,
>>> Vijay
>>>
>>>
>>
>> _______________________________________________
>> Gluster-users mailing list
>> Gluster-users at gluster.org
>> http://supercolony.gluster.org/mailman/listinfo/gluster-users
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20141116/31168b16/attachment.html>


More information about the Gluster-users mailing list