[Gluster-users] Quota

Vijaikumar M vmallika at redhat.com
Tue Jul 14 08:27:02 UTC 2015



On Tuesday 14 July 2015 01:15 PM, Geoffrey Letessier wrote:
> Hi Vijay,
>
> It’s probably due to the fact I set the brick logs to CRITICAL.
>
> I can’t confirm it to you this because, as you can read in one my 
> yesterday emails, one time yesterday evening I met the same trouble 
> with no quota set (but all //tmp/<temporary_mounts> were stil here, so… )
>
> That said, I feel the problem come from quota when transport-type is 
> set to RDMA… I dont met this issue with TCP transport type.
>
> In addition I note a strange performance with RDMA (TCPoIB provides 
> bigger performances in my tests (small files operations))
>
Regarding Quota performance, some improvements are done in 3.7.3.

Adding RDMA team in the loop, to provide more information on RDMA 
performance.

Thanks,
Vijay


> Thanks,
> Geoffrey
> ------------------------------------------------------
> Geoffrey Letessier
> Responsable informatique & ingénieur système
> UPR 9080 - CNRS - Laboratoire de Biochimie Théorique
> Institut de Biologie Physico-Chimique
> 13, rue Pierre et Marie Curie - 75005 Paris
> Tel: 01 58 41 50 93 - eMail: geoffrey.letessier at ibpc.fr 
> <mailto:geoffrey.letessier at ibpc.fr>
>
> Le 14 juil. 2015 à 05:33, Vijaikumar M <vmallika at redhat.com 
> <mailto:vmallika at redhat.com>> a écrit :
>
>> Hi Geoffrey,
>>
>> In the brick logs, there is no graph information and also no client 
>> has connected to the bricks (volume may not be mounted).
>> You mentioned that you were able to mount and use volume with quota 
>> disabled, can you confirm that without quota volume was mounted 
>> successfully and no error in the client log?
>>
>>
>> Thanks,
>> Vijay
>>
>>
>>
>> On Monday 13 July 2015 10:21 PM, Geoffrey Letessier wrote:
>>> Hi Vijay,
>>>
>>> OK, no problem.
>>> This issue only appears when I enable to quota feature on my volume. 
>>> If I disable it, the problem disappears in the same time…
>>>
>>> Here (in attachments) you can find my brick logs.
>>>
>>> Let me know if you need something else.
>>> Good luck :) and thanks by advance,
>>> Geoffrey
>>>  ------------------------------------------------------
>>> Geoffrey Letessier
>>> Responsable informatique & ingénieur système
>>> UPR 9080 - CNRS - Laboratoire de Biochimie Théorique
>>> Institut de Biologie Physico-Chimique
>>> 13, rue Pierre et Marie Curie - 75005 Paris
>>> Tel: 01 58 41 50 93 - eMail: geoffrey.letessier at ibpc.fr 
>>> <mailto:geoffrey.letessier at ibpc.fr>
>>>
>>> Le 13 juil. 2015 à 18:41, Vijaikumar M <vmallika at redhat.com 
>>> <mailto:vmallika at redhat.com>> a écrit :
>>>
>>>> Hi Geoffrey,
>>>>
>>>> Can we get brick-logs from all the bricks?
>>>>
>>>> Thanks,
>>>> Vijay
>>>>
>>>> On Monday 13 July 2015 08:15 PM, Atin Mukherjee wrote:
>>>>>
>>>>> -Atin
>>>>> Sent from one plus one
>>>>> On Jul 13, 2015 8:08 PM, "Geoffrey Letessier" 
>>>>> <geoffrey.letessier at cnrs.fr <mailto:geoffrey.letessier at cnrs.fr>> 
>>>>> wrote:
>>>>> >
>>>>> > Hi Atin,
>>>>> >
>>>>> > First of all, thanks for your reply and for your help.
>>>>> > Now, after having set features.quota-deem-statfs to on, my df hang:
>>>>> >
>>>>> > # gluster volume set vol_home features.quota-deem-statfs on;
>>>>> > volume set: success
>>>>> > [root at lucifer ~]# df
>>>>> I am not a quota expert, Vijaikumar can help you on this. However 
>>>>> 'strace df' might give you a hint where exactly its stuck.
>>>>> >
>>>>> > An idea? :)
>>>>> >
>>>>> > Thanks in advance,
>>>>> > Geoffrey
>>>>> > ------------------------------------------------------
>>>>> > Geoffrey Letessier
>>>>> > Responsable informatique & ingénieur système
>>>>> > UPR 9080 - CNRS - Laboratoire de Biochimie Théorique
>>>>> > Institut de Biologie Physico-Chimique
>>>>> > 13, rue Pierre et Marie Curie - 75005 Paris
>>>>> > Tel: 01 58 41 50 93 - eMail: geoffrey.letessier at ibpc.fr 
>>>>> <mailto:geoffrey.letessier at ibpc.fr>
>>>>> >
>>>>> >> Le 13 juil. 2015 à 16:29, Atin Mukherjee 
>>>>> <atin.mukherjee83 at gmail.com <mailto:atin.mukherjee83 at gmail.com>> a 
>>>>> écrit :
>>>>> >>
>>>>> >> Please bump up the op version by Gluster volume set all 
>>>>> cluster.op-version 30702
>>>>> >>
>>>>> >> -Atin
>>>>> >> Sent from one plus one
>>>>> >>
>>>>> >> On Jul 13, 2015 7:43 PM, "Geoffrey Letessier" 
>>>>> <geoffrey.letessier at cnrs.fr <mailto:geoffrey.letessier at cnrs.fr>> 
>>>>> wrote:
>>>>> >>>
>>>>> >>> Dear all,
>>>>> >>>
>>>>> >>> I just upgraded my storage cluster from 3.5.3 to the latest 
>>>>> version of GlusterFS (3.7.2-3) but when I want to activate quota 
>>>>> info my volume I obtain this result:
>>>>> >>> # gluster volume quota vol_home enable
>>>>> >>> quota command failed : Volume quota failed. The cluster is 
>>>>> operating at version 3. Quota command enable is unavailable in 
>>>>> this version.
>>>>> >>>
>>>>> >>> Can you give me your help?
>>>>> >>>
>>>>> >>> Thanks in advance,
>>>>> >>> Geoffrey
>>>>> >>> ------------------------------------------------------
>>>>> >>> Geoffrey Letessier
>>>>> >>> Responsable informatique & ingénieur système
>>>>> >>> UPR 9080 - CNRS - Laboratoire de Biochimie Théorique
>>>>> >>> Institut de Biologie Physico-Chimique
>>>>> >>> 13, rue Pierre et Marie Curie - 75005 Paris
>>>>> >>> Tel: 01 58 41 50 93 - eMail: geoffrey.letessier at ibpc.fr 
>>>>> <mailto:geoffrey.letessier at ibpc.fr>
>>>>> >>>
>>>>> >>>
>>>>> >>> _______________________________________________
>>>>> >>> Gluster-users mailing list
>>>>> >>> Gluster-users at gluster.org <mailto:Gluster-users at gluster.org>
>>>>> >>> http://www.gluster.org/mailman/listinfo/gluster-users
>>>>> >
>>>>> >
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Gluster-users mailing list
>>>>> Gluster-users at gluster.org
>>>>> http://www.gluster.org/mailman/listinfo/gluster-users
>>>>
>>>
>>>
>>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20150714/4fe14cd0/attachment.html>


More information about the Gluster-users mailing list