[lustre-discuss] MDT partition getting full

[Radu Popescu]

Hi

anyone has any other ideas? I’m still forced to restart servers every few weeks.

Thanks,
Radu


> On 01 May 2015, at 21:29, Mohr Jr, Richard Frank (Rick Mohr) <rmohr at utk.edu> wrote:
> 
> 
>> On Apr 29, 2015, at 6:23 PM, Mohr Jr, Richard Frank (Rick Mohr) <rmohr at utk.edu> wrote:
>> 
>> 
>>> On Apr 29, 2015, at 6:01 AM, Alexander Zarochentsev <alexander.zarochentsev at seagate.com> wrote:
>>> 
>>> 
>>> can you increase mds parameters osp.*.max_rpcs_in_progress from 4096
>>> (default) to, say, 65536, and check whether those llog files are being
>>> deleted faster?
>>> 
>> 
>> I increased the value, and I will let you know what happens.
> 
> 
> I checked the current state of the files, and this is what I am seeing in the O/1/d14 directory:
> 
> 58129  100644 (1)      0      0    8256 13-May-2014 16:51 14
>  58162  100644 (1)      0      0    8256 13-May-2014 16:51 46
>  58197  100644 (1)      0      0    8256 13-May-2014 16:51 78
>  58237  100644 (1)      0      0   38720 13-May-2014 19:28 110
>  58271  100644 (1)      0      0   39488 13-May-2014 19:28 142
>  58305  100644 (1)      0      0   38912 13-May-2014 19:28 174
>  58343  100644 (1)      0      0   38272 13-May-2014 19:28 206
>  58396  100644 (1)      0      0   38400 13-May-2014 19:28 238
>  58429  100644 (1)      0      0   37184 13-May-2014 19:28 270
>    179  100644 (17)      0      0   4153280 24-Apr-2015 04:14 43278
>    188  100644 (17)      0      0   4153280 24-Apr-2015 12:03 43310
>    206  100644 (17)      0      0   4153280 24-Apr-2015 18:42 43246
>   1304  100644 (17)      0      0   4153280 26-Apr-2015 06:47 43630
>   1285  100644 (17)      0      0   4153280 25-Apr-2015 10:17 43470
>    120  100644 (17)      0      0   4153280 25-Apr-2015 16:49 43502
>    202  100644 (17)      0      0   4153280 26-Apr-2015 11:53 43662
>    124  100644 (17)      0      0   4153280 26-Apr-2015 20:44 43694
>   1327  100644 (17)      0      0   4153280 27-Apr-2015 15:30 43822
>  12671  100644 (17)      0      0   4153280 29-Apr-2015 12:53 44558
>   9991  100644 (17)      0      0   4153280 27-Apr-2015 15:13 43790
>  10060  100644 (17)      0      0   4153280 28-Apr-2015 04:34 43886
>  10039  100644 (17)      0      0   4153280 28-Apr-2015 13:03 43918
>  10019  100644 (17)      0      0   4153280 29-Apr-2015 03:01 44046
>  10131  100644 (17)      0      0   4153280 29-Apr-2015 00:15 44014
>  12661  100644 (17)      0      0   2562624 30-Apr-2015 15:55 44622
>  12724  100644 (17)      0      0   1222528  1-May-2015 03:11 44654
> 
> New files are still being created, and the old ones are not being deleted.  However, after making the change, it looks like the two most recent files are not as large as the previous ones.  I don’t know if this is coincidence or a direct result of the new setting.  
> 
> Of course, there is still the question of where these files are coming from and why they aren’t going away.
> 
> --
> Rick Mohr
> Senior HPC System Administrator
> National Institute for Computational Sciences
> http://www.nics.tennessee.edu
> 
> _______________________________________________
> lustre-discuss mailing list
> lustre-discuss at lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20150514/71f0e6ec/attachment.htm>