JISCMail - GRIDPP-STORAGE Archives

Print

As requested in meeting

[gsongara@lxplus411]~% lcg-infosites --vo biomed space | grep se2.ppgrid1.rhul.ac.uk
    3719    -3860        0        0        0        0 -                      se2.ppgrid1.rhul.ac.uk

[root@se2 ~]# dpm-listspaces

POOLS:

commonpool
    CAPACITY: 642.07T     RESERVED: 642.20T     UNAVAIL (free/used): 0.00/0.00
                          USED: 0.00            FREE: 3.38T (0.5%)
    Space Tokens: ATLASDATADISK, ATLASGROUPDISK, ATLASHOTDISK,
                  ATLASLOCALGROUPDISK, ATLASPRODDISK,
                  ATLASSCRATCHDISK
    Authorized FQANs: all VOs
    Space Type: Any                       Retention Policy: Replica
    Number of file systems : 34           FS selection policy: maxfreespace

SPACE RESERVATIONS:

ATLASDATADISK   ID=c56f0f41-a8c7-4270-be29-a8ab92a2e641
    CAPACITY: 551.00T     RESERVED: 551.00T     UNAVAIL (free): 0
                          USED: 484.52T         FREE: 66.48T (12.1%)
    Space Type: Any       Retention: Replica    Latency: Online
    Lifetime: Infinite
    Authorized FQANs: atlas/Role=production
    Pool: commonpool

ATLASGROUPDISK ID=552b7d3c-3caf-45a7-9c1e-277f6040b3d9
    CAPACITY: 200.00G     RESERVED: 200.00G     UNAVAIL (free): 0
                          USED: 1.10M           FREE: 200.00G (100.0%)
    Space Type: Any       Retention: Replica    Latency: Online
    Lifetime: Infinite
    Authorized FQANs: atlas/Role=production
    Pool: commonpool

ATLASHOTDISK    ID=1f070a88-b1b5-40e4-a707-ecffc0a86762
    CAPACITY: 4.00T       RESERVED: 4.00T       UNAVAIL (free): 0
                          USED: 1.79T           FREE: 2.21T (55.1%)
    Space Type: Any       Retention: Replica    Latency: Online
    Lifetime: Infinite
    Authorized FQANs: atlas/Role=production
    Pool: commonpool

ATLASLOCALGROUPDISK     ID=ebc937c4-677e-477a-930c-e293bfa361a1
    CAPACITY: 35.00T      RESERVED: 35.00T      UNAVAIL (free): 0
                          USED: 27.81T          FREE: 7.19T (20.5%)
    Space Type: Any       Retention: Replica    Latency: Online
    Lifetime: Infinite
    Authorized FQANs: atlas
    Pool: commonpool

ATLASPRODDISK   ID=261f2148-9bd1-4b8c-8699-f6440eb9ff7c
    CAPACITY: 8.00T       RESERVED: 8.00T       UNAVAIL (free): 0
                          USED: 1.27T           FREE: 6.73T (84.1%)
    Space Type: Any       Retention: Replica    Latency: Online
    Lifetime: Infinite
    Authorized FQANs: atlas/Role=production
    Pool: commonpool

ATLASSCRATCHDISK        ID=44f98d9b-ee65-4d83-8185-9d9992aab27c
    CAPACITY: 44.00T      RESERVED: 44.00T      UNAVAIL (free): 0
                          USED: 20.17T          FREE: 23.83T (54.2%)
    Space Type: Any       Retention: Replica    Latency: Online
    Lifetime: Infinite
    Authorized FQANs: atlas
    Pool: commonpool

On Tue, Jul 17, 2012 at 9:39 AM, Govind Songara <[log in to unmask]> wrote:

Hi Sam,

Yes these is no down node.
Lost disk server was removed from dpm and database cleaned as suggested by developer.

Cheers
Govind

On Fri, Jul 13, 2012 at 2:11 PM, Sam Skipsey <[log in to unmask]> wrote:
Well, indeed - nodes which are "RDONLY" have their total size
subtracted from the available space.
But I assume it isn't this, since Govind doesn't mention any nodes being down?

Sam

On 13 July 2012 13:16, J Coles <[log in to unmask]> wrote:
> Hi Govind
>
> Manchester had a problem with DPM reporting negative space when some nodes were set to read only. I'm sure someone at the meeting would have commented if there was possibly a common cause, but I thought I would mention it in case they had not.
>
> Jeremy
>
>
> On 13 Jul 2012, at 12:12, Govind Songara wrote:
>
>> Looks like no one has faced similar issue before ?
>>
>>
>>
>> On Tue, Jul 3, 2012 at 3:00 PM, Govind Songara <[log in to unmask]> wrote:
>> Hi,
>>
>> In today dteam meeting, we talked about this issue.
>>
>> Non atlas disk space is publishing negative value.
>> lcg-infosites --vo biomed space | grep se2.ppgrid1.rhul.ac.uk
>> 3724 -3866 0 0 0 0 - se2.ppgrid1.rhul.ac.uk
>>
>> I thought it would be because few month ago we lost one disk server.
>> Atlas data were declared lost and replicated on other servers, but not other VO's data.
>>
>> I found some of dpns entry for other's VO and cleanup using dpns-rm.
>> But there are no change in negative number.
>>
>> Any suggestions?
>>
>> Thanks
>> Govind
>>
>>
>>