Apologies for replying to myself, but when I checked back to my notes
from when I fixed the same issue last November I found that oddly (at
least to me) GlueCEPolicyMaxCPUTime is set from resources_default.cput,
not from resources_max.cput. Sorry for not checking this before sending
the earlier mail.
John
On 27/02/2014 15:12, John Hill wrote:
> Hi Mark,
> I had a similar issue here a while back. On PBS/Torque you need to
> ensure that resources_max.cput is set for each queue (I've set it to be
> the same as resources_default.cput). While you are at it, you might
> check that resources_max.walltime is also set.
>
> John
>
> On 27/02/2014 14:20, Mark Slater wrote:
>> Hi All,
>>
>> It seems our publishing is off for LHCb but I'm not sure how to fix it.
>> It seems the MaxCPU time is not publishing correctly and indeed:
>>
>> [root@epgr02 ~]# ldapsearch -LLL -x -h localhost -p 2170 -b o=grid |
>> grep 999999
>> ...
>> GlueCEPolicyMaxCPUTime: 999999999
>> ...
>> GlueCEPolicyMaxCPUTime: 999999999
>> ...
>>
>> How can I get this fixed? Is there some dynamic setting/yaim problem
>> going on here?
>>
>> Thanks!
>>
>> Mark
|