Hello...
For the record, this issue was solved offline.
For Pavel problem, it was due to the fact that user tomcat was not
allowed to query the GE server because of certificate issues.
For Rod problem, it was due to the fact that BUpdaterSGE daemon was not
inheriting proper GE environment.
Both problems are now documented in
https://wiki.italiangrid.it/twiki/bin/view/CREAM/KnownIssues#CREAM_jobs_are_cancelled_with_st
Cheers
Goncalo
On 12/03/2012 10:59 AM, Pavel Weber wrote:
> Hi Goncalo,
>
> Yes, I contacted the ge list directly although with another problem
> related to ge cream (late file staging problem), which turned out to
> be the reason=3 later on. So the problems can be merged now. Thank you.
>
> Cheers,
>
> Pavel
>
> On 12/03/2012 11:33 AM, Gonçalo Borges wrote:
>> Hi Dimitri...
>>
>> Is Pavel Weber (from KIT also) working with you? Because he has
>> contacted the GE list directly, and we are providing support to him
>> via that way...
>>
>> Cheers
>> Goncalo
>>
>> On 11/29/2012 03:42 PM, Dimitri Nilsen wrote:
>>> Hi,
>>>
>>> where there any solution found for this issue?
>>> we are running in the same problem now.
>>>
>>> Regards
>>>
>>> On 10/25/2012 12:23 PM, Andreas Haupt wrote:
>>>> Hi Torsten,
>>>>
>>>> On Wed, 2012-10-24 at 22:35 +0200, Torsten Harenberg wrote:
>>>>>> Is the same queue used for both atlas analysis and atlas
>>>>>> production ?
>>>>>
>>>>> no.. different queues "atlasanl.q" and "atlasprd.q". atlasanl.q
>>>>> works, atlasprd.q not.
>>>>
>>>> Are the SGE queues configured identically? Especially the
>>>> "shell_start_mode" option might be worth a look.
>>>>
>>>> Cheers,
>>>> Andreas
>>>
>>>
>>
>>
|