Hello Cristina,
It is not the response time of the support I was concerned about in my
reply, but the fact that this problem could have been fixed already in
2011. This would have saved me and the support units quite some time
last week, as I would not have had the problem.
Kind regards,
Fokke
On 04-02-13 10:35, Cristina Aiftimiei wrote:
> Fokke Dijkstra wrote:
>> Hi Jeff and Cristina,
>>
>> I've been bitten by the issue Jeff describes last week (and probably
>> earlier), see GGUS ticket 90905. It would have been nice if this
>> patch had been available in EMI earlier.
> you point to a "very urgent" tkt opened on 28.01.2013.
> If you are not happy with the importance it is given and the velocity
> to which it is handeld with please let the developers know through the
> tkt.
>
> Thank you very much for your understanding,
> Cristina
>
>>
>> Kind regards,
>>
>> Fokke
>>
>> On 01-02-13 17:23, Cristina Aiftimiei wrote:
>>> Hi Jeff,
>>>
>>> thank you very much for your offer!
>>> I've added in CC the team under whose responsibility are now the
>>> various scheduler plugins.
>>> Please get in contact with them, or they will get in contact with
>>> you :-) if there are more details to be discussed.
>>> From what I know lately there were a lot of changes done on this
>>> side for EMI 3, I'm sure the Cream team has more details.
>>>
>>> Thank you again,
>>> Cristina
>>>
>>> Jeff Templon wrote:
>>>> Hi
>>>>
>>>> Perhaps you could accept, before the project ends, the most recent
>>>> version of the pbs plugin for the dynamic scheduler. It contains a
>>>> fix made back in April 2011 for a parse error that can occur if one
>>>> of the jobs on the LRMS has a stagein or stageout error. My own
>>>> attempt at getting it added to EMI way back when was thwarted by
>>>> ETICS.
>>>>
>>>> https://ndpfsvn.nikhef.nl/cgi-bin/viewvc.cgi/pdpsoft/nl.nikhef.pdp.dynsched-pbs-plugin/tags/REL-2.3.0/
>>>>
>>>>
>>>> The thing now being released by EMI is also being labelled 2.3.0,
>>>> but the relevant component in that release (pbsServer.py) is
>>>> identical (with formatting exceptions) to my release 2.2.1, which
>>>> does not contain the fix.
>>>>
>>>> J "better nate than lever" T
>>>>
>>>
>>>
>>
>>
>
>
--
Fokke Dijkstra <[log in to unmask]>
High Performance Computing & Visualisation
Donald Smits Center for Information Technology, University of Groningen
Postbus 11044, 9700 CA Groningen, The Netherlands
+31-50-363 9243
|