Hi Benjamin.
Hard to say offhand, without digging deeper, but you could be right about
your system suffering from transient issues. DNS is a usual suspect and
you should probably check whether the time is synchronised between the two
machines too. I recommend using NTP and a caching-only nameserver. See
if they fix your issues.
Regards,
Prashanth Chengi
National PARAM SuperComputing Facility
System Administration and Networking Group
C-DAC Pune
--
"Don't ever take a fence down until you know the reason it was put up."
-G.K.Chesterton
On Thu, 10 Feb 2011, "L?hnhardt, Benjamin" wrote:
Hello everyone,
I have some trouble with globus jobs, that are running from a client and
should be executed on a globus server (nimrod.med.uni-goettingen.de). The
client gets the following error soap message:
<soapenv:Fault xmlns:soapenv="http://www.w3.org/2003/05/soap-envelope";>
<soapenv:Code>
<soapenv:Value>env:Server</soapenv:Value>
</soapenv:Code>
<soapenv:Reason>
<soapenv:Text xml:lang="en">Activity
'Romanus_d0ea7b40-3454-11e0-80a7-b9b884222201_0000002213': Exception during
WS-GRAM invocation</soapenv:Text>
</soapenv:Reason>
<soapenv:Node>nimrod.med.uni-goettingen.de/LSF</soapenv:Node>
<soapenv:Detail>factoryType=LSF
operationName=software:medigrid-fsl-probtrackx-pack-v3
executable=/opt/medigrid/medigridbvapp/DTI1.0_ROMANUS/gwes_fsl_probtrackx_pa
ck_v3.sh
factoryEndpoint=https://nimrod.med.uni-goettingen.de:8443/wsrf/services/Mana
gedJobFactoryService
resourceName=hardware:nimrod.med.uni-goettingen.de/LSF
GSSException: Failure unspecified at GSS-API level [Caused by:
nimrod.med.uni-goettingen.de]</soapenv:Detail>
</soapenv:Fault>
Has anybody an idea/hint what this GSSException means and how to handle it?
On server side (Globus version 4.0.8) we checked the container.log. There is
(obviously) no entry for that. I also activated the GRAM debug modus
(log4j.category.org.globus.exec=DEBUG in
$GLOBUS_HOME/container-log4j.properties), but I did not found such an entry.
How can I get more detailed information about the gram process on server
side?
This error happens after several similar jobs were successfully conducted,
so it should be a temporary problem. Can (a temporary) dns resolving problem
causes this?
Regards,
Benjamin
--
Benjamin L?hnhardt
UNIVERSIT?TSMEDIZIN G?TTINGEN
GEORG-AUGUST-UNIVERSIT?T
Abteilung Medizinische Informatik
Robert-Koch-Stra?e 40
37075 G?ttingen
Briefpost 37099 G?ttingen
Telefon +49-551 / 39-22842
[email protected]
www.mi.med.uni-goettingen.de
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
