Hi,
I observe strange behavior of the infosystem by cream ce with SGE.
there is one queue and about 6 VOs supported.
doing an ldapsearch I could see more or less correct values for running
and waiting jobs according to VO:
[root@gridka24 ~]# ldapsearch -xLLL -b o=grid -p 2170 -h
cream-ge-4-kit.gridka.de | grep GlueCEStateRunningJobs
GlueCEStateRunningJobs: 628
GlueCEStateRunningJobs: 0
GlueCEStateRunningJobs: 0
GlueCEStateRunningJobs: 0
GlueCEStateRunningJobs: 0
GlueCEStateRunningJobs: 0
GlueCEStateRunningJobs: 0
GlueCEStateRunningJobs: 0
GlueCEStateRunningJobs: 174
but if I run an lcg-infosites for this host, I always see same values
for each VO:
# lcg-infosites --is cream-ge-4-kit.gridka.de --vo alice ce
# CPU Free Total Jobs Running Waiting ComputingElement
----------------------------------------------------------------
0 1148 628 628 0
cream-ge-4-kit.gridka.de:8443/cream-sge-core1
# lcg-infosites --is cream-ge-4-kit.gridka.de --vo atlas ce
# CPU Free Total Jobs Running Waiting ComputingElement
----------------------------------------------------------------
0 1148 628 628 0
cream-ge-4-kit.gridka.de:8443/cream-sge-core1
# lcg-infosites --is cream-ge-4-kit.gridka.de --vo cms ce
# CPU Free Total Jobs Running Waiting ComputingElement
----------------------------------------------------------------
0 1148 628 628 0
cream-ge-4-kit.gridka.de:8443/cream-sge-core1
where there some major changes in infosystem now?
running
/usr/libexec/glite-info-dynamic-sge --lrmsinfo -c /etc/lrms/scheduler.conf
on cream ce gibs also normal results.
regards
Dimitri
--
Dimitri Nilsen, Dipl.-Ing(FH)
Karlsruhe Institute of Technology (KIT)
Steinbuch Centre for Computing
Kaiserstr. 12
76131 Karlsruhe, Germany
Tel.: +49 721 608 28607
Email: [log in to unmask]
|