Since Thursday early evening when half my worker nodes were switched off
with out warning due to air con problems my cluster as reported by
http://goc.grid.sinica.edu.tw/gstat/t2ce01.physics.ox.ac.uk/GIISQuery_Us
age_cpu_.html
has been pretty dodgy.
I have now marked the missing nodes as offline on the ce with the
pbsnodes -o command.
qstat commands seems to take longer than normal to complete.
chkconfig --list on the ce reveals :
[root@t2ce01 root]# chkconfig --list|grep globu
globus-gatekeeper 0:off 1:off 2:on 3:on 4:on 5:on
6:off
globus-gridftp 0:off 1:off 2:off 3:off 4:off 5:off 6:off
globus-mds 0:off 1:off 2:off 3:off 4:off 5:off 6:off
Should mds be running on the ce?
Any suggestions of what to check.
Thanks Pete
----------------------------------------------------------------------
Peter Gronbech Unix Systems Manager Tel No. : 01865 273389
Department of Particle Physics Fax No. : 01865 273418
University of Oxford,
Keble Road, Oxford OX1 3RH, UK E-mail : [log in to unmask]
----------------------------------------------------------------------
-----Original Message-----
From: NOVAK Judit [mailto:[log in to unmask]]
Sent: 27 November 2004 01:25
To: [log in to unmask]
Cc: SA1 Followup Mailing List
Subject: GIIS at OXFORD-01-LCG2
Dear Site Administrators,
GIIS doesn't publish Computing Element.
Could you have a look at it, please?
Thanks in advance,
Judit Novak
CERN, IT-GD-GIS
|