Hi everyone,
we are currently on the procedure of installing a fresh umd-2 cream + sbdii instance on HG-03-AUTH. The only supported VO at the moment is dteam. Within site-info.def file the line defining the acceptable fqans is the following one:
DTEAM_GROUP_ENABLE="dteam /dteam /dteam/ROLE=lcgadmin /dteam/ROLE=production"
What we are observing is that after yaim configuration the site bdii is publishing the appropriate information. However, within /var/log/bdii/bdii-update.log the following error is printed:
---
Traceback (most recent call last):
File "/usr/libexec/lcg-info-dynamic-scheduler", line 447, in <module>
ql2, vol2, dndictglue2 = parseGLUE2ComputingShareFile(config)
File "/usr/libexec/lcg-info-dynamic-scheduler", line 316, in parseGLUE2ComputingShareFile
vogrp = tmpl[2].strip()
IndexError: list index out of range
---
After some digging around it turns out that this error is generated when executing the following script (either as root or as ldap user on the node).
/usr/libexec/lcg-info-dynamic-scheduler -c /etc/lrms/scheduler.conf
Going further into the problem we see that the error is generated through an included file in /etc/lrms/scheduler.conf :
# cat /etc/lrms/scheduler.conf
[Main]
...
static_glue2_ldif_file_computingshare: /var/lib/bdii/gip/ldif/ComputingShare.ldif
…
It turns out that when the /var/lib/bdii/gip/ldif/ComputingShare.ldif file is generated some empty values are initialized.
We are not sure whether this should be expected (as I mentioned above information published seems perfectly OK). However, there seems to be a problem with the way the fqans are defined in the site-info.def. Any ideas/suggestions are more than welcome.
Thank you,
Paschalis
Scientific Computing Support Unit
--
Scientific Computing Center, Aristotle University of Thessaloniki
University Campus, GR 541 24, School of Sciences, Building 22A
Tel. +302310998988, Fax. +302310994309, http://www.grid.auth.gr
|