Hi Peter,
The messages you see below just means that the GIIS monitor could not find
those DN entries that should normally be present.
Dn: GlueCEUniqueID=...
should be present if your CE info is being properly published.
DN: 'dn: GlueCESEBindGroupCEUniqueID=...
should be present to help find the close SE for a given CE.
When I query your site GIIS, I get only your siteinfo. Are your CE and SE
GRIS functioning properly?
For 2_3_0, you can also use a BDII on port 2170 instead of the GIIS. The
query would be:
ldapsearch -x -H ldap://lunegw.lancs.ac.uk:2170 -b o=grid.
When I do this query, I get information other grid sites as well.
This could be caused by setting "auto" update to yes.
You can send in the relevant configuration and we can try to figure out the
problem.
Best Regards,
Min
-----Original Message-----
From: LHC Computer Grid - Rollout [mailto:[log in to unmask]] On
Behalf Of Peter Love
Sent: Thursday, December 16, 2004 1:20 PM
To: [log in to unmask]
Subject: [LCG-ROLLOUT] 2_3_0 upgrade, info system
At LANCS, we have 1 problem and 1 tiny bug after the 2_3_0 upgrade.
Alert from the GOC GIIS monitor:
DN: 'dn: GlueCEUniqueID='
'GlueCEAccessControlBaseRule: (VO:.+|/.+CN=.+)' ()
DN: 'dn: GlueCESEBindGroupCEUniqueID='
'GlueCESEBindGroupCEUniqueID: .+' ()
'GlueCESEBindGroupSEUniqueID: .+' ()
This isn't mentioned in the GOC wiki, and I'm not sure the problems are
relevant
for 2_3_0 with the new GIP system described in the release notes. A
short summary of the info system in 2_3_0 would be useful. I'm currently
taking this webpage as most relevant:
http://lfield.home.cern.ch/lfield/trouble.html
I assume this is a minor bug in redhat73-cfg.h and should be commented
out:
/* lcfgrpms is only for certification TB since we have not enough
space on lcfgng server for repository */
nfsmount.nfsdetails_lcfgrpms /export/local/linux/7.3/RPMS
lxshare086d.cern.ch:/shift/lxshare086d/data05/lcfgng.lxshare0204/linux/7.3/R
PMS \ ro,nfsvers=2,nolock
Peter
--
Lancaster University
|