On Fri, 19 Feb 2010 19:47:40 +0100
Maarten Litmaath wrote:
> Hola Arnau,
Hi Maarten,
[...]
>
> That is because there are stale "slapd" processes:
[...]
>
> And CREAM ce08 will have the same problem. So, what happened on Jan
> 22?
Mmmm.. lcg-CEs had some ldap processes belonging to user 100. On that
day I changed edguser uid, so, after changing it and reconfig, ldap
restart was unable to kill preivous process...
ce08 did show 45? (I don't have my own mail here). ce08 had 2 ldap
processess belonginf to edguser ... so, it should show 34 always...
Anyway, I've killed old proc in old CEs. and reconf... let's see if
tomorrow publication is still fine. Now is ok!
[arnaubria@ui02 ~]$ for a in ce05 ce06 ce07 ce08; do ldapsearch -x -h $a.pic.es -p 2170 -b mds-vo-name=resource,o=grid|grep GlueHostApplicationSoftwareRunTimeEnvironment|grep cms> $a; done;wc -l ce0*
34 ce05
34 ce06
34 ce07
34 ce08
As, always, thanks Maarten,
Arnau
|