Hi,
sorry, i forgot to explain, that the Networker client isn't running as a
package, just simply as client. So there are no control scripts or
.nsr_cluster entrances for NetWorker. There are running several other
packages on the server and if i trigger a save on a normel filesystem (ex:
/opt), there will be up to 10 times of "cmgetconf" for every configurated
package on the maschine before writing on tape starts. Don't know where the
sense of this behavior is.

Klaus

-----Ursprüngliche Nachricht-----
Von: Kalbus Johanes [mailto:Johanes.Kalbus AT itellium DOT com]
Gesendet am: Dienstag, 8. Oktober 2002 11:57
An: 'Legato NetWorker discussion'; Grohal Klaus
Betreff: AW: [Networker] Networker client in HP-UX cluster

Hello,

we had the same problem until we entered all cluster packages in the
/etc/cmcluster/.nsr_cluster file
All packages run on the cluster MUST be defined there, even those you don't
back up.
.nsr_cluster is a description of the cluster packages for networker. And if
you keep it up to date you could use
the saveset All again on your cluster clients

Each line is like:
clientname:IPofTheClient:/dir1:/dir2:/dir3

Here it helped - if you have cluster packages without own IPs you are out of
luck - support suggested to assign it an own IP even if it is not used -
that helped.

Johanes


-----Ursprüngliche Nachricht-----
Von: Klaus Grohal [mailto:klaus.grohal AT SIEMENS DOT COM]
Gesendet: Montag, 7. Oktober 2002 16:33
An: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Betreff: [Networker] Networker client in HP-UX cluster


hi spezialists,

we have a strange problem on several NetWorker Clients 6.1.2 running on HP-
UX.
Since we updated the clients (2 weeks ago), our syslog.log is flooded with
the message: /usr/sbin/cmgetconf -p <pakage-name>. We get these message
about 17.000 times in 12 hours. It is definitely invoked by NetWorker
6.1.2, because on one client we got back to 5.5.5 an the messages where
gone.
The second problem is: if we do a recover by multiple ssids (for example
recovering an oracle database which was backuped by nmo) it last over 1
hour before networker starts reading from the tapes. During this time, it
seems that NetWorker is collecting informations about the cluster (also a
lot of cmgetconf).

The NSR-Clients are running under MC/Service-Guard and backup works fine.

Any ideas ??

regards
Klaus

--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=