Control-M server Network Status = Reconnecting

Diablau · Post by **Diablau** » 13 Aug 2009 2:56

Hello all!
We've cleared the AJF yesterday. Our Ctrl-M server is a cluster so we've used the procedure to clear the AJF with a cluster as server. I don't know if it is the reason for what's happening, but mention it just in case. This morning we've been surprised to see that our ctrl-m server network status is reconnecting (but never reconnect...). If I connect on it, the menu works and when I make a Ping from server to agent it fails, the reverse Ping works.
Any ideas?
Thanks!

Diablau · Post by **Diablau** » 13 Aug 2009 3:41

Seems to be cluster instability...

philmalmaison · Post by **philmalmaison** » 13 Aug 2009 7:32

do you mean ctm_menu ?
if you can't ping your agent, and as you can before, it seems to be a network configuration that can have been changed by somebody ?
check all necessary system file, /etc/... on unix
use network comand to trace the communication beetween twice
(traceroute,ifconfig -a,netstat -na | grep "<port_number>")
and the date of any change of system files

regards
philmalmaison

Diablau · Post by **Diablau** » 13 Aug 2009 9:23

Yes, I mean ctm_menu. In fact, we've been able to get the connection back, but suddenly it has failed again... When we give the control to the other server (remember they're clusters) the ctrl-m server suddenly becomes available.???
Do you speak French philmalmaison? Because I do, and maybe it will be clearer in French???

philmalmaison · Post by **philmalmaison** » 14 Aug 2009 9:46

comment un cluster peut il être instable ?
reprends moi si je me trompe, un cluster dans ton cas qu'estce que c'est ?
2 machines physiques, se basculant en cas de failover du noeud actif, les file system control-m nécessaire au redémarrage sur le nouveau serveur actif non ?
alors il faut sur la 2 eme machine la même configuration securité, network pour joindre les agents intervenants et connection à l'EM ...
Les mêmes ports de communication ...
etc ...
quelle base de donnée ?
fichier interface si sybase ?
quelle cluster ?

regards
philmalmaison

Zakeer · Post by **Zakeer** » 17 Aug 2009 6:05

Hi,

I think it should be the network issue between the Control-M Server and EM try to bring down the Control-M server using CTM_menu and make sure none of the process are running if so, then kill those process and start the server if the issue still persists try to restart the EM server using root_menu.

Hope it helps

Zakeer

Diablau · Post by **Diablau** » 17 Aug 2009 8:12

Thank you all! Merci beaucoup!

I'll see what I can do, and if I find the reason why, I'll let you know. The problem is that it is stable now and still last Friday... Also there isn't any details about these incidents in the EventViewer... Anyways, if it happen again we'll do some tests...