Home » Infrastructure » Unix » Veritas Cluster Server
Veritas Cluster Server [message #184004] Mon, 24 July 2006 16:29 Go to next message
toshidas2000
Messages: 120
Registered: November 2005
Senior Member
How do we fail the crontab, once the database fail to a different box. Crontab should copy to failover server??

Report message to a moderator

Re: Veritas Cluster Server [message #195654 is a reply to message #184004] Sat, 30 September 2006 06:12 Go to previous messageGo to next message
Frank Naude
Messages: 4581
Registered: April 1998
Senior Member
Use an external scheduler like Control-M. If you cannot, you will have to write some nifty shell script to do it for you.

Report message to a moderator

Re: Veritas Cluster Server [message #204041 is a reply to message #184004] Fri, 17 November 2006 08:01 Go to previous message
henderd
Messages: 6
Registered: November 2006
Junior Member
I'll assume you are running active/passive in this case.

What I do is run a test in the crontab line to see if a filesystem is present on the node. If say /u02 is only visible on one node at a time then you would have an entry like this in the crontab:

0 0 1 * * test -d /u02/oradata && /path/to/script/script.sh

So if the direcory /u02/oradata is not there then the script won't run.

You can't test for /u02 as even if the fs is mounted on the other node the /u02 directory will still be there

Means you need both crontabs kept up to date but it works.

HTH

Report message to a moderator

Previous Topic: Need HP - Unix documentation
Next Topic: runInstaller : cannot execute
Goto Forum:
  

[ Syndicate this forum (XML) ] [ RSS ]

Current Time: Mon Nov 25 02:49:52 CST 2024
.:: Forum Home :: Blogger Home :: Wiki Home :: Contact :: Privacy ::.