12-28-2010 05:19 PM
Hello All- Looking for a way to get notification of a supervisor failover. We have 6500s with Sup 720s, and 4500s with Sup II+ and Sup V. Any suggestions are welcome.
Many thanx
Solved! Go to Solution.
12-28-2010 08:54 PM
For an external script interacting with network devices, I prefer using Expect (http://www.nist.gov/el/msid/expect.cfm), but Perl with the relevant CPAN modules would also work well.
EEM could be a little more preferrable than an external script, since it lives right on the IOS device itself.
http://www.cisco.com/en/US/products/ps6815/products_ios_protocol_group_home.htm
There're several EEM experts on this forum, so in case they miss this thread, you can always start a new one with a list of your requirements, and usually you'll get a tailored solution well nigh unobtainable through regular support channels.
12-28-2010 07:51 PM
Hello Michael,
Below post talks about failover notification
https://supportforums.cisco.com/message/3207478
Regards
mahesh
12-28-2010 08:09 PM
Thanx Mahesh- The article talks about CatOS and we are running IOS. Since we do have some SNMP traps enabled already, I tried the command "snmp-server enable traps redundancy" -- redundancy is not one of the available keywords.
12-28-2010 09:01 PM
For cat6k running IOS, you can configure "snmp-server enable traps module" in the case of the SUP going cleanly "up" or "down".
You can also poll for moduleStatus and moduleStandbyStatus of the CISCO-STACK-MIB, as Joe discussed in https://supportforums.cisco.com/message/3053835
Thirdly, IOS generates a hodge-podge of syslog messages in different severity levels for possible SUP issues. The problem is as a customer, I don't have access to IOS/CatOS source codes, so I have no way of knowing all the current and future key words Cisco chooses to use, concerning a subject such as SUP failures.
Even after doing all three above, IME I could still miss out some corner cases for SUP going bad. That's why I prefer the "show redundancy state" in a script, internal or external to IOS. Besides, the scripting output will be comparing apple-to-apple with what the network staff look at.
12-28-2010 08:07 PM
The problem is SUP failovers could happen for a variety of reasons and in a variety of "styles". Worse, the secondary SUP could go kaputt, something that usually goes unnoticed until there's a need to fail over to it, autonomously or manually. So I think the surest way to check redundant SUP status, including timely notification of secondary SUP failures, is to have either an external script or EEM applet/policy parse the output of "show redundancy state" periodically.
12-28-2010 08:14 PM
Thanx Bear, the more I look, the more this seems the solution. Would you have any suggestions as to scripting methodology? And what is EEM?
12-28-2010 08:54 PM
For an external script interacting with network devices, I prefer using Expect (http://www.nist.gov/el/msid/expect.cfm), but Perl with the relevant CPAN modules would also work well.
EEM could be a little more preferrable than an external script, since it lives right on the IOS device itself.
http://www.cisco.com/en/US/products/ps6815/products_ios_protocol_group_home.htm
There're several EEM experts on this forum, so in case they miss this thread, you can always start a new one with a list of your requirements, and usually you'll get a tailored solution well nigh unobtainable through regular support channels.
12-28-2010 09:06 PM
Outstanding Bear, many thanx
And yet, you have opened up a whole new world which will suck down even more of my time...
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide