ArubaOS and Controllers

Reply
Contributor I
Posts: 33
Registered: ‎04-12-2007

How to config a Persistent LMS?

Hey guys,

I will work with a customer the next few days who wants me to configure backup LMS in their HQ for several branches. However, all the controllers work as Masters and as far as I know there would be no problem to keep them working that way as far as the backup LMS controller has the appropiate config and licensing to receive the branch APs. Am I wrong?

But as I know that networks in their branch offices are not as reliable as I would want them to be. I'm looking forward to have tons of support calls regarding APs are working on their Backup LMS instead of the LMS.

Is there a good practice to keep APs working with their LMS as long as they can? What we don't want is to have hoping APs all the time.


Regards,

Erick Müller
Aruba Employee
Posts: 509
Registered: ‎07-03-2008

Re: How to config a Persistent LMS?

You could increase the bootstrap threshold in the AP system profile to the max of 65535. Heartbeats are sent every second, so that would allow about 18 hours of missed heartbeats to the LMS before flipping to the backup LMS. I think you may also need to match that number in the max-request-retries field as well.

Problem with that is, if the LMS is truly down, the AP won't go to the backup LMS for over 18 hours, so you'll want to think about using a more practical time.

Also, if an AP does go to its backup LMS, leave preemption off and it won't go back to the primary LMS until you manually take action. This will help with AP flapping.
Contributor I
Posts: 33
Registered: ‎04-12-2007

Re: How to config a Persistent LMS?

Yes I should think in a more practical time, hehehe. Thank you very much Mike


You could increase the bootstrap threshold in the AP system profile to the max of 65535. Heartbeats are sent every second, so that would allow about 18 hours of missed heartbeats to the LMS before flipping to the backup LMS. I think you may also need to match that number in the max-request-retries field as well.

Problem with that is, if the LMS is truly down, the AP won't go to the backup LMS for over 18 hours, so you'll want to think about using a more practical time.

Also, if an AP does go to its backup LMS, leave preemption off and it won't go back to the primary LMS until you manually take action. This will help with AP flapping.


Search Airheads
Showing results for 
Search instead for 
Did you mean: