02-15-2013 08:53 AM
I have a Master-Local setup. LMS is configured such that the local is primary and master is backup. When testing a local failure, all of my APs roll to the backup LMS as planned. Outside of rebooting the master, how can I force the APs to terminate back to the backup? Is it as simple as re-enabling the preemption temporarily? I don't want to always have preemption enabled in the event a controller is in the process of failing, and reboots itself every so often.
Solved! Go to Solution.
02-15-2013 09:04 AM
(Aruba650) (AP system profile "test") #lms-h?
lms-hold-down-period Amount of time AP needs to stay on backup LMS before
switching to primary. Range: 1-3600. Default: 600.
This stops the AP's from switching back during a bootup.
The other way of doing this (if the master and local can be on the same subnet) is to use VRRP rather than bkup-lms-ip, we can set a delay before VRRP wil preempt.
(Aruba650) (config) #vrrp 1
(Aruba650) (config-vrrp)#preempt delay ?
<seconds> Delay in seconds (0-60)
So we can wait for the controller to come up before it takes over again + there is interface tracking etc.
02-18-2013 04:40 PM
So after 600 seconds my APs will automatically attempt to reconnect to the primary, which is fine but if my primary LMS isn't healthy the APs will end up back at the backup LMS. I'd rather be able to specifically have them migrate when I know that the primary LMS is healthy. Possibly a blended approach would work as well, auto reterminate at the primary and then if another failure within x hours, wait for the manual command.
Until then, we'll go with the lms-hold-down-period