Remote Networking

Reply
Occasional Contributor II
Posts: 11
Registered: ‎08-13-2009

All RAPs Down when failed over to Secondary Controller

We upgraded to version 6.0.0.1 yesterday from version 5.0.0.1. When we rebooted the primary controller, the secondary controller took over. When that happened, all of the RAPs went down and wouldn't come back up until we rebooted the secondary controller and made the primary controller the master again.

While the Secondary controller was in control, I looked at monitor->access points->IPsec down. All of the RAPS had the IP address of the Master controller listed as their "Switch IP" rather than the VRRP address. When we moved them back over to the original Primary, they came back up with the VRRP address listed (correctly) as their "Switch IP".

Our initial set up for a rap is to give it a public address to connect to, x.x.0.1 which has a static route set up on our PIX that points to the VRRP private address of the controllers:

static (inside,outside) x.x.0.1 172.30.161.3 netmask 255.255.255.255.

172.30.161.3 is our VRRP address, 172.30.161.1 is our primary and 172.30.161.2 is our secondary.

I can't figure out why, when the secondary controller takes over the RAPS are still looking for 172.30.161.1, considering we've never told them to use that address, we've used the vrrp address for everything.

Thoughts?
Aruba Employee
Posts: 99
Registered: ‎09-08-2010

Re: All RAPs Down when failed over to Secondary Controller

Does the RAP AP Group(s) contain a unique AP System Profile, or is it default? Does this AP System Profile point to specific LMS and Backup LMS IP Addresses?
Occasional Contributor II
Posts: 11
Registered: ‎08-13-2009

Re: All RAPs Down when failed over to Secondary Controller

The RAPs do have their own AP system Profile and no, they don't have either of the LMS IPs filled in. We actually had no less than 4 TAC engineers on the phone for like 3 hours setting this up for us because we were mentally deficient in the RAP department (heh) and no one ever mentioned putting it in there. Do you think that's our problem?
Guru Elite
Posts: 21,010
Registered: ‎03-29-2007

Re: All RAPs Down when failed over to Secondary Controller

Steph,

the VRRP will NOT survive a translation through a firewall (if that is what you have). What you can do:

Create a 1:1 translation for the loopback or literal address of the primary controller. Create a 1:1 translation for the loopback or literal address of the secondary controller. Create a DNS a-record that has two addresses externally (the two external addresses for the translations you just created). Provision the RAP to the dns a-record and it will try the first, then the second if the first is down.


Colin Joseph
Aruba Customer Engineering

Looking for an Answer? Search the Community Knowledge Base Here: Community Knowledge Base

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