I ran into an interesting problem last week… I had a group of 15 Aruba IAPs essentially loose their configuration and revert to the factory default configuration. The issue occurred after a power outage so that gave me a few hints as to what might have occurred.
After some troubleshooting and sleuthing I discovered that there there was a 16’th IAP that was connected to the network but wasn’t able to obtain an IP address because the DHCP scope had been exhausted. When the network switches and routers were rebooted this IAP was able to obtain an IP address and because it had the higher MAC address it became the Master Virtucal Controller and the remaining 15 IAPs all ended up downgrading themselves and adopting the configuration of the 16th IAP (a factory default configuration).
That left me with the question of how do the Aruba Instant APs election of the Virtual Controller Master actually work?
As documented in the Aruba Airheads Community pages;
- look into IP scope. IAP with default-ip cannot win
- look into 3G capability
- look into AP class, Arran (IAP-13x) gets higher priority
- look into uptime, older AP gets higher priority
- look into AP MAC, bigger address gets higher priority
Since all the IAPs had just been rebooted (power outage) it was left to the IAP with the “biggest” MAC address and that just turned out to be the IAP that was originally unable to connect to the VC because it wasn’t able to obtain an IP address.
Puzzle solved.
Cheers!
Reference;
How is master IAP elected in an IAP cluster?
Hey Michael, you should always have a master IAP statically configured in order to get some control about this situations.
When the VC comes up, the master static config always overrides all the others requeriments.
I use a the best and more physical controlled/accesible IAP on the VC to be the master IAP.
dont be worried about it, if it is lost this master the second better one (best plataform IAP avaliable will be the master) Tested several times.
the Worst thing can happen if the network has flaps, sometimes the VC splits in two or three parts, but when the network achieve stability againg, the VC rebuilt itself. (you can also help to this, doing a full reboot of all aps -poe shutdown)
Thanks for the comment Adolfo.
We already set the VC IP address… in short it’s a virtual IP that gets moved to the VC IAP.
You can check out the configuration in this post – https://blog.michaelfmcnamara.com/2017/03/aruba-instant-access-point-example-configuration-walkthrough/
Cheers!