03-19-2015 08:29 AM
Last night, we were performing a clean-up on a site which is in a two controller Master-standby setup.
We ran into a few issues when doing these controllers. When we added the loopback to the standby controller it was flagged to reboot. When we tested this in the lab we did not have to do this. I think the reason is there was no controller-ip command set or it was set to the loopback, but no loop back was assigned.
We ran into this on the primary controller as well but the odd thing was before we were able to do this, there was on old entry for a local controller. We did not remember this being set up as a master –local but there was a local ip setting on the controller, no big deal. We have removed these before (actually last week at three sites without issue). When we generated the command to remove localip 0.0.0.0, the controller flaked out and became a local. (show role info) At that point everything was working but in a local –standby config. We could not make any changes.
We failed everything over to the standby. (no user impact) Then added the loopback and was forced to reload the controller. After thinking about this, it appears the controller-ip was assigned to the loopback but there was no loop back ip address so when we added the loopback address the controller ip was essentially changed which required a reboot.
After the reboot, we cleaned everything back up and the site is fully restored operating under the new group.
We have two SC-3 controllers here that are attached Port Channel to two Alcatle 7450 ESS-7 SA routers.