Wireless Access

I have a cluster, I built and added a third controller but when I add the controller to the cluster aps move to it but never come up. With in 10 minutes they move back to the original controller they were on.

During this, I never see clients connect and pass traffic on an ap that moved to the new controller.

I have even manually moved an ap to this controller without it being in the cluster. Same behavior, moves over, never shows an "up" status until it moves back to one of my original two controllers.

Any help appreciated!

Thanks

Have you executed any of the "show lc-cluster" commands  on the MD to see if that MD fully joined the cluster?

Yes, that is what is stumping me... Please see below

(homearmc01) [MDC] *#show lc-cluster group-membership

Cluster Enabled, Profile Name = "Home-Cluster"
Redundancy Mode On
Active Client Rebalance Threshold = 50%
Standby Client Rebalance Threshold = 75%
Unbalance Threshold = 5%
AP Load Balancing: Enabled
Active AP Rebalance Threshold = 50%
Active AP Unbalance Threshold = 5%
Active AP Rebalance AP Count = 30
Active AP Rebalance Timer = 1 minutes
Cluster Info Table
------------------
Type IPv4 Address Priority Connection-Type STATUS
---- --------------- -------- --------------- ------
peer 10.17.1.18 128 L2-Connected CONNECTED (Member, last HBT_RSP 27ms ago, RTD = 0.978 ms)
self 10.17.1.19 128 N/A CONNECTED (Member)
peer 10.17.1.20 128 L2-Connected CONNECTED (Leader, last HBT_RSP 87ms ago, RTD = 1.024 ms)
(homearmc01) [MDC] *#show lc-cluster vlan-probe
status Cluster VLAN Probe Status

(homearmc01) [MDC] *#show lc-cluster vlan-probe stat

Cluster VLAN Probe Status
-------------------------
Type IPv4 Address REQ-SENT REQ-FAIL ACK-SENT ACK-FAIL REQ-RCVD ACK-RCVD VLAN_FAIL CONN-TYPE START/STOP
---- --------------- -------- -------- -------- -------- -------- -------- --------- --------- ----------
peer 10.17.1.18 30 0 29 0 29 30 0 L2 Conn 0/ 0
peer 10.17.1.20 1 0 2 0 2 1 0 L2 Conn 0/ 0
(homearmc01) [MDC] *#

Do the access points in that AP group have an LMS-IP?

Yes, each controller in the cluster including the new one has a primary/backup LMS IP and it is pingable from the new controller via cli. See pic.

I keep thinking there must be something still needing manually configured that would not normally sync once joined to the cluster.... Not finding anything yet.

I just noticed in the MM dashboard it is not showing any uplink information on the new controller. Everything in the GUI looks to match the other controllers including interfaces, vlans... etc. I have confirmed the vm is configured the same as the other controllers in vsphere.

If you have a single cluster, you don't need a primary or backup LMS.  You should remove those ip addresses.  Access points, when they find a cluster automatically download all the ip addresses of the controllers in the nodelist.

The WAN uplink does not come into play in this scenario.

Did you add the Controller to the cluster in both the Cluster configuration and at the MD level of the controller?

Okay, I will remove the LMS IPs. Is it having a negative impact having them at the moment? Included screen shots shows what I have set regarding add the new controller into the cluster. I never see any download data transfer either, see pic.

If you have a single cluster, you should not have an LMS or backup LMS.  Removing them removes two variables.

Why do you have so many VLANs excluded?

I have removed the LMS settings and will retest, this environment was put in place a couple months before I started with the company this year by a third party. I'm not sure why exactly they have so many excluded.

They basically have every vlan in use excluded

That should not be.  It means that the controllers will not check for connectivity between each other for those clients which is not good.  I would remove every VLAN that each controller should be able to connect to.

Also, SSH into the controller that is not working and type "show log system 50" to see if there are any errors that are related to your issue.