Network Management

We have been running few small sites with Aruba instant clusters. We just installed AirWave so we could get better monitoring for our networks.

We installed the AirWave and Instant virtual controllers and access points all connected to AirWave without issues.

However every device is showing status up and configuration Unknown. Also no data from clients or anything else is coming to AirWave server.

AirWave is on DMZ and ports HTTPS (443) and AMON (8211) is opened from sites to DMZ. Do we need open more ports for just monitoring? Or is the problem something else?

Do we need to open any ports from AirWave server to sites where Instant cluster resides?

You should only need port 443 inbounds.  Did your devices show up in "New Devices" first?  If yes, what did you do after that?  If you moved them to a folder, you should be able to see users on it, at minimum.

Did you also read the guide here?  https://support.arubanetworks.com/Documentation/tabid/77/DMXModule/512/Command/Core_Download/Default.aspx?EntryId=14106

Yes the devices showed on the "New devices", we then moved them to folder for that site.

If you do a tcpdump of https traffic, are you seeing the IAPs contacting the AMP?  Can the AirWave server ping the IPs of the IAPs? (and same for the reverse from the IAPs to AirWave).  You might also check /var/log/pound to make sure that the pound service is catching the IAP traffic.

Next steps after that would be to open a support case, and possibly enable additional logging using:

# qlog enable swarm_debug

This logs output into /var/log/amp_diag/swarm_debug

We have only opened https and amon ports, so can't ping the devices from AirWave or vice versa. But if I open ICMP port then offcourse ping works.

Looks like you'll probably want to enable the swarm_debug logging from the AMP, and then send the output to support so that they can help analyze and troubleshoot.

Now it shows configuration Good and I can see clients from different sites. Maybe it just took some time, but it is working now. 

That's good news.  It's possible that the Async Logger Client processes got held up and eventually restarted themselves.