Global Mint

[Phil_storey]

Hi all and Happy New Year

I have been trying to figure out why 2 AP's that service clients in a remote building are not showing up in the RFS view, I came across some information about Mint. So from the controller I did the ping

ping 172.xxx.xxx.xxx dont-fragment size 1500

PING 172.xxx.xxx.xxx (172.xxx.xxx.xxx) 1500(1528) bytes of data.
From 172.xxx.xxx.xxx ( contoller ) icmp_seq=1 Frag needed and DF set (mtu = 1500)
From 172.xxx.xxx.xxx ( contoller ) icmp_seq=1 Frag needed and DF set (mtu = 1500)
From 172.xxx.xxx.xxx ( contoller ) icmp_seq=1 Frag needed and DF set (mtu = 1500)
From 172.xxx.xxx.xxx ( contoller ) icmp_seq=1 Frag needed and DF set (mtu = 1500)
From 172.xxx.xxx.xxx ( contoller ) icmp_seq=1 Frag needed and DF set (mtu = 1500)
0 packets transmitted, 0 received, +5 errors

so as per the guide I dropped it by 8 each time until I got a response

ping 172.xxx.xxx.xxx dont-fragment size 1472
PING 172.xxx.xxx.xxx (172.xxx.xxx.xxx) 1472(1500) bytes of data.
1480 bytes from 172.xxx.xxx.xxx: icmp_seq=1 ttl=64 time=2.10 ms
1480 bytes from 172.xxx.xxx.xxx: icmp_seq=2 ttl=64 time=6.91 ms
1480 bytes from 172.xxx.xxx.xxx: icmp_seq=3 ttl=64 time=4.26 ms
1480 bytes from 172.xxx.xxx.xxx: icmp_seq=4 ttl=64 time=3.18 ms
1480 bytes from 172.xxx.xxx.xxx: icmp_seq=5 ttl=64 time=3.51 ms
--- 172.xxx.xxx.xxx ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 4002ms
rtt min/avg/max/mdev = 2.103/3.996/6.919/1.619 ms

The Client AP is at the end of a wireless Bridge that utilises 2 AP 7532, The bridge shows up in the mesh point view under stats.

I connected to one of the AP's in the remote building and had a play around with the MTU settings the AP briefly appeared in the GUI view on the RFS.

So Should I change the Global Mint MTU size ? the Guide I was looking at is this one
https://extremeportal.force.com/ExtrArticleDetail?an=000089595

but the command to alter the Mint MTU does not see to work for wing 5.8

Also the mint ping does not seem to work in wing 5.8

I'm not sure if the VLANS ( Informations ) are carried in the MINT protocol across the wireless bridge as VLAN 1 is working fine and clients are getting IP addresses but VLAN 10 which has and ADSL router on it to give out IP addresses fail to connect, this is why I was looking at the MINT protocol.

But If I give a device a static IP in the range of the ADSL router DHCP server they work. The Network Switches that the AP's connect to are AVAYA ERS500xx and I did read that there was a bug in the firmware that was malforming the DHCP request from the DHCP servers/devices. So to ensure this was not the root cause the IP DHCP-Snooping was disabled.as per another guide from AVAYA

I may be looking in completely the wrong place but any advise is gratefully received

[Phil_storey]

Spoke too soon one of the AP's the one I swopped out has disappeared from the RFS GUI , On The AP7532's I'm seeing Dot11i handshake timeouts for anything trying to connect to VLAN10

[Phil_storey]

Hi Chris
I have Both AP's now checking in to the RFS, I ended up adding the controllers to the AP's under controller adoption. I also created a new profile for these two AP's only in the RFS and under adoption I set it to pool 1 routing level 1 and added the host ( RFS )
This seems to have worked,
Over in the remote building I was able to get an IP on VLAN 1 which has never been an issue, and on VLAN 10 which has been the issue, But this was only on the AP on the first floor, So I connected to the AP's in the remote building as they were getting IP addresses and compared the setting, on the AP I could not get an IP on VLAN 10 under Access Point it was showing no information - i.e Name /location Version etc but the one that was working did. So I have swopped the AP out and then tried to connect to VLAN 10 and I got an IP, So Just waiting on users turning up now and seeing if the MU's will connect to Vlan 10.

[ckelly]

Regarding the tree view having the appearing/dissapearing issue....this tells me that there's an RF Domain Manager election issue. This is what you'll see happen in the tree.

How are the APs adopted? Level-1MiNT? Level-2 MiNT?
The fact that you are tunneling traffic tells me either your APs are level-1 MiNT adopted or you are tunneling over Level-2 MiNT...so I can't tell.

If the APs are level-1 MiNT adopted, then the controller should be the RF Domain Manger, in which case you shouldn't be seeing the weird issue withe the APs in the tree.

If the APs are level-2 MiNT adopted, then you need to make sure that you have defined the Control-VLAN for that RF Domain (configuration for the Control VLAN is in the RF Domain setup). If you don't, that tree issue will occur.

[Phil_storey]

Hi Chris
this gets stranger by the second, So I have changed the MTU value, in the tree view one of the AP's is now showing but not the other , If I connect to the AP and look at the events, I can see devices trying to connect to the VLAN1 & VLAN10, but failing with dot11i handshake timeouts and association denied invalid MDIE
crazy