Wireless Access

Have a 620 with 6.1.3.0 code. Have a AP connected ont eh g1/0 interface, but getting the following message:

AP rebooted Fri Dec 31 16:04:25 PST 1999; SAPD: Unable to contact switch. Called by sapd_hello_cb:4

That would indicate a network error, but the AP is getting an IP and knows the master IP address. The controller can ping the AP also. Monitoring shows the status as up, but can't provision and AP keeps rebooting. Any thoughts an why the AP will not provision?

This can happen if you accidentally have Vlan 1 (or the lowest Vlan number) as a non-routable ip address.  type "show controller-ip" to see what the controller-ip is.  No matter what Vlan ip address the AP contacts the controller on initially, it gets re-routed to the controller-ip and if it is non routable, you will get that symptom.  Find out what the output of that command is, first.

The address is the ip of the controller on vlan 4, which is the same vlan as the AP. I have other controllers with the same setup with no issues?

(Controller) #show controller-ip

Switch IP Address: #.#.4.18

Switch IP is configured to be Vlan Interface: 4

Switch IPv6 address is not configured.

Can you get into the console of the AP to see what is going on?

Yes, was able to fix. Had some profile issues. Thanks!

What were the profile issues?  I'm having similar issues, but the AP provisions for a little while then goes down and I must reboot the controller and other such silly methods.   I'm pretty green to the world of Aruba and I'm working on getting things sorted out...

When did this start happening or did it ever work?

It did work!  In fact when we were on older ArubaOS code (2.4.8.14) everything worked fine.  Since then, due to suits giving pressure in the deadline department, we decided to skip submitting the code to Aruba for revision before OS upgrades and now we are on 3.4.4.0 and some of the code is preserved, but the 1 AP we have connected for just testing right now will not stay up.  I understand the code should have been submitted to Aruba for revision but at this point I'm pondering a full wipe...  Any help is greatly appreciated.

Get on the commandline in enable mode and type "show log system 50" to see if you can see anything that is happening with that AP.

I will paste what the output was below the break.  I still have not read the output yet, but I'll work on that once I post this...

----------------------------------------------

Aug 1 08:45:21 :315382: <CRIT> |fpapps| Reboot Cause: User reboot.
Aug 1 08:45:28 KERNEL: klogd started: BusyBox v0.60.4 (2011.04.05-22:26+0000)
Aug 1 08:45:28 KERNEL:
Aug 1 08:45:28 KERNEL: Aruba Networks Model A5xxx
Aug 1 08:45:28 KERNEL: ArubaOS Version 3.4.4.0-FIPS (build 27739 / label #27739)
Aug 1 08:45:28 KERNEL: Built by p4build@stan on 2011-04-05 at 15:34:51 PDT (gcc version 3.4.1)
Aug 1 08:45:28 KERNEL: Copyright (c) 2002-2011, Aruba Networks, Inc.
Aug 1 08:45:28 KERNEL: i2c-dev.o: Registered 'System I2C Interface' as minor 0
Aug 1 08:45:28 KERNEL: VFS: Disk change detected on device ide0(3,3)
Aug 1 08:46:14 :399801: <ERRS> |cts| An internal system error has occurred at file sapi.c function sapi_rsp line 247.
Aug 1 08:46:17 :313078: <ERRS> |fpapps| XSec Vlan Interface 340 not found line 877
Aug 1 08:46:17 :313078: <ERRS> |fpapps| XSec Vlan Interface 340 not found line 877
Aug 1 08:46:17 :313078: <ERRS> |fpapps| XSec Vlan Interface 4095 not found line 877
Aug 1 08:46:17 :313078: <ERRS> |fpapps| XSec Vlan Interface 4095 not found line 877
Aug 1 08:46:17 :313256: <WARN> |fpapps| Route resolve returned an Error
Aug 1 08:46:19 :399801: <ERRS> |pim| An internal system error has occurred at file sapi.c function sapi_rsp line 247.
Aug 1 08:46:26 :301406: <ERRS> |snmp| SNMP agent timed out when sending a request to application Station Management Low Priority for object 202
Aug 1 08:46:31 :325022: <ERRS> |authmgr| Bogus VLAN ID:4095 received.
Aug 1 08:46:32 :301406: <ERRS> |snmp| SNMP agent timed out when sending a request to application Station Management Low Priority for object 202
Aug 1 08:46:42 :304001: <ERRS> |stm| Unexpected stm (Station managment) runtime error at data_path_handler, 579, data_path_handler: recv - Network is down
Aug 1 08:46:45 :301406: <ERRS> |snmp| SNMP agent timed out when sending a request to application STM for object 602
Aug 1 08:46:49 :301406: <ERRS> |snmp| SNMP agent timed out when sending a request to application WMS for object 725

Hmm... Looks like you have a FIPS version of ArubaOS.  What model access point is this, and how do access points discover the controller?

That is correct.

The one and only controller is an Aruba 6000 FIPS (even thought the OS says Aruba 5XXX) with only 1 AP-70 that should be discovering via the AP discovery protocol.  Sorry working off what little knowledge I have about Aruba right now.

Is the AP directly connected to gig1/0?  If not, is there a firewall or packet filtering ACL between the AP and controller that could be allowing ICMP but blocking other ports and protcols that might be needed?