Wireless Access

Hi,

we have two controller Aruba3200, in Master-Local configuration.

Since few days one AP93 continuosly reboot: it gets an IP address from controller's DHCP server, start with correct configuration, but after a variable period of time (1 minute, 10, minute, 15 minute,....) it goes down and reboot. We have tried to associate the AP with bth the controller. In the log message I see this:

"nanny[834]: <303022> <WARN> |AP AP35-Vetro@192.168.150.98 nanny| Reboot Reason: No reboot message found."

I think it isn't a problem about configuration of AP because others APs with the same parameters don't have any problems. And also the AP works correctly for a time before rebooting.

Does anyone have some idea of the cause of this issue?

Can this be a bug of the AP?

Thanks,

Massimo

1. What ARUBAOS are you using?
2. Are u using CONTROL-PLANE SECURITY?
Do u have LMS / BACKUP LMS Configured in your ap-default system profile?
3.Are u sure your DHCP SERVER (that is on the controller) isn't giving the same address to other unit/user on site? (might happen if broadcast is blocked in some old ArubaOS versions)
4.Can u check if this AP is connected via 10Mbps 100Mbps or 1000Mbps connectivity? - you might an lan infrastructure issue. -> to be sure take this AP and connect it to a different port/location - if the problem stops - so u got infrastructure problems.

Hope i gave u some ideas on how to solve you issue. :smileywink:
Update me if further assistance needed.

Me.

Thanks for your suggestions kdisc98!

I answer to your questions here:

1. What ARUBAOS are you using?

The controllers have the 6.1.3.5 version

2. Are u using CONTROL-PLANE SECURITY?

Yes, i'm using CONTROL-PLANE SECURITY, but i don't think this is the problem because in the past 5 months it works fine. Also, the AP MAC Address is in the whitelist campus AP.

If the problem is CONTROL-PLANE SECURITY, i think that the AP doesn't go UP for a some minutes.

Do u have LMS / BACKUP LMS Configured in your ap-default system profile?

Yes, I configure LMS / BACKUP LMS in the ap system profile. Other APs in fact works fine, so I think this is not the problem

3.Are u sure your DHCP SERVER (that is on the controller) isn't giving the same address to other unit/user on site? (might happen if broadcast is blocked in some old ArubaOS versions)

It seems that the IP isn't duplicated. But i must verify this. Broadcast isn't blocked

4.Can u check if this AP is connected via 10Mbps 100Mbps or 1000Mbps connectivity? - you might an lan infrastructure issue. -> to be sure take this AP and connect it to a different port/location - if the problem stops - so u got infrastructure problems.

I monitor from remote site the switch on wich the AP is connected; there isn't any error on port where the AP is connecteded.

This afternoon I will go to the client and so I can verify more completly if the problem is in the lan infrastructure (even if I have some doubt on this solution...)

 I note that when the problematic AP goes UP in the LOCAL controller, if i execute the commmand "show ap database" in the CLI of the MASTER controller this AP isn't displayed in the list; is it normal?

In the LOCAL controller CLI, with the same command, it is displayed instead.

But if i execute "show switches", the controllers status is UPDATE SUCCESSFUL.

"I note that when the problematic AP goes UP in the LOCAL controller, if i execute the commmand "show ap database" in the CLI of the MASTER controller this AP isn't displayed in the list; is it normal?

In the LOCAL controller CLI, with the same command, it is displayed instead.

But if i execute "show switches", the controllers status is UPDATE SUCCESSFUL."

• IF your AP connecting to your local - and establisjh connecitivy in front of he Local you will not see it in the show ap databse. > GO VIA GUI to monitroing page and then to ALL ACCESS POINTS.

• It's Seems your AP losing connecitivy to your LMS and connecting to his BACKUP-LMS every once in a while.

• How does this AP finding is controller? by IP or by name resolving? (if it by name resoloving - check your DNS server also)

update me.

I just came back from client site. I assume that the problem is on the AP, probably an AP bug.In fact I connected throught console to the AP and when it rebooted  I saw this messages:

 # ar7240 data bus error: cause 0x9080001c
Data bus error, epc == c077a50c, ra == c077a514
Oops[#1]:
Cpu 0
$ 0   : 00000000 00000001 deadc0de b00080ec
$ 4   : 83dc0000 000080f0 00000000 80301de4
$ 8   : 83dc0000 1000001f 80690000 80690000
$12   : 80330000 80690000 80330000 80690000
$16   : 00000000 00000001 83de0000 83dc0000
$20   : 000bfc30 00000001 83dda860 00000000
$24   : 00000000 c075c710
$28   : 80300000 80301d80 c0751018 c077a514
Hi    : 000000e4
Lo    : 00000311
epc   : c077a50c ar5416AniArPoll+0xa0/0x570 [ath_hal]     Tainted: P
ra    : c077a514 ar5416AniArPoll+0xa8/0x570 [ath_hal]
Status: 1000d403    KERNEL EXL IE
Cause : 9080001c
PrId  : 00019374
Modules linked in: ath_pktlog wlan_wep wlan_tkip wlan_ccmp ath_pci ath_dev spectrum dfs ath_rate_atheros ath_hal wlan asap_mod drvlog_mod ag7240_mod
Process swapper (pid: 0, threadinfo=80300000, task=80302180)
Stack : 80690000 80333914 5149c84d 00000002 83dc21d0 80301de4 00000000 00000001
        83d80000 83dc0000 000bfc30 00000001 806971bc 00000000 83cb0280 c06b4354
        80301dd0 c0150000 80301e10 00000000 0000000a 80333930 00000000 80144928
        00000064 00000000 00000000 ffffffa3 00000000 1b2e09d0 83d8397c 83d83994
        80301e50 806969b4 806969b4 80330000 806971bc 00200200 802d6a40 c06b9b08
        ...
Raw Call Trace:
[<c06b4354>] ath_calibrate+0x59c/0x674 [ath_dev]
[<80144928>] ktime_get+0x18/0x40
[<c06b9b08>] _ath_internal_timer_handler+0xb4/0x1ac [ath_dev]
[<c06b9a54>] _ath_internal_timer_handler+0x0/0x1ac [ath_dev]
[<80133d6c>] run_timer_softirq+0x18c/0x1f8
[<8012f350>] __do_softirq+0x84/0x138
[<8012f470>] do_softirq+0x6c/0x74
[<8010a6b4>] do_IRQ+0x24/0x34
[<8010666c>] ar7240_interrupt_receive+0xec/0x100
[<801088c4>] r4k_wait+0x0/0xc
[<802c2188>] packet_poll+0x0/0x9c
[<8010aa68>] cpu_idle+0x38/0x50
[<801088c8>] r4k_wait+0x4/0xc
[<801d5104>] idr_cache_ctor+0x0/0xc
[<8037a7ec>] start_kernel+0x1d4/0x1ec
[<8037a288>] unknown_bootoption+0x0/0x31c

Stack : 80690000 80333914 5149c84d 00000002 83dc21d0 80301de4 00000000 00000001
        83d80000 83dc0000 000bfc30 00000001 806971bc 00000000 83cb0280 c06b4354
        80301dd0 c0150000 80301e10 00000000 0000000a 80333930 00000000 80144928
        00000064 00000000 00000000 ffffffa3 00000000 1b2e09d0 83d8397c 83d83994
        80301e50 806969b4 806969b4 80330000 806971bc 00200200 802d6a40 c06b9b08
        ...
Unwound Call Trace:
[<c077a50c>] ar5416AniArPoll+0xa0/0x570 [ath_hal]
[<c06b4354>] ath_calibrate+0x59c/0x674 [ath_dev]
[<c06b9b08>] _ath_internal_timer_handler+0xb4/0x1ac [ath_dev]
[<80133d6c>] run_timer_softirq+0x18c/0x1f8
[<8012f350>] __do_softirq+0x84/0x138
[<8012f470>] do_softirq+0x6c/0x74
[<8010a6b4>] do_IRQ+0x24/0x34
[<8010666c>] ar7240_interrupt_receive+0xec/0x100

I never see nothing like this on APs!

I replaced the AP and all seems to work fine  now.

Please open a case with TAC.  it probably will have to be replaced

Thanks for update us.

try to do to this unit factory reset - and then try agian.

if yet not solved - you should open a TAC and replace that unit.

Me

Anyone knows what was the root cause for this?