ArubaOS and Controllers

Reply
Contributor II

Watchdog Timeouts and Controllers Rebooting, Please help!

Over the last week we've had a couple of our local controllers reboot on their own. Looking in the Show Tech a couple of things are sticking out, I'm hoping somebody can help me out.

First it looks like the controller rebooted due a Watchdog Timeout:

Switch uptime is 2 hours 15 minutes 52 seconds
Reboot Cause: Watchdog timeout or hard reset.


It also looks like the Heartbeat misses between the Local & Master are increasing (5 misses in a couple hours already):

Missed -> HB Resp from Master
-----------------------------
IP Address HB Req HB Resp Cfg Terminate Peer Reset Total Missed Last Sent Missed Last Synced/Last Missed
---------- ------ ------- ------------- ---------- ------------ ---------------- -----------------------
10.48.14.10 795 795 0 0 5 0 Sun Apr 25 08:40:42 2010


Lots of ICMP errors in the Datapath App Stats:

show datapath application counters

Datapath Application Statistics
-------------------------------
NAPT Allocation Failures 0
NAT Allocation Failures 0
NAT TCP Sequence Failures 0
ICMP Errors Received 531
ICMP Errors Denied 507


And lastly here is the output from the top of the errorlog

show log errorlog all
Apr 25 08:37:57 |fpapps| Reboot Cause: Watchdog timeout or hard reset.
Apr 25 08:40:02 |publisher| Pubsub send message code 0 source port 8378 to destination port 8345 failed, errno Connection timed out
Apr 25 08:40:05 |publisher| Pubsub send message code 0 source port 8378 to destination port 8384 failed, errno Connection timed out
Apr 25 08:40:44 |profmgr| An internal system error has occurred at file sapi.c function sapi_rsp line 247.

Any ideas? Thanks.
Guru Elite

Unscheduled reboot

Any controller that has an unscheduled reboot you should open a case so that support can get to the bottom of it. That is the best way to deal with that.
******************
Answers and views expressed by me on this forum are my own and not necessarily the position of Aruba Networks or Hewlett Packard Enterprise.
******************
Contributor II

Re: Watchdog Timeouts and Controllers Rebooting, Please help!

Will do, thanks :)
Frequent Contributor II

Re: Watchdog Timeouts and Controllers Rebooting, Please help!

Definitely open a support ticket. Do a tar crash (the TAC will ask for it).

Just for your own FYI: You can upzip/untar it on your local computer and look around. You can sometimes determine what caused it (which process anyway) by drilling down through the crash file. Gives you a bit of insight into the inner workings of the AOS/controller.
Search Airheads
cancel
Showing results for 
Search instead for 
Did you mean: