All mbufs exhausted/Random spontaneous reboots started today

[Einstein Oliveira]

Chris Byrnes wrote:
> Sorry for the crosspost, not sure where this should go.  Please cc me in
> replies because I am not subscribed.
> 
> Server was up almost 134 days and spontaneously rebooted this morning saying
> "All mbufs exhausted, please see tuning(7)."  Had never seen it before, read
> up on it and figured something just happened causing a lot of memory usage
> or something, chalked it up to an abnormality.
> 
> Took the opportunity to upgrade to the latest -STABLE and rebooted.
> 
> Server was up for 30 minutes and just rebooted AGAIN with the same
> message(s)..
> 
> Jan 11 16:32:19 /kernel: All mbufs exhausted, please see tuning(7).
> Jan 11 16:32:50 last message repeated 33 times
> Jan 11 16:34:44 last message repeated 127 times
> 
> Do you have any idea what's causing this and how I can monitor/stop it?
> 

Use netstat -m to chek the current configuration/load.

It's something like this:

----
191/8272/65536 mbufs in use (current/peak/max):
         186 mbufs allocated to data
         5 mbufs allocated to packet headers
170/8128/16384 mbuf clusters in use (current/peak/max)
^^^^^^^^^^^^^^
18324 Kbytes allocated to network (37% of mb_map in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines
----

You can modify the number of mbufs with sysctl variable 
kern.ipc.nmbclusters.

Usually setting this on /boot/loader.conf

ex.:

kern.ipc.nmbclusters=16384

Check also tuning(7) ;)

> Thank you..
> 
> 
> Chris
> 

-- 
Einstein Oliveira                 einstein at yawl.com.br
______________________________________________________
Yawl Internet Ltda.            http://www.yawl.com.br/