[ previous ] [ next ] [ threads ]
 
 From:  Chris Buechler <cbuechler at gmail dot com>
 Cc:  m0n0wall at lists dot m0n0 dot ch
 Subject:  Re: [m0n0wall] 1.2b8 hanged: all mbufs exhausted
 Date:  Wed, 15 Jun 2005 14:30:52 -0400
On 6/15/05, Bjoern Euler (Lists at edain) <lists at edain dot de> wrote:
> 
> there are no worms/viruses on my net. Its a small home network with
> three windows clients and one freebsd server. On that a mldonkey was
> running when the m0n0wall died but I don't think this has something to
> do with it.
> 

yeah, that shouldn't create even close to enough traffic to run out of mbufs.  


> My m0n0wall box is a P3 system with two Intel Lan Management Pro
> adapters and 256MB of RAM. I should be able to handle my 1024/128kbit
> DSL line...
> 

yeah, handle that DSL plus a T3 probably.  :)  


> Yesterday evening I configured OpenVPN on my box with a client link to
> one of my friends. Maybe that is problem?

Might have something to do with it.  I know out of mbuf can also be
caused by bugs of some sort, or other issues, but am not familiar
enough with the kernel internals to know just what..


> The output of netstat -m before I replaced it:
> 
> $ netstat -m
> 3943/3968/17856 mbufs in use (current/peak/max):
>      3010 mbufs allocated to data
>      4 mbufs allocated to packet headers
>      929 mbufs allocated to socket names and addresses
> 388/576/4464 mbuf clusters in use (current/peak/max)
> 2144 Kbytes allocated to network (16% of mb_map in use)
> 0 requests for memory denied
> 0 requests for memory delayed
> 0 calls to protocol drain routines
> 

Looks normal for your type of setup.  Peaks nowhere near the max (when
you ran this).

-Chris