Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 29 Apr 1999 16:36:42 -0500 (EST)
From:      Alfred Perlstein <bright@rush.net>
To:        Bryce Newall <data@dreamhaven.net>
Cc:        FreeBSD Questions List <freebsd-questions@FreeBSD.ORG>
Subject:   Re: No buffer space
Message-ID:  <Pine.BSF.3.96.990429163527.7628E-100000@cygnus.rush.net>
In-Reply-To: <Pine.NEB.4.10.9904291011230.608-100000@ds9.dreamhaven.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 29 Apr 1999, Bryce Newall wrote:

> On Wed, 28 Apr 1999, Alfred Perlstein wrote:
> 
> > > Apr 28 15:20:01 calico named[13140]: socket(SOCK_RAW): No buffer space
> > > available
> > 
> > recompile with a higher maxusers, or do a search on the lists for
> > NMBCLUSTERS to find out how to increase them and what the recommended 
> > increase is.
> 
> MAXUSERS was originally set to 64, and that was causing some other
> problems, so we went to an extreme and pushed it up to 512.  That was
> apparently way too high, and was causing the system to reboot
> spontaneously several times a day.  We backed it down to 128, which
> stabilized that part, but now we're back to having buffer space problems
> again.  NMBCLUSTERS is set to 6000 at the moment... think that's too low?
> 
> I searched the mailing lists, and found someone referencing their output
> from netstat -m.  Here's what I have:
> 
> 2194/3136 mbufs in use:
>         1001 mbufs allocated to data
>         1193 mbufs allocated to packet headers
> 691/1460/6000 mbuf clusters in use (current/peak/max)
> 3312 Kbytes allocated to network (50% in use)
> 0 requests for memory denied
> 0 requests for memory delayed
> 0 calls to protocol drain routines
> 
> I really don't know what all that means... :)  But it looks like the top
> line is saying that 2194 out of 3136 are in use?  Or am I reading that
> wrong?  If I'm reading it correctly, would increasing NMBCLUSTERS give me
> more room to work?

Well... if things are better now, leave them be. :)  if not, try bumping
NMBCLUSTERS again.

I would also suggest cvsup'ing to -stable to fix the problems with high
maxusers if you are a 3.x system.

-Alfred



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.990429163527.7628E-100000>