*BSD News Article 61887


Return to BSD News archive

Newsgroups: comp.unix.bsd.freebsd.misc
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.cs.su.oz.au!inferno.mpx.com.au!news.mel.aone.net.au!imci4!newsfeed.internetmci.com!in2.uu.net!twwells!bill
From: bill@twwells.com (T. William Wells)
Subject: Re: Networking problems...  No buffer space for sendto.
Organization: None, Mt. Laurel, NJ
Message-ID: <DMpn90.F5o@twwells.com>
References: <gnn.824074349@wrs.com>
Date: Tue, 13 Feb 1996 10:46:12 GMT
Lines: 40

In article <gnn.824074349@wrs.com>, George Neville-Neil <gnn@wrs.com> wrote:
:       I've just installed FreeBSD 2.1 on my WinBook (486DX/33 laptop
:     with 16M/340M) and I'm having some problems with the networking.  I
:     can bring up the network and mount NFS volumes over it.  BTW The
:     network card is a 3COM 3C589C Combo PCMCIA card.  After a while the
:     network connection hangs/goes away. I can't ping the machine from
:     outside, and when I try to ping to the outside from the console I
:     get a "sendto: no buffer space" error.  Anyone else seen this?
:       I've included the MINION kernel config file at the end of this
:     message.

When this is occuring, try doing a netstat -m.

58 mbufs in use:
	1 mbufs allocated to data
	44 mbufs allocated to packet headers
	11 mbufs allocated to protocol control blocks
	2 mbufs allocated to socket names and addresses
1/32 mapped pages in use                        <-----------this line
71 Kbytes allocated to network (12% in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

Look at the marked line. The first number is the number of pages
currently in use. The second number is the number of pages
allocated, which can be larger than the first number.

Normally, when the first number exceeds the second number, the
kernel automatically allocates more. However, there is a limit to
this, that depends on the number of users configured.

If you see these two numbers are equal (or the first is slightly
smaller than the second), then what has happened is that you have
run out of mbufs. Increasing maxusers will increase the maximum
memory for mbugs.  It can also be set explicitly with option
NNMBCLUSTERS .

2.1 seems to run out of mbufs more often than the previous
versions. I suspect a kernel bug somewhere....