Date: Tue, 25 Jan 2000 15:42:04 +0200 (EET) From: Yaroslav Terletsky <ts@polynet.lviv.ua> To: Alfred Perlstein <bright@wintelcom.net> Cc: FreeBSD Questions Mailing List <freebsd-questions@FreeBSD.ORG>, altq@csl.sony.co.jp Subject: Re: Panic with 3.4: rtfree Message-ID: <Pine.BSF.3.96.1000125153837.43912C-100000@NetSurfer.lp.lviv.ua> In-Reply-To: <20000124125902.E26520@fw.wintelcom.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 24 Jan 2000, Alfred Perlstein wrote:
> * Yaroslav Terletsky <ts@polynet.lviv.ua> [000124 09:44] wrote:
> >
> > Dear Coders,
> >
> > One of my routers being upgraded to 3.4-STABLE (as of Dec 30 1999)
> > crashes *very* often with panic("rtfree");
> > (With old 2.2.2 branch it works ok.)
> >
> > It is cool in testing evnironment but after few minutes in real cond.
> > (I think due to higher traffic and RT manipulations) it crashes.
> > My RT is about 70 entries in size, maxusers 32, NMBCLUSTERS=2048.
> > I'm also using ALTQ 2.0 (it does not patch sys/net/route.c).
> >
> > Any suggestions why it is and how to fix this problem?
>
> Why do you have such a low maxusers/NMBCLUSTERS setting?
>
> I can understand the maxusers, but lowering NMBCLUSTERS on a router
> doesn't seem like the right thing to do.
>
> -Alfred
Look at /usr/src/sys/conf/param.c, the formula for NMBCLUSTERS is:
/* maximum # of mbuf clusters */
#ifndef NMBCLUSTERS
#define NMBCLUSTERS (512 + MAXUSERS * 16)
#endif
So for 32 users the number is just 1024 and I have increased this
to 2048 as ALTQ's README suggested.
Yaroslav
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.3.96.1000125153837.43912C-100000>