Date: Fri, 28 May 1999 12:38:04 -0700 (PDT) From: Doug White <dwhite@resnet.uoregon.edu> To: Kevin Day <toasty@home.dragondata.com> Cc: current@FreeBSD.ORG Subject: Re: -Current still leaking mbuf's Message-ID: <Pine.BSF.4.03.9905281123530.11808-100000@resnet.uoregon.edu> In-Reply-To: <199905271847.NAA18802@home.dragondata.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 27 May 1999, Kevin Day wrote:
> I tried doubling whatever it was that putting maxusers at 256 set it at. (I
> can get the exact number later). I'm running with no NMBCLUSTERS setting,
> just with maxusers at 128 at the moment.
Okay, you may want to bump NMBCLUSTERS.
> > You might want to collect some netstat -m stats as time goes on. In
> > addition to being easier to read, it may give you some hints as to how
> > high you want to go with mbuf clusters.
>
> I added a cron job to to netstat -m every half hour... Right now, after 10
> hours of being up:
>
> 130/1686/2560 mbuf clusters in use (current/peak/max)
^^^^
This is what you care about. If the peak pops above 1670 (2/3 of 2560),
you should try bumping NMBCLUSTERS to 4096 or maybe 8192.
> This didn't happen in 2.2.8 or 3.1, so I'm trying to figure out what's
> causing it. :)
NFS could be a little less efficient than in previous versions.
[...]
> #11 0xc015c382 in m_retryhdr (i=0, t=1) at ../../kern/uipc_mbuf.c:297
> 297 panic("Out of mbuf clusters");
Doug White
Internet: dwhite@resnet.uoregon.edu | FreeBSD: The Power to Serve
http://gladstone.uoregon.edu/~dwhite | www.freebsd.org
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" 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.4.03.9905281123530.11808-100000>