Date: Fri, 16 Oct 1998 12:54:05 +1000 From: Andy Farkas <andyf@speednet.com.au> To: Studded <Studded@gorean.org> Cc: "Mikhail A. Sokolov" <mishania@demos.net>, Pecsenyanszky Istvan <pisti@c3.hu>, freebsd-isp@FreeBSD.ORG Subject: Re: kernel panic (out of mbuf's) Message-ID: <3626B54D.49AE0C41@speednet.com.au> References: <Pine.SGI.4.05.9810132035350.19435-100000@visio.c3.hu> <19981014011323.61957@demos.su> <362664A0.3209B97E@gorean.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Studded wrote: > > "Mikhail A. Sokolov" wrote: > > > > a) increase maxusers, as handbook teaches, b) options "NMBCLUSTERS=4096". > > When you exceed ti, try next 2x value, but don't raise NMBCLUSTERS to > > >12288 (even though you hardly need it that high). > > Why should you not raise it above 12,288? What evidence do you have to > support that? This is an issue near to my heart, so I would appreciate > any details you can provide. > > I do agree however that leaving maxusers at 512 and increasing > NMBCLUSTERS is the answer. IIRC, with maxusers at 512 there are around > 8,000 mbuf clusters, so something higher than that would be the place to > start. According to the FreeBSD FAQ, http://www.freebsd.org/FAQ/FAQ92.html, "NMBCLUSTERS=4096" is the maximum. -- :{ andyf@speednet.com.au Andy Farkas System Administrator Speed Internet Services http://www.speednet.com.au/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3626B54D.49AE0C41>