Date: Fri, 18 Oct 2013 15:27:04 +0100 From: Frank Leonhardt <frank2@fjl.co.uk> To: freebsd-questions@freebsd.org Subject: Re: warning: total configured swap (8960911 pages) exceeds maximum recommended amount (8243200 pages). Message-ID: <52614538.8090309@fjl.co.uk> In-Reply-To: <201310181254.r9ICshUi050154@mech-cluster241.men.bris.ac.uk> References: <201310181254.r9ICshUi050154@mech-cluster241.men.bris.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On 18/10/2013 13:54, Anton Shterenlikht wrote: >> Fri, 18 Oct 2013 05:41:54 -0700 (PDT) >> Date: Fri, 18 Oct 2013 13:41:53 +0100 >> From: RW <rwmaillists@googlemail.com> >> To: freebsd-questions@freebsd.org >> Subject: Re: warning: total configured swap (8960911 pages) exceeds maximum >> recommended amount (8243200 pages). >> >> On Thu, 17 Oct 2013 19:14:33 +0100 >> Frank Leonhardt wrote: >> >>> On 17/10/2013 17:01, RW wrote: >>>> The hard limit that's mentioned there is i386 only. I think that may >>>> have changed a bit since 8.2. >>>> >>> This was running 8.2 on AMD64, and it's definitely there (and with >>> the same value as the i386 version). The code's there on 9.1, as is >>> the manifest constant in param.h and it's still the same value. I >>> haven't tried 9.2 yet. So what am I missing about i386? >> >> The changed was merged into 9-stable in September 2012, 9.1 was >> branched in August. > Looking at the article > ( http://blog.frankleonhardt.com/2011/large-swap-files-on-freebsd-die-with-mystery-killed-howto-add-lots-of-swap-space/ ) > the default value of kern.maxswzone has also > changed meaning. > > On r256683 I get: > > # sysctl kern.maxswzone > kern.maxswzone: 0 > # > > So I'm not sure, do I set this variable > to the *total* swap size? > Or to the excess, above the "maximum > recommended amount" reported by dmesg? > > As RW has pointed out, things may have moved on since the blog post (which was specific about it being AMD64 V8.2). However, back then maxswzone was the size of the radix tree, which doesn't correlate directly with the number of structures. I'm sure wikipedia will explain. As I said at the time, I didn't have a good way of calculating values - just up it and see. I think I flagged this whole situation as a bit naff at the time - it's okay for me to go to the kernel source for answers but this doesn't count as "user friendly", so I'm glad it's been addressed. However, if it's changed I'm to answer questions on the new version. Regards, Frank.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?52614538.8090309>