Date: Sat, 11 Jan 1997 19:20:46 +0100 (MET) From: Christoph Kukulies <kuku@gilberto.physik.rwth-aachen.de> To: kimc@w8hd.org (Kim Culhan) Cc: kuku@gilberto.physik.rwth-aachen.de, freebsd-hackers@freefall.freebsd.org Subject: Re: Xaccel leaks? Message-ID: <199701111820.TAA08255@gilberto.physik.rwth-aachen.de> In-Reply-To: <Pine.BSF.3.95.970111124340.19559C-100000@moonpie.w8hd.org> from Kim Culhan at "Jan 11, 97 01:02:56 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
>
>
> On Sat, 11 Jan 1997, Christoph Kukulies wrote:
>
> >
> > On one of my machines (some older 2.2-current) - where one user has the habit of
> > leaving many X apps like ghostview/gs standing on for days - top is showing
> > that Xaccel is now at 11M SIZE (9000K RES). swapinfo shows 37% of 44M swap.
> >
> > What could I do about it? Approach Xinside? Rebuild ghostview with another
> > malloc lib? Run XFree86?
> >
> > In that machine we have a Matrox Millenium 4MB and XFree86 support was a bit
> > sparse until recently.
>
> This machine is a Pentium 100 running 3.0-current and Xinside's v2.1 X
> server, and has a Millenium 4MB and 32MB main memory.
>
> Top shows Xaccel is now at 7.92M SIZE (7156K RES) swapinfo shows
> 38% of 131072M swap.
131 GB swap is a lot :-)
>
> This is running the CDE with vic, vat, wb and sdr plus a few 'dtterms'
>
> What would be reasonable to expect?
The problem with the machine was that it often ran out of swap space -
OK, 44 MB swap isn't much but is has 32MB memory and for a single user
with a couple of color_xterms this should suffice.
Since that happened I'm logging tops output to a file every five minutes
watching it from time to time.
>
> regards
> --kim
>
> kimc@w8hd.org
>
>
>
--Chris Christoph P. U. Kukulies kuku@gil.physik.rwth-aachen.de
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701111820.TAA08255>