Date: Mon, 12 May 2003 12:16:35 +0200 From: Stijn Hoop <stijn@win.tue.nl> To: Greg Lehey <grog@FreeBSD.org>, freebsd-questions@FreeBSD.org Subject: Re: what process is eating my swap? -> SOLVED Message-ID: <20030512101635.GD20217@pcwin002.win.tue.nl> In-Reply-To: <20030509141200.GC7374@pcwin002.win.tue.nl> References: <20030509141200.GC7374@pcwin002.win.tue.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
--Hf61M2y+wYpnELGG Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, May 09, 2003 at 04:12:00PM +0200, Stijn Hoop wrote: > My swapspace is 99% used, but I can't find a way to figure out what proce= ss is > using it. top & ps don't seem to list swap usage of processes, and systat > -swap only lists the fact that 99% is in use. [snip] > - I'm using vinum for two RAID-5 arrays on 8 ATA disks, and was > rsync'ing from one to the other at the time the kernel complained > about lack of swapspace: >=20 > [stijn@sandcat] </> dmesg | grep swap > swap_pager_getswapspace: failed > pid 33932 (rsync), uid 0, was killed: out of swap space >=20 > Could vinum somehow 'leak' swap space? My swap partition is on a regular > SCSI disk, not on a vinum volume, so this seems farfetched but I'm a bit > at a loss here. >=20 > Anyone have any clue? I need to have a bit of virtual memory on this > machine... I figured it out, and I thought I'd let you know what the problem was: -> don't _EVER_ build a vinum RAID-5 with a blocksize of 512 bytes (and try to do something with it) Actually, that wasn't my intention of course but I forgot to end my plex si= ze with a 'k'. While doing the rsync to this RAID-5 it was, I think, the buffer cache that couldn't keep up with the number of blocks vinum had to read/write, and so = my virtual memory kept filling up. At least, now that I've rebuild the RAID with the proper blocksize, I'm not running into swap space problems anymore. Greg, does it ever make any sense to build a RAID with a blocksize < 128k? The manpage says it doesn't; maybe vinum could issue a warning while creating such a plex? The output of ls tells you that the subdisks are at offsets of 512 B <ytes> of course, but the missing 'k' is rather hard to spot. I'm off to shoot myself in the foot somewhere else then... --Stijn --=20 Fairy tales do not tell children that dragons exist. Children already know dragons exist. Fairy tales tell children the dragons can be killed. -- G.K. Chesterton --Hf61M2y+wYpnELGG Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (FreeBSD) iD8DBQE+v3SDY3r/tLQmfWcRAroHAJ4ttfuAXB2ZKwDzIeYr+TVUBOuRUgCbBjSg 4p5frGS4jyTUhowZtX8MYrQ= =emSN -----END PGP SIGNATURE----- --Hf61M2y+wYpnELGG--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030512101635.GD20217>