Date: Mon, 25 Apr 2005 17:38:07 +0200 From: =?ISO-8859-1?Q?Rasmus_Br=F8gger?= <Rasmus.Brogger@uni-c.dk> To: freebsd-stable@freebsd.org Subject: Re: background_fsck=no does not work? Message-ID: <426D0EDF.50500@uni-c.dk> In-Reply-To: <426C85EA.5090906@speechpro.com> References: <DD8197BD35BCC63A18AF62CB@palle.girgensohn.se> <426BA8FA.3080602@samsco.org> <C863DF88E98B5ECA262E825E@palle.girgensohn.se> <426C85EA.5090906@speechpro.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Igor Robul wrote: > Palle Girgensohn wrote: > >> >> No, but the machine crashes a lot, sometimes a couple of times in a >> day (oddly mostly on week-ends, and sometimes it can be stable for a >> week or two). I'm clueless to what's causing it, but suspect that the >> amd64 stuff is broken, at least for this machine (dual Xeon). I wanted >> foreground fsck since I thought there might be a problem with the file >> systems, but I guess I was wrong... >> > I had frequent panics which I tracked to time I had started rsync (with > large data) over em0 gigabit netcard. > I have reinserted card and it works fine now, so I think this was > "conductor problem". > > _______________________________________________ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org" I run quite a lot of rsync of large archives, and I've had no problems at all. System is a 5.3-RELEASE-p2 on a Dell 1750 (Dual Xeon 2.8 with 2GB ram), not em but bge gigabit driver. The system peaks with about 150 connected rsync sessions, the archive to sync is about 90GB and some 200k files. The netload ranges between 10 mbit/s to 180 mbit/s. At peaks it hits a load about 30. I've done very little tuning, but the disk system runs via a Qlogic HBA to gain faster IO from a SAN. Only problem I've had was when about 130 rsync clients connecting within the same second. It saturated the server for some 10 min. at a load above 150, it didn't crash however and recoverd slowly. This was fixed by changing the dial-up schedule on the clients. 5.3 is indeed very stable :) Cheers, Rasmus
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?426D0EDF.50500>