Date: Wed, 6 Feb 2008 14:52:40 -0500 (EST) From: doug@safeport.com To: Jerry McAllister <jerrymc@msu.edu> Cc: Frank Staals <frankstaals@gmx.net>, FreeBSD <freebsd-questions@freebsd.org> Subject: Re: buildworld releng7 exterme performance loss Message-ID: <20080206144920.G12870@fledge.watson.org> In-Reply-To: <20080206193916.GA20244@gizmo.acns.msu.edu> References: <47A3720D.1080702@gmx.net> <47AA08E8.1020502@gmx.net> <20080206193916.GA20244@gizmo.acns.msu.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 6 Feb 2008, Jerry McAllister wrote: > On Wed, Feb 06, 2008 at 08:22:16PM +0100, Frank Staals wrote: > >> Frank Staals wrote: >>> Every 2 weeks or so I rebuild my sources on my laptop ( Dell Latitude >>> D630 ). Last wednesday I wanted to update my system again since it was >>> allready a couple of weeks ago I rebuilded my sources. There were no >>> abnormalities during the build/install. But when I restarted my system >>> the system performance was absolutely *horrible*. When I try typing >>> anything in a terminal window the lag is about 20 seconds before the >>> command even shows up. Switching ttys takes at least as long. When >>> logging in remotely the performance is better; giving a random command >>> through the ssh session sometimes even allows a command run on a >>> local-console to run/finish ( or at least thats what the screen shows ). >>> >>> Even trying to shutdown the system using shutdown -p now doesn't have >>> any effect: I get the message "System is going down NOW" but nothing >>> realy happens, I just get my prompt back and I can continue entering >>> commands. When I push the power-off button the system actually starts >>> shutting down but it stops at "Writing entropy file ." leaving my >>> only choise to power off the system the hard way. When booting the >>> system to single-user mode however it reacts normally and also >>> shutting down works as it should >>> >>> Last 2 days I tried redownloading the sources from a different mirror >>> ( cvsup2.freebsd.org instead of cvsup.nl.freebsd.org ) to rule out >>> outdated/corrupted sources. But I still get the same behaviour. >>> Gladly I still had my previous sources so I managed to restore my >>> system in a working state. Those sources are from about 3 weeks ago. >>> Does anyone know what could have caused this behaviour and how to fix >>> it ? Below is some system information, however I'm not sure what I >>> should provide to help out, if I should test something I'll gladly do. > > Sounds like the system is spending a lot of time waiting for some > resource - I don't know which one[s]. > > Have you perused the system logs? Can you be running out of memory > or swap space? Is the system built with debugging turned on? > > I'm not sure what else to suggest. if you have not already done so turn on all.log (see etc/syslog.conf) > ////jerry > > >>> >>> Working system (sources from about 3 weeks ago, rebuilded today ) : >>> >>> frank@Rena# uname -a >>> FreeBSD Rena.FStaals.net 7.0-PRERELEASE FreeBSD 7.0-PRERELEASE #0: Fri >>> Feb 1 20:23:11 CET 2008 root@:/usr/obj/usr/src/sys/RENAKERNEL i386 >>> >>> dmesg: http://fstaals.net/junk/rena/dmesg_old.txt >>> >>> With today's ( 1 frebruary 2008 ) sources : >>> >>> frank@Rena# cat /root/uname >>> FreeBSD 7.0-PRERELEASE FreeBSD 7.0-PRERELEASE #0: Fri Feb 1 18:32:59 >>> CET 2008 frank@Rena.FStaals.net:/usr/obj/usr/src/sys/RENAKERNEL i386 >>> >>> dmesg: http://fstaals.net/junk/rena/dmesg_2008_01_02.txt >>> >>> Kernel config used for both the builds : >>> >>> http://fstaals.net/junk/RENAKERNEL.txt >>> >> Updated my sources again today. This time even build with a GENERIC >> kernel but still the same result. Is there nobody who can help me with >> this ? >> >> -- >> -Frank Staals if you have not already done so turn on all.log (see etc/syslog.conf)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080206144920.G12870>