PDA

View Full Version : system freezes after a period, something to do with system clock



alexlambdax
17th July 2004, 04:33 PM
hi,

My system freezes for no reason after it has been idle for a while (like 8-10 hours). I look at the kernal logs /var/log/message and found this strange line:

Jul 16 16:37:03 localhost smartd[2209]: System clock time adjusted to the past. Resetting next wakeup time.

and when i try to log into my computer at 10pm, the system clock was at 16:37, and it freezes during iinitialization and i had to do a hard reset.

any idea why this is happening, what smartd is, how i can disable it, etc?

i'm running FC2 on a P4 hyperthreading system, using an smp kernel

thanks
Alex

friez
17th July 2004, 05:12 PM
smartd is a daemon that monitors the Self-Monitoring, Analysis and
Reporting Technology System(S.M.A.R.T.) built into ATA-3 and later IDE
and SCSI-3 Hard Drives. This is used to check the reliability of the
hard drive and predict drive failures. smartd is daemon designed to
notify users of S.M.A.R.T. errors via SYSLOG interface. it can be disable by going to running the command
system-config-services then stop the service and uncheck it so it does not start at boot but i don't believe smartd has anything to due with the system clock :confused:

alexlambdax
18th July 2004, 03:09 AM
My system freezes again...this time the log looks more interesting

who is user cyrus or user news? Do you think apm is the problem. RIght now i have the kernel parameter acpi=oldboot. maybe i should add apm=off?

Thanks


Jul 17 04:02:02 localhost su(pam_unix)[4495]: session opened for user cyrus by (uid=0)
Jul 17 04:02:02 localhost su(pam_unix)[4495]: session closed for user cyrus
Jul 17 04:02:03 localhost logrotate: ALERT exited abnormally with [1]
Jul 17 04:05:33 localhost su(pam_unix)[4892]: session opened for user news by (uid=0)
Jul 17 04:05:33 localhost su(pam_unix)[4892]: session closed for user news
Jul 17 07:14:50 localhost kernel: [drm:i830_wait_ring] *ERROR* space: 131048 wanted 131064
Jul 17 07:14:50 localhost kernel: [drm:i830_wait_ring] *ERROR* lockup
Jul 17 08:26:23 localhost udev[5312]: removing device node '/udev/vcs7'
Jul 17 08:26:23 localhost udev[5313]: removing device node '/udev/vcsa7'
Jul 17 08:26:24 localhost gpm[2493]: O0o.oops(): [console.c(83)]:
Jul 17 08:26:24 localhost gpm[2493]: Opening console failed.
Jul 17 07:14:56 localhost network: Shutting down interface eth0: succeeded
Jul 17 07:14:56 localhost kernel: mtrr: base(0xf0020000) is not aligned on a size(0x400000) boundary
Jul 17 07:14:56 localhost kernel: mtrr: 0xf0000000,0x8000000 overlaps existing 0xf0000000,0x400000
Jul 17 07:14:57 localhost network: Shutting down loopback interface: succeeded
Jul 17 07:15:02 localhost apmd[2147]: System Standby
Jul 17 09:59:16 localhost syslogd 1.4.1: restart.

ilja
18th July 2004, 03:21 AM
cyrus is a mail programm and news is afaik a news-server. Just a guess, try to stop them and also try to run chrootkit.

friez
18th July 2004, 03:51 AM
:confused: the problem seem to be with your onboard graphic

kernel: [drm:i830_wait_ring] *ERROR* space: 131048 wanted 131064
which is causing your system to lock up
this was a known issue in fc1 but has supposed to be fixed with an update kernel
http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=84097 (http://)
try use the vanilla kernel instead :(

alexlambdax
18th July 2004, 05:05 AM
I try enabling framebuffer, coz my screen saver doesn't work if i don't. I was looking through the forum, and someone suggested having a setting vga=0x318 as a kernel parameter (i have no idea why it is 0x318, i just put it there coz it seems to sort of work). Is this the way to enable the framebuffer? Before i did this, the system didn't freeze unless i have a screensaver on. But now with the vga=0x318, my system freeze at the login screen after a while. and like friez said, it's a graphic card issue.

any suggestions?

many thanks


kernel /boot/vmlinuz-2.6.5-1.358smp ro root=LABEL=/ vga=0x318 apm=off acpi=oldboot rhgb quiet