Fedora Core 2 freeze after several minutes/hours

[Jim Cornette]

Lee Un Pang wrote:
> On Fri, 23 Jul 2004 20:38:07 -0400, Jim Cornette wrote

>>
>>I suggest runlevel 3 and dropping the GUI until it is needed.
>>
>>http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=128358
>>


> 
> Dear Jim:
>   I am the same problem as you in Dell computer. Do you have any solution on
> this? since i must use X to run some special application.

Not yet. Maybe smartd, acpi, drm. I am using guesses to figure out what 
is going wrong. I see AGP seems to be somehow in the picture too w/ a 
posting on fedora-test-list back in February.

>   By the way, did you try FC1? Do you have this problem by using FC1?
> 

There was discussion on the google search related to 2.4 kernel builds. 
I think that this points to this problem affecting systems for quite awhile.

> Thanks a lot
> Lewis
> 

I wish I could help, but I have no solution yet.

With the excerpt from the log below. I think that the drm:i830_wait_ring 
is getting an unacceptable address and then tries three time to get the 
address it wants. It then crashes X and nothing recaptures the display 
hardware. (give it up, you ain't gettin' 131064!)


[drm] Initialized i830 1.3.2 20021108 on minor 0:
mtrr: base(0xe8020000) is not aligned on a size(0x400000) boundary
ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 10 (level, low) -> IRQ 10
PCI: Setting latency timer of device 0000:00:1f.5 to 64
intel8x0_measure_ac97_clock: measured 49434 usecs
intel8x0: clocking to 48000
[drm:i830_wait_ring] *ERROR* space: 131052 wanted 131064
[drm:i830_wait_ring] *ERROR* lockup
[drm:i830_wait_ring] *ERROR* space: 131004 wanted 131064
[drm:i830_wait_ring] *ERROR* lockup
[drm:i830_wait_ring] *ERROR* space: 129740 wanted 131064
[drm:i830_wait_ring] *ERROR* lockup
[drm:i830_wait_ring] *ERROR* space: 130996 wanted 131064
[drm:i830_wait_ring] *ERROR* lockup

This leads the user to believe that there computer is locked up. The 
screen never gets updated until the computer recycles through a hot boot.

I did a search for the drm:i830_wait_ring and came up with the below 
linked google result. Some were from February in the test list. In his 
case, he was using the graphical boot loader and gdm kept giving him the 
problem. There were discussions relating to the 2.4 kernel so the 
problem could still affect FC1 which is 2.4 kernel based.

http://www.google.com/search?q=drm%3Ai830_wait_ring&btnG=Search&hl=en&lr=&ie=UTF-8

Funny thing is that the only error in my logs related to smartd below.

smartd: smartd shutdown failed

then below:

smartd[2047]: smartd version 5.30 Copyright (C) 2002-4 Bruce Allen
smartd[2047]: Opened configuration file /etc/smartd.conf
smartd[2047]: Configuration file /etc/smartd.conf parsed.
smartd[2047]: Device: /dev/hda, No such device or address, open() failed

(Of course, there are only sda devices. It should fail.)

smartd[2047]: Unable to register ATA device /dev/hda at line 30 of file 
/etc/smartd.conf
smartd[2047]: Unable to register device /dev/hda (no Directive -d 
removable). Exiting.

Jim