Red Hat Bugzilla – Bug 29030
Filesystem crashes after remounting root fs several times
Last modified: 2005-10-31 17:00:50 EST
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows 98; Win 9x 4.90)
After mounting the root file system a couple of times (i would say over 10
times), it crashes on boot-up, i get an incredible amount of errors
reported by fsck and get dropped to admin shell. I re-installed fisher 3
times and it did the same the 3 times
Steps to Reproduce:
2.Re-boot about 10 times
Actual Results: The fs crashes :(
Expected Results: It should not :P
Well, i'd like to have more infos but i just can't use the system
afterwards, all logs are gone many files are missing... A big mess...
Could you please describe the hardware in your machine? In particular,
I'd like to know whether you use IDE or SCSI, and if IDE, which chipset
you have, and if you use SCSI, which SCSI controller you have.
We (Red Hat) should really try to resolve this before next release.
This is my hardware :
ASUS A7V MotherBoard (VIA Appolo KT133 Chipset)
ThunderBird 1100Mhz CPU
ATI Rage128 32MB DDR AGP
Creative SB Live! Value2.0
Fujitsu 20.4GB IDE (HD is ATA/100 but running ATA/66)
Creative 52X IDE CD-Rom
D-Link DFE538-TX PCI Network (RTL8139 Based)
Digital DC21140 PCI Network (Tulip)
256MB SDRAM PC-133
Latest ASUS BIOS (1005D)
Note: I have a Promise Ultra ATA/100 Controller on-board but im not using it
since Linux didn't support it (does fisher?)
That's about all..
Hope it may be of some help
Note : I forgot to mention that the 3rd time I installed fisher, i changed my
drive for a Maxtor 30.7G i took from another of my computers... so obviously,
my HD is not the cause... (and Windows and QNX wich resides on seperate
partitions never had any problems)
A few questions:
When you say "reboot", are you talking about clean shutdown/reboot, or a hard
reboot such as /sbin/reboot -f, pressing the reset button or doing a power cycle?
Secondly, do you have an example of the log messages when the error happens, and
are there any disk controller messages amongst the filesystem messages?
Finally, has this hardware been stable under older versions of the kernel?
Yes, I'm talking about a clean reboot using reboot (or poweroff for a cold
I don't have the logs, they get deleted or unreadable when the fs crahses
(Essentially, fsck gives me a HUGE list of addreses on the ext2 fs that is
And finally, nope, my controller has not had any problems with previous kernels
Can you try this with a Wolverine kernel to see if that makes any difference?
You don't need to upgrade any of the other components. Email me at
firstname.lastname@example.org if you need a hand getting the new kernel up.
Okay, i just re-installed fisher and downloaded the wolverine kernel from
ftp.redhat.com. I updated the kernel with the new one. The system looks fine. I
added a line to the rc.local file to reboot the computer, so the OS will go on
an infinite reboot loop :P
I'll check on it times to times to see if it crashes again with the new kernel.
Ok, I was away for the last few days and left the computer running in it's
infinite reboot loop. Fine! The system has not crashed yet :P