Re: ext2 filesystem corruption?!?!??

Dave Wreski (
Sat, 5 Apr 1997 22:57:45 -0500 (EST)


I don't think we have someone who is keeping track of the reports that
(hopefully) will come in regarding working and non working configuration
with regard to ext2 corruption. Do you have a web page to keep these on,
and know the people that should be contacted for them to reference when
doing testing/development (Ted Ts'o and Remy Card?) ?

Anyway, my P180 doesn't have this problem. I modified your one-hellova
wicked 'script' to run infinitely. I started it last night, and stopped
it after 10hours:



crashme () {
badblocks ... &
badblocks ... &
badblocks ... &
badblocks ...

while [ 1 ]
i=`expr $i + 1`
echo "Iteration: $i `uptime`"

Here is a known working configuration. Please let me know if this is
sufficient information, ie, more info on Award BIOS options. I don't have
access to the box that was crashing for me during the middle of last month
(was off-hours cdrom/file/atalk/NT consulting job)

- kernel 2.0.29-ISS4
- 64 meg 60ns EDO
- PPro 180
- NE2000 clone
- RedHat 4.0/4.1 mix (4.0 libs, 4.1 updates)
- Load averaged 4.5-5.0 (peaks of 10-12)
- APM disabled in BIOS
- Jaz external connected to AHA1510 (50pin Fast SCSI connection)
- Micropolis 1.2g attached to AHA2940uw as well, but tests run only on
- AHA1510 and NE2k as modules, everything else compiled in kernel
- Acer AP65 1.20 Award BIOS 440FX

processor : 0
cpu : 686
model : Pentium Pro
vendor_id : GenuineIntel
stepping : 7
fdiv_bug : no
hlt_bug : no
fpu : yes
fpu_exception : yes
cpuid : yes
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 11 mtrr pge mca cmov
bogomips : 179.40

Host: scsi0 Channel: 00 Id: 01 Lun: 00
Vendor: SEAGATE Model: ST32151W Rev: 0322
Type: Direct-Access ANSI SCSI revision: 02

Adaptec AIC7xxx driver version: 4.0/3.2/4.0

Compile Options:

Adapter Configuration:
SCSI Adapter: AHA-2940 Ultra
(AIC-788x chipset)
Host Bus: Wide
Base IO: 0x6000
IRQ: 15
SCBs: Used 4, HW 16, Page 16
Interrupts: 1841642
Serial EEPROM: True
Extended Translation: Disabled
SCSI Bus Reset: Enabled
Ultra SCSI: Enabled
Target Disconnect: Enabled

Hope thats adequate. Also, I've never really been involved in
bug-reports/bug-tracing. I would assume this takes precendence over
further development until this is resolved?

It sure is weird that it only happens to some. Even some people who run
news feeds don't exhibit this problem..