> I agree, but so far we haven't limited things down very much. One extremely
> important thing I think people need to keep in mind is the following. Your
> choice of low-level controller driver for the attached hard drives can greatly
> impact this problem. I had a machine with an IDE drive and a Triton II
> chipset. When I enabled the Triton DMA support, I would get corruption,
> reliably, and under light load even. When I disabled the special drivers, my
> corruption went away. The same could very well be true of SCSI drivers (you
> could end up getting corruption if you use the PAGE_ENABLE flag in the aic7xxx
> driver, or possibly some other scsi controllers, but I wouldn't know about
> them). I think people who are reporting filesystem corruption under ext2 fs
> need to start also reporting the hard drive controller and driver in use so
> some of us can look for any patterns in what's happening.
Yes, of course. I'm sorry; I should have included more information. Here
it is (I hope the problem is just a driver issue :)
express:/proc/scsi$ cat scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: SEAGATE Model: ST31200N SUN1.05 Rev: 8722
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 01 Lun: 00
Vendor: SEAGATE Model: ST12550N Rev: 0013
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 02 Lun: 00
Vendor: SEAGATE Model: ST15150W Rev: 0020
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 03 Lun: 00
Vendor: SEAGATE Model: ST15150W Rev: 0023
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 05 Lun: 00
Vendor: HP Model: T4000s Rev: 1.06
Type: Sequential-Access ANSI SCSI revision: 02
SCSI Adapter:
BusLogic BT-956C
Bus 0, device 14, function 0:
SCSI storage controller: BusLogic MultiMaster (rev 6).
Fast devsel. IRQ 11. Master Capable. Latency=240.
I/O at 0x6000.
The drive arrangement is different than I thought (I'm not the sysadmin,
sadly, so I'm doing this in my spare time).
Filesystem 1024-blocks Used Available Capacity Mounted on
/dev/sda2 932549 344980 539390 39% /
/dev/sdb1 2020873 1240275 676151 65% /home
/dev/sdc1 4052132 3267785 574701 85% /var/spool/dnews
/dev/sdd1 4052132 3361956 480530 87% /var/spool2
/var/spool2 is running a small caching server, for experimental purposes.
I am very interested in this thread, naturally, so if I can provide any
other information I'll be right on it.
ejt