Re: [SCARED] Is ext2 unreliable?

From: Sasi Peter (sape@iq.rulez.org)
Date: Tue May 16 2000 - 10:04:20 EST


On Tue, 16 May 2000, Alan Cox wrote:

> 1. You are using usb patches

For my mouse, yes

> 2. You are using raid patches

For all of my partitions, 'cause they are striped/mirrored on the
different disks.

> 3. You are using ide patches.

Because the machine is loaded, and I nedd DMA for all of the drives + I
have a CDROM too.

> It could be a bug in any of the above. It could be a problem with the disk.
> It could be a problem with the ram. It could be a problem with the cabling.

I do not know too much about the disk layout of ext2, but what I have
written about the fsck run makes me think it should be a kernel bug
(errors are somewhat consistent, like zeroed counts, and all blocks bitmap
diferring)

> It could be a bug in the mmainstream kernel.

I think so too, and that is why I am writing here.

> You can eliminate some of these by load testing the machine (memtest86
> then kernel building). Try running without the usb/raid patches. etc

I cannot run it w/o the patches, but the box is under heavy pressure all
the time, ~160 samba connections and ~80 NFS connections concurrently,
doing ~6.6MBytes/s sustained transfer rate on the eth. interface.

-- SaPE

Peter, Sasi <sape@sch.hu>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue May 23 2000 - 21:00:10 EST