File system corruption under 2.4.21-pre5-ac1
I'm seeing filesystem corruption on a number of intel SE7501wv2's
under 2.4.21-pre5-ac1. The systems are running Cerberus (ctcs). They fail the kcompile, and memtst tests.
At 1st I thought it was memory, but the systems are fine under
memtest86, and the Cerberus memtst failures disappear if I turn off my swap devices. The kcompile failures seem to point to random file corruption from copying the kernel tree. (Cerberus copies a new copy of the kernel for each compile.)
I suspect the aic79xx driver as:
1)I've running ctcs with the same kernel on 100+ SE7501wv2 with ide drives. 2)This occurs under reiserfs, and ext3.
3)I've been using 2.4.21-pre5-ac1 to burnin systems for quite some time.
Signature · 4 lines
There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory <[email protected]>
Quoted 3 lines
I'm seeing filesystem corruption on a number of intel SE7501wv2's under
2.4.21-pre5-ac1. The systems are running Cerberus (ctcs). They fail the kcompile, and memtst tests.
Are you running the aic79xx driver version embedded in that kernel version or the latest from my site?
http://people.FreeBSD.org/~gibbs/linux/SRC/
--
Justin
Quoted 10 lines · Justin T. Gibbs
Quoted 4 lines
I'm seeing filesystem corruption on a number of intel SE7501wv2's
under
2.4.21-pre5-ac1. The systems are running Cerberus (ctcs). They fail the kcompile, and memtst tests.Are you running the aic79xx driver version embedded in that kernel version or the latest from my site?
That seems to have fixed the issues. All of the systems burned in
over night with out issue. Thanks.
Signature · 4 lines
There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory <[email protected]>