Skip to contentSkip to conversation

Memory Corruption

DISCUSSION Chris Mason 2 participants 2 messages Aug 29, 2002 – Sep 8, 2002 VFS

lore
Share & tools
lore
#0 lore raw
Quoted 11 lines · David Boutcher
Quoted 11 lines · Dave Boutcher
Quoted 8 lines

Hi,

I'm chasing a wierd memory corruption problem on a ppc64 system. The first byte of a slab_t structure keeps getting stepped on (zeroed, actually.) This happens during a testcase that copies a large file called "junk" between file systems (a mix of ext2 and reiser) on a 2.4.13 kernel. I know that's REALLY REALLY old, but it's whats in SuSE's SLES-7 release that we have customers running...

Quoted 8 lines
Quoted 3 lines

Any chance the test case involves renames?

-chris

So I posted my problem with memory corruption a few weeks ago....and the problem turned out to be a REALLY old/moldy set of userland reiser tools. I don't know exactly why that caused memory corruption in the kernel, but updating the tools fixed everything right up.

Well, that shouldn't fix it ;-) Which version of reiserfsprogs were you running before?

Are the filesystems getting checked during boot at all (you would see reiserfsck messages during boot)?

-chris

#1 lore raw

Chris Mason
<[email protected]> To: David Boutcher/Rochester/IBM@IBMUS
cc: [email protected], [email protected]
08/29/2002 08:40 Subject: Re: [reiserfs-list] Memory Corruption
AM

Quoted 20 lines · David Boutcher
Quoted 11 lines
Quoted 10 lines · Dave Boutcher
Quoted 2 lines

Hi,

I'm chasing a wierd memory corruption problem on a ppc64 system. The first byte of a slab_t structure keeps getting stepped on (zeroed, actually.) This happens during a testcase that copies a large file called "junk" between file systems (a mix of ext2 and reiser) on a 2.4.13 kernel. I know that's REALLY REALLY old, but it's whats in SuSE's SLES-7 release that we have customers running...

Quoted 8 lines

Any chance the test case involves renames?

-chris

So I posted my problem with memory corruption a few weeks ago....and the problem turned out to be a REALLY old/moldy set of userland reiser tools. I don't know exactly why that caused memory corruption in the kernel, but updating the tools fixed everything right up.

Quoted 5 lines

Well, that shouldn't fix it ;-) Which version of reiserfsprogs were you running before?

Are the filesystems getting checked during boot at all (you would see reiserfsck messages during boot)?

The fixed tools show:
[root@linux102 /root]# reiserfsck -V

<-------------reiserfsck, 2002------------->
reiserfsprogs 3.6.3

The broken tools show:
[root@linux102 /root]# /sbin/reiserfsck.org -V

<-------------reiserfsck, 2002------------->
reiserfsprogs 3.x.1a

[root@linux102 /root]# ls -l /sbin/reiserfsck*

-rwxr-xr-x    1 root     root       719431 Aug 21 10:16 /sbin/reiserfsck*
-rwxr-xr-x    1 root     root       265740 Feb 28  2002

/sbin/reiserfsck.org*
[root@linux102 /root]#
This was Turbolinux...the broken tools are the ones shipped with their 6.5 distribution.

Dave B