Red Hat Bugzilla – Bug 198671
slab corruption / slab error in slab_destroy_objs(): cache `bio'
Last modified: 2007-11-30 17:11:37 EST
Description of problem:
Seen on one of my HP ia64 Integrity servers running a proprietary HP test suite.
Since all the messages are rather lengthy I wil upload it as an attachment.
But here are some of what think are the more interesting messages:
Slab corruption: (Not tainted) start=e0000040765c2828, len=104
kernel BUG at mm/slab.c:608!
kjournald: bugcheck! 0 
slab error in slab_destroy_objs(): cache `bio': start of a freed object was
Version-Release number of selected component (if applicable):
This took about 10 hours of stress testing before it was seen. I will update
this BZ if I see it again.
Steps to Reproduce:
Created attachment 132317 [details]
kernel messages from slab corruption runing the "busy2" test on hpdiablo1
We haven't seen this again, have we? Doug, any objections to closing this out?
I have not seen this in a while, yes I am OK with closing it.