Bug 785182 - DM RAID: Performing multiple suspend/resume cycles on a dm RAID device causes memory corruption
Summary: DM RAID: Performing multiple suspend/resume cycles on a dm RAID device causes...
Keywords:
Status: CLOSED DUPLICATE of bug 789409
Alias: None
Product: Red Hat Enterprise Linux 6
Classification: Red Hat
Component: kernel
Version: 6.2
Hardware: Unspecified
OS: Unspecified
high
medium
Target Milestone: rc
: ---
Assignee: Jonathan Earl Brassow
QA Contact: Red Hat Kernel QE team
URL:
Whiteboard:
Depends On:
Blocks: 732458 739162
TreeView+ depends on / blocked
 
Reported: 2012-01-27 14:58 UTC by Jonathan Earl Brassow
Modified: 2012-08-27 14:54 UTC (History)
5 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2012-02-10 22:22:21 UTC
Target Upstream Version:

Attachments (Terms of Use)
my crashlog. The kernel is 2.6.32-220.4.1.el6.x86_64 (4.89 KB, text/plain; charset=ISO-8859-2)
2012-02-06 09:58 UTC, Mikuláš Patočka 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Description Jonathan Earl Brassow 2012-01-27 14:58:49 UTC 
Steps to reproduce:
1) lvcreate --type raid1 -m 1 -L 500M -n lv vg
2) dmsetup suspend vg-lv
3) dmsetup resume vg-lv  ** KERNEL OPPS **

It may be necessary to repeat steps 2 & 3.
Comment 2 Mikuláš Patočka 2012-02-06 09:56:56 UTC 
I managed to reproduce the bug. In function bitmap_daemon_work, just after the "/* skip this page unless it's marked as needing cleaning */" comment, page is a wild pointer, pointing to 0x0000006f0000006f. This should be forwarded to the developers of md subsystem, because they have best knowledge of their code.
Comment 3 Mikuláš Patočka 2012-02-06 09:58:11 UTC 
Created attachment 559583 [details]
my crashlog. The kernel is 2.6.32-220.4.1.el6.x86_64
Comment 5 Jonathan Earl Brassow 2012-02-10 22:22:21 UTC 
Missed this bug when creating bug 789409.  That bug has all the acks already and i've described the patch I sent upstream there.

*** This bug has been marked as a duplicate of bug 789409 ***
Note You need to log in before you can comment on or make changes to this bug.