Red Hat Bugzilla – Bug 141091
[PATCH] disk ordering fails raidstart
Last modified: 2007-11-30 17:07:05 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030922
Description of problem:
Reported via IT#53591
Customer: IBM-IGS/HSBC Bank Brazil
If you have a mirrored raid device ( raid level 1 ), and one disk
encounters fatal errors, the raid device continues to operate on the
other disk, as expected.
However if the device is stopped and restarted then the raidstart will
fail if the failing hardware is the first listed in the raidtab ( e.g.
raid-disk 0 ).
If the first raid-disk is operational, then the md device will start
irrespective of the status of the second raid-disk.
The customer encountered the problem on an EMC array; it is also
recreated using my test machine on both AS2.1 and RHEL 3.
The problem seems to be with raidstart, which only attempts the first
listed disk irrespective of the raid level.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Create an md device using 2-disk mirroring (say sdb1, sda1). Make
sure sdb1 is raid-disk 0.
2. Start the md device as usual - proves everything works.
3. Shutdown the system and remove sdb1.
4. Reboot and run raidstart. It will fails.
Actual Results: The raidstart command will fail since it can't find sdb1.
Expected Results: The md should be able to pick up sda1 and operational.
Created attachment 107526 [details]
raidstart patch to fix this issue.
Patch submitted by email@example.com and tested by me. Red Hat TAM is Mike
AS2.1 bugzilla: 141095.
Anyword on this issue ?
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.