Red Hat Bugzilla – Bug 117354
(SCSI MD)error when using software RAID-1 + LVM
Last modified: 2007-11-30 17:10:37 EST
Description of problem:
Sporadically get following error message:
sym53c895-0:1: ERROR (81:0) (8-0-0) (f/9f) @ (script 2a0:48000000).
sym53c895-0: script cmd = f25c0004
sym53c895-0: regdump: da 00 00 9f 47 0f 01 0a 00 08 81 00 80 00 0f 02
00 a8 83 1f 02 ff ff ff.
sym53c895-0: Downloading SCSI SCRIPTS.
I'm using both software RAID and LVM -- the LVM is built on top of a
RAID-1 drive pair. I've only noticed this problem when accessing an
LVM partition: /home and /usr/local are logical volumes on /dev/md5,
/boot is ext3 on /dev/md0. Running fsck -c will throw the error about
1 time in 3 on one of the lv's, but has yet to do so on a non-lvm
I suspect this is NOT a hardware problem, as I had just scanned both
drives for bad blocks prior to repartitioning and installing FC1.
This system (an HP LPr PII/450 512M 2x9.1G SCSI) was previously
running RH 7.3 with a straight ext3 partitioning scheme -- no RAID or
LVM -- without exhibiting this problem.
Version-Release number of selected component (if applicable):
Random / sporadic
Steps to Reproduce:
1. umount /home ; fsck -c /dev/Volume00/lv_home
Sometimes it throws the error message, sometimes it doesn't.
I've done everything I can think of to rule out hardware errors.
Re-installed FC1 on the affected box on 3/13/04 using LVM but without
software RAID. Problem seems to have gone away. This leads me to
believe that the problem was caused by the software RAID.
Thanks for the bug report. However, Red Hat no longer maintains this version of
the product. Please upgrade to the latest version and open a new bug if the problem
The Fedora Legacy project (http://fedoralegacy.org/) maintains some older releases,
and if you believe this bug is interesting to them, please report the problem in
the bug tracker at: http://bugzilla.fedora.us/