From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.2.1) Gecko/20030225 Description of problem: There is a bug is the function super_90_load() in md.c Let me paste the required code change first and then explain what is happening, @@ -582,22 +582,24 @@ static int super_90_load(mdk_rdev_t *rde if (calc_sb_csum(sb) != sb->sb_csum && calc_sb_csum_common(sb) != sb->sb_csum) { printk(KERN_WARNING "md: invalid superblock checksum on %s\n", b); goto abort; } rdev->preferred_minor = sb->md_minor; rdev->data_offset = 0; - - if (sb->level == MULTIPATH) + if (sb->level == LEVEL_MULTIPATH) rdev->desc_nr = -1; else rdev->desc_nr = sb->this_disk.number; As you see it is checking for MULTIPATH instead of LEVEL_MULTIPATH, rdev->desc_nr will be set to sb->this_disk.number which will be the same for each path (this should actually be set to -1). Eventually bind_rdev_to_array will fail to bind any but the first path into the array. This bug has been fixed in 2.6.12-rc2. Version-Release number of selected component (if applicable): kernel-2.6.9-5.Elsmp How reproducible: Always Steps to Reproduce: 1.Create a multipath array 2.Stop the array 3.Assemble the array back into service. I use mdadm for the purpose but I guess raidtools would too expose the bug. Additional info:
I just upgraded my system to RHEL4 U1 and found that this bug has been fixed there.