I have what is probably a weird RAID setup. I have one plain RAID-0
array (/dev/md1), one plain linear array (/dev/md0), and another RAID-0
array (/dev/md2) consisting of a disk and the linear array.
It appears that what is happening is that linux tries to initialize
/dev/md2 before /dev/md0 is fully initialized. Of course, since
/dev/md0 doesn't exist yet, linux says "too few disks (1 of 2) -
aborting!"
After a boot, I have to run "raidstop /dev/md2 ; raidstart /dev/md2" in
order to make md2 work properly. I have never had to do this before under
any 2.6 kernel. The last working kernel I used was 2.6.1-mm2. BTW, I
compile everything into the kernel and don't use modules for RAID.
Below is my /etc/raidtab and the relevant parts of my dmesg output.
Please let me know if you need any more information.
thanks,
Jim Faulkner
my /etc/raidtab
-------------------------------------
raiddev /dev/md0
raid-level linear
nr-raid-disks 2
chunk-size 128
persistent-superblock 1
device /dev/hdg1
raid-disk 0
device /dev/hdh1
raid-disk 1
raiddev /dev/md1
raid-level 0
nr-raid-disks 2
chunk-size 8
persistent-superblock 1
device /dev/hda1
raid-disk 0
device /dev/hde1
raid-disk 1
raiddev /dev/md2
raid-level 0
nr-raid-disks 2
chunk-size 4
persistent-superblock 1
device /dev/hdc1
raid-disk 0
device /dev/md0
raid-disk 1
-------------------------------------
kernel output during boot-time raid detection
---------------------------------------------
md: Autodetecting RAID arrays.
md: autorun ...
md: considering hdh1 ...
md: adding hdh1 ...
md: adding hdg1 ...
md: hde1 has different UUID to hdh1
md: hdc1 has different UUID to hdh1
md: hda1 has different UUID to hdh1
md: created md0
md: bind<hdg1>
md: bind<hdh1>
md: running: <hdh1><hdg1>
md: considering hde1 ...
md: adding hde1 ...
md: hdc1 has different UUID to hde1
md: adding hda1 ...
md: created md1
md: bind<hda1>
md: bind<hde1>
md: running: <hde1><hda1>
md1: setting max_sectors to 16, segment boundary to 4095
raid0: looking at hde1
raid0: comparing hde1(117218176) with hde1(117218176)
raid0: END
raid0: ==> UNIQUE
raid0: 1 zones
raid0: looking at hda1
raid0: comparing hda1(117218176) with hde1(117218176)
raid0: EQUAL
raid0: FINAL 1 zones
raid0: done.
raid0 : md_size is 234436352 blocks.
raid0 : conf->hash_spacing is 234436352 blocks.
raid0 : nb_zone is 1.
raid0 : Allocating 4 bytes for hash.
md: considering hdc1 ...
md: adding hdc1 ...
md: created md2
md: bind<hdc1>
md: running: <hdc1>
md2: setting max_sectors to 8, segment boundary to 2047
blk_queue_segment_boundary: set to minimum fff
raid0: looking at hdc1
raid0: comparing hdc1(60026752) with hdc1(60026752)
raid0: END
raid0: ==> UNIQUE
raid0: 1 zones
raid0: FINAL 1 zones
raid0: too few disks (1 of 2) - aborting!
md: pers->run() failed ...
md :do_md_run() returned -22
md: ... autorun DONE.
---------------------------------------------------
kernel output from when I run "raidstop /dev/md2 ; raidstart /dev/md2"
after boot-up
------------------------------------------------------
md: md2 stopped.
md: unbind<hdc1>
md: export_rdev(hdc1)
md: autorun ...
md: considering md0 ...
md: adding md0 ...
md: adding hdc1 ...
md: created md2
md: bind<hdc1>
md: bind<md0>
md: running: <md0><hdc1>
md2: setting max_sectors to 8, segment boundary to 2047
blk_queue_segment_boundary: set to minimum fff
raid0: looking at md0
raid0: comparing md0(63512640) with md0(63512640)
raid0: END
raid0: ==> UNIQUE
raid0: 1 zones
raid0: looking at hdc1
raid0: comparing hdc1(60026752) with md0(63512640)
raid0: NOT EQUAL
raid0: comparing hdc1(60026752) with hdc1(60026752)
raid0: END
raid0: ==> UNIQUE
raid0: 2 zones
raid0: FINAL 2 zones
raid0: zone 1
raid0: checking hdc1 ... nope.
raid0: checking md0 ... contained as device 0
(63512640) is smallest!.
raid0: zone->nb_dev: 1, size: 3485888
raid0: current zone offset: 63512640
raid0: done.
raid0 : md_size is 123539392 blocks.
raid0 : conf->hash_spacing is 120053504 blocks.
raid0 : nb_zone is 2.
raid0 : Allocating 8 bytes for hash.
md: ... autorun DONE.
----------------------------------------------------
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
