Very small bug report:
mdadm assemble --force fails to assemble at the very last step after
having adjusted the event count:
root@storage1:/root# mdadm --assemble /dev/md106 --config
/etc/mdadm/mdadm.secondary.conf --force
mdadm: forcing event count in
/dev/mapper/mpathid-350014ee0ac5a112e-part1(3) from 1058 upto 1061
mdadm: forcing event count in
/dev/mapper/mpathid-350014ee056feaaf4-part1(4) from 1058 upto 1061
mdadm: forcing event count in
/dev/mapper/mpathid-350014ee001a93ede-part1(6) from 1058 upto 1061
mdadm: clearing FAULTY flag for device 1 in /dev/md106 for
/dev/mapper/mpathid-350014ee0ac5a112e-part1
mdadm: clearing FAULTY flag for device 5 in /dev/md106 for
/dev/mapper/mpathid-350014ee056feaaf4-part1
mdadm: clearing FAULTY flag for device 8 in /dev/md106 for
/dev/mapper/mpathid-350014ee001a93ede-part1
mdadm: Marking array /dev/md106 as 'clean'
mdadm: /dev/md106 assembled from 8 drives - not enough to start the array.
mdadm is wrong here: 8 devices is indeed the total number of devices for
this array.
the last command started it inactive like this:
md106 : inactive dm-101[0](S) dm-24[9](S) dm-107[6](S) dm-103[5](S)
dm-123[4](S) dm-95[3](S) dm-105[2](S) dm-106[1](S)
4193255424 blocks super 1.2
(maybe the blocks count is also wrong? Not adjusted for parity maybe?
4193255424 / 8.0 * 6.0 == 3144941568.0 almost correct)
it can be fixed simply by doing:
root@storage1:/root# mdadm --stop /dev/md106
mdadm: stopped /dev/md106
root@storage1:/root# mdadm --assemble /dev/md106 --config
/etc/mdadm/secondary.conf --no-degraded
mdadm: /dev/md106 has been started with 8 drives.
now it works:
md106 : active raid6 dm-101[0] dm-24[9] dm-107[6] dm-103[5] dm-123[4]
dm-95[3] dm-105[2] dm-106[1]
3144938496 blocks super 1.2 level 6, 512k chunk, algorithm 2
[8/8] [UUUUUUUU]
bitmap: 0/1 pages [0KB], 1048576KB chunk
Regards
EW
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
