Hi all
Am running into some problems with gentoo on sparc (TI UltraSparc II
(BlackBird). Am trying to convert a sun server in the office to be the
corporate email server. Everything works flawlessly with the exception
of software raid (raid 1). Riad5 just did not work !!
ok mmkraid worked as expected
>>>>
web etc # mkraid /dev/md0
handling MD device /dev/md0
analyzing super-block
>>>
proc info seems to be ok
>>>
web etc # cat /proc/mdstat
Personalities : [raid1] [raid5]
read_ahead 1024 sectors
md0 : active raid1 scsi/host0/bus0/target11/lun0/part1[2]
scsi/host0/bus0/target10/lun0/part1[1] scsi/host0/bus0/target9/lun0/part1[0]
8079744 blocks [2/2] [UU]
[>....................] resync = 0.9% (77632/8079744)
finish=25.7min speed=5175K/sec
unused devices: <none>
>>>
raidstop is also ok
>>>
web etc # web etc # raidstop /dev/md0
>>>
Proc after raid stop
>>>
web etc # cat /proc/mdstat
Personalities : [raid1] [raid5]
read_ahead 1024 sectors
unused devices: <none>
>>>
So far so good but on any reboot the raid does not come up(fsck prompt).
on raidstart, i get some funky error message
>>>
web etc # raidstart /dev/md0
/dev/md0: Invalid argument
web etc # raidstart -a
/dev/md0: Invalid argument
web etc #
>>>
My data is not lost (uhmm i think) It's just sitting thre prettily
(Thinkin again). The only way to start up the raid seems to be....
>>>
web etc # mkraid /dev/md0
handling MD device /dev/md0
analyzing super-block
web etc # cat /proc/mdstat
Personalities : [raid1] [raid5]
read_ahead 1024 sectors
md0 : active raid1 scsi/host0/bus0/target11/lun0/part1[2]
scsi/host0/bus0/target10/lun0/part1[1] scsi/host0/bus0/target9/lun0/part1[0]
8079744 blocks [2/2] [UU]
[>....................] resync = 2.7% (219200/8079744)
finish=25.4min speed=5146K/sec
unused devices: <none>
web etc #
web etc # mount /dev/md0 /var/virtual/
web etc # ls /var/virtual/
lost+found mailbox
web etc #
>>>
I can change the init scripts to call mkraid instead of raidstart but
what if mkraid does at some point later trash my data ?? maybe on a
kernel or raidtools upgrade? Any one see something that i missed?
btw here's my
uname -a
Linux foo.bar.co.ke 2.4.26 #4 SMP Tue Jun 8 19:16:40 EAT 2004
sparc64 sun4u TI UltraSparc II (BlackBird) GNU/Linux
mkraid --version
mkraid version 0.90.0
Thanx
Lmwangi
--
gentoo-sparc@gentoo.org mailing list
Am running into some problems with gentoo on sparc (TI UltraSparc II
(BlackBird). Am trying to convert a sun server in the office to be the
corporate email server. Everything works flawlessly with the exception
of software raid (raid 1). Riad5 just did not work !!
ok mmkraid worked as expected
>>>>
web etc # mkraid /dev/md0
handling MD device /dev/md0
analyzing super-block
>>>
proc info seems to be ok
>>>
web etc # cat /proc/mdstat
Personalities : [raid1] [raid5]
read_ahead 1024 sectors
md0 : active raid1 scsi/host0/bus0/target11/lun0/part1[2]
scsi/host0/bus0/target10/lun0/part1[1] scsi/host0/bus0/target9/lun0/part1[0]
8079744 blocks [2/2] [UU]
[>....................] resync = 0.9% (77632/8079744)
finish=25.7min speed=5175K/sec
unused devices: <none>
>>>
raidstop is also ok
>>>
web etc # web etc # raidstop /dev/md0
>>>
Proc after raid stop
>>>
web etc # cat /proc/mdstat
Personalities : [raid1] [raid5]
read_ahead 1024 sectors
unused devices: <none>
>>>
So far so good but on any reboot the raid does not come up(fsck prompt).
on raidstart, i get some funky error message
>>>
web etc # raidstart /dev/md0
/dev/md0: Invalid argument
web etc # raidstart -a
/dev/md0: Invalid argument
web etc #
>>>
My data is not lost (uhmm i think) It's just sitting thre prettily
(Thinkin again). The only way to start up the raid seems to be....
>>>
web etc # mkraid /dev/md0
handling MD device /dev/md0
analyzing super-block
web etc # cat /proc/mdstat
Personalities : [raid1] [raid5]
read_ahead 1024 sectors
md0 : active raid1 scsi/host0/bus0/target11/lun0/part1[2]
scsi/host0/bus0/target10/lun0/part1[1] scsi/host0/bus0/target9/lun0/part1[0]
8079744 blocks [2/2] [UU]
[>....................] resync = 2.7% (219200/8079744)
finish=25.4min speed=5146K/sec
unused devices: <none>
web etc #
web etc # mount /dev/md0 /var/virtual/
web etc # ls /var/virtual/
lost+found mailbox
web etc #
>>>
I can change the init scripts to call mkraid instead of raidstart but
what if mkraid does at some point later trash my data ?? maybe on a
kernel or raidtools upgrade? Any one see something that i missed?
btw here's my
uname -a
Linux foo.bar.co.ke 2.4.26 #4 SMP Tue Jun 8 19:16:40 EAT 2004
sparc64 sun4u TI UltraSparc II (BlackBird) GNU/Linux
mkraid --version
mkraid version 0.90.0
Thanx
Lmwangi
--
gentoo-sparc@gentoo.org mailing list