All,
Here's a reprise of this ancient message. But there's method
in my madness. We need to do just this and are hitting a wall.
We have stock RedHat 6.0 with the 18 disk patch by Lance
Robinson, and no other kernel, module or raidtools changes. We
tried removing disk fifteen (sdo, scsi 2 0 2 0) in an attempt
to recover from a down disk without rebooting.
To force it to recognize the disk loss I ran
dd bs=1024 count=100 skip=1000000 if=/dev/md0 of=/dev/null
and it kept trying to rebuild and complaining no disk there,
and finally gave up hanging that terminal (a "D" in the ps ax).
At 01:12 AM 7/8/99 +0100, you wrote:
>On 7 Jul 1999, m.allan noah wrote:
>
> > 5. Reboot to get linux to recognize the new partition
>
> you hit the nail on the head man. no need to copy things to the replacement
> disk. just get the partition table right and readable by linux (requires
the
> reboot in 5). then use raidhotadd.
>
>reboot.. yuk! this is linux not nt! :)
>
>what you do is:
>
>1. unmount everything that's directly on the disk that isn't raid.
>2. swapoff if you have swap on that disk
Didn't understand this. Our system is on hda and we can't
unmount it.
>3:
>
>echo "scsi remove-single-device c b t l" > /proc/scsi/scsi
Tried it and it didn't work. But I noticed /proc/mdstat had
an (F) after sdo1. So I tried from another terminal
raidhotremove /dev/md0 /dev/sdo1
which hung THAT terminal and gave me a D in ps ax.
THEN I tried
echo "scsi remove-single-device 2 0 2 0" >/proc/scsi/scsi
and that worked; it was gone from /proc/scsi/scsi.
>
>where c=controller, b=bus/channel, t=target/scsi id, l=lun of the
>failed disk. Eg disk id-3 on the first channel of the first scsi
>controller would be:
>
>echo "scsi remove-single-device 0 0 3 0" > /proc/scsi/scsi
>
>4. replace failed disk
>5. get linux to see new disk:
>
>echo "scsi add-single-device c b t l" > /proc/scsi/scsi
That worked. New disk at end of /proc/scsi/scsi list.
>
>(replace c b t l as in above example)
>
>6. partition new disk.
No dice. fdisk hung the terminal and gave the "D" in ps ax.
>7. raidhotadd.
Trying this one gave an "S" instead of a "D" and kill -9 replied
with
ioctl lock interrupted, reason-4, cmd 2344
>
>We never.. never.. never.. ever.. reboot linux!!
I guess we will have to - which is really a bummer. We were
hoping to make our RAID product more friendly to customers.
Larry Dickson
PS. We would really appreciate, if possible, some route that
does NOT require massive kernel recompiles and leading-edge
alpha code.
>
>(well unless the hardware is on fire or something. then ... maybe)
>
>:)
>
>--
>Paul Jakma
>[EMAIL PROTECTED] http://hibernia.clubi.ie
>PGP5 key: http://www.clubi.ie/jakma/publickey.txt
>-------------------------------------------
>Fortune:
>Sauron is alive in Argentina!
>
>
