Mailing List Archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[tlug] Replacing software-raided disks with dd, underlying-partitions-and-all: Will this work?



I've got a Linux server (which I'll call "server 1") with a pair of
disks divided up into partitions, then the partitions paired up into
RAID 1 partitions using software RAID. (No LVM in between or
anything.)

cat /proc/mdstat
Personalities : [raid1]
md3 : active raid1 sdb2[1] sda2[2](F)
      8193024 blocks [2/1] [_U]

md5 : active raid1 sdb3[1] sda3[2](F)
      4096448 blocks [2/1] [_U]

md6 : active raid1 sdb5[1] sda5[2](F)
      4096448 blocks [2/1] [_U]

md2 : active raid1 sdb6[1] sda6[2](F)
      2048192 blocks [2/1] [_U]

md1 : active raid1 sdb7[1] sda7[2](F)
      2048192 blocks [2/1] [_U]

md4 : active raid1 sdb8[1] sda8[2](F)
      57512576 blocks [2/1] [_U]

md0 : active raid1 sdb1[1] sda1[2](F)
      128384 blocks [2/1] [_U]

As you can see, the /dev/sda disk is dead and has fallen out of the RAID array.

I also have another Linux server ("server 2") in exactly the same
predicament, with the same sized disks, only a slightly different
partition arrangement.

Dell no longer sell disks of the original size, so I thought I'd take
the opportunity to buy a new pair of bigger disks top put in server 1,
and move the surviving disk from server 1 to server 2.

This sounds like it should be easy, but have I misunderstood anything
about how software RAID 1 works, or are there any dd / partitioning /
RAID 1 gotchas I should be aware of?

I'm thinking:

1) Check my backups

2) Replace the dead disk (sda) in server 1 with a new, bigger disk,
boot from a LiveCD and do:

dd if=/dev/sdb of=/dev/sda

3) Reboot server 1, which I expect to be magically working with both
disks back in the RAID array, without even needing to tell it I've
recovered the disk. (I'm not sure exactly why I think that will work,
but I do.)

4) Replace the originally alive disk (sdb) in server 1 with the other
new, bigger one, boot from a LiveCD and do:

dd if=/dev/sda of=/dev/sdb

5) Reboot server 1, and find everything magically working, as in step (3).

6) Make a new partition on each disk out of the empty space, then
RAID1 them together.

7) Replace the dead disk (sda) in server 2 with the live disk I took
out of server 1, boot from a LiveCD and do:

dd if=/dev/sdb of=/dev/sda

8) Reboot server 2, and find everything magically working, as in steps
(3) and (5).


-- 
Edmund Edgar
Founder, KK Social Minds
Educational Technology for the Web and Virtual Worlds

ed@example.com
+81 090 3912 3380
Skype: edmundedgar
Second Life: Edmund Earp
Linked In: edmundedgar
Twitter: @edmundedgar
http://www.socialminds.jp


Home | Main Index | Thread Index

Home Page Mailing List Linux and Japan TLUG Members Links