mdadm ( 2xSATA ) RAID 1, Fedora Core 5 - Not keeping system running

G'day All,
I've created a RAID1 with 3 partitions on it, with 2 SATA HD's 250Gig.

[root@localhost /]# cat /proc/mdstat
Personalities : [raid1]
md1 : active raid1 sda1[0] sdb1[1]
      104320 blocks [2/2] [UU]
     
md2 : active raid1 sdb2[1] sda2[0]
      1028096 blocks [2/2] [UU]
     
md0 : active raid1 sda3[0] sdb3[1]
      243063360 blocks [2/2] [UU]
     
unused devices: <none>
[root@localhost /]#

To what I can tell they seemed fine, but when I unplug one of the hard drives (SATA) to simulate a drive failure, the whole system stops (freezes). And it only comes back when I plug it back in. Checked the RAID Status and is fine.
My understanding of a RAID 1 is that because the two Hard drives are Mirrored no matter which hard drive I pull the System should still be operational.

All I want is Hard Drive redundancy in the case of a drive failure.
Thank you in advance.


Cheers,
Arthur
LVL 6
Arthur_MinoAsked:
Who is Participating?
 
rindiCommented:
This only works reliably if you are using a real hardware raid controller. With software raid you first boot a non raid system, so if you remove the wrong disk from the array, it can't be seen by the bootloader or kernel... One possibility would be to have the OS itself on a non raid system, and use the raid disks for the data, this wold ensure that raid would be loaded after the OS. Maybe the following links can help.

http://www.linuxdevcenter.com/pub/a/linux/2002/12/05/RAID.html
http://tldp.org/HOWTO/ATA-RAID-HOWTO/index.html
http://tldp.org/HOWTO/Boot+Root+Raid+LILO.html
0
 
Arthur_MinoAuthor Commented:
G'day Rindi,

what do you think of this article?

http://www.dirigo.net/tuxTips/avoidingProblems/GrubMdMbr.php
0
 
rindiCommented:
This should help, but you would need to make sure that if a disaster happens the partition on the still functioning HD is active, and if you still can't boot properly, move the HD to the other connection, so that the system looks for the correct devices when booting. You can use a recovery or linux livecd which has fdisk or a similar utility on it to make the partition active.
0
Upgrade your Question Security!

Your question, your audience. Choose who sees your identity—and your question—with question security.

 
Arthur_MinoAuthor Commented:
G'day Rindi,


As long as the two hard-drives are mirroring it should work as a fall back. But the question is would I need to reboot to bring the system up after a hard drive failure? Or will the system continue to work. Because at the moment I'm not getting any robustness out of two hard drives.

I think I might end up purchasing a SATA Raid PCI Card. I can't belive this has been so hard to do with Linux.

Do you know if this is the case with RAID 5 as well?

Thank you for all your help,

Cheers,

Arthur
0
 
rindiCommented:
If a raided disk fails while the system is running, it should continue running until you shut it down. But to reboot it may often need some tweaking like mentioned above.

Raid 5 would have similar problems.
0
 
Arthur_MinoAuthor Commented:
so the freezing i'm experiencing when pulling out a hard disk from a soft raid 1 is abnormal? it should keep running?
0
 
rindiCommented:
This depends on the hardware. It is difficult to simulate a broken disk while the system is running. The controller and HD must at least support hotswapping, and your OS should then also have ACPI support for that hardware. Just removing connectors from the HD isn't the same as a disk breaking or going offline. I recently saw a howto where you can properly simulate a disk crash on a software raid. I'm not sure if this is explained in one of the links I posted above, but if not I don't remember where that was. Maybe you need to look for readme's on mdadm...
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.