LinuxQuestions.org
Help answer threads with 0 replies.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Distributions > Red Hat
User Name
Password
Red Hat This forum is for the discussion of Red Hat Linux.

Notices

Reply
 
Search this Thread
Old 07-10-2004, 08:25 PM   #1
Ashron
LQ Newbie
 
Registered: Jul 2004
Posts: 1

Rep: Reputation: 0
Moving Raid0 on Redhat Linux


Hi,

I have been using linux software raid for raid0 for a long time.
I decided to add more space with an additional controller, but it broke my raid setup.

Some back ground:

System used to be setup as follows:

hdb: Maxtor 90840D5, ATA DISK DRIVE (with OS)
hdc: SAMSUNG CD-R/RW SW-240B, ATAPI CD/DVD-ROM drive.
hde: WDC WD400BB-00AUA1, ATA DISK drive
hdg: WDC WD400BB-00AUA1, ATA DISK drive

/dev/md0 was made of /dev/hdg1 and /dev/hde1 in raid0.


When I added an ATA133 controller, and two additional drives, the kernel started to detect them as follows:

hdb: Maxtor 90840D5, ATA DISK DRIVE (with OS)
hdc: SAMSUNG CD-R/RW SW-240B, ATAPI CD/DVD-ROM drive
hde: Maxtor 6Y250P0, ATA DISK drive
hdg: Maxtor 6Y250P0, ATA DISK drive
hdi: WDC WD400BB-00AUA1, ATA DISK drive
hdk: WDC WD400BB-00AUA1, ATA DISK drive

When the machine boots, it wants to look at /dev/hdg1 to setup the array, and it breaks when it sees that hde is zero size. (the new maxtor drive have not yet been formatted)

it works it way down to hdk, but then it realizes that:

md: device name has changed from hdg1 to hdk1 since last import!
md0: former device hde1 is unavailable, removing from array!

and then it does not go forward.

I have tried to edit /etc/raidtab so it reflects the new changes to the devices. i.e

raiddev /dev/md0
raid-level 0
persiatent-superblock 1
chunk-size 4
nr-raid-disks 2
device /dev/hdk1
raid-disk 0
device /dev/hdi1
raid-disk 1


where they used to be /dev/hdg1, and /dev/hde1, for raid-disk 0 and 1 respectively.

It does not seem to have any effect.

The research I have done so far, suggests that the problem is due to the presistant superblock on the array. When the kernel tries to mount /dev/md0 it is look at the superblock on one of the disks, and binds the other drives. The superblock has old info.

I would like to know is, what is the procedure of rebuilding a raid0 in linux, when the disks consisting of the array have been moved to another system, and have a different letters due to their position on the controller.

Or, if there is a way to make the kernel in linux detect certain IDE devices before others, so that drives are recognized in the older order making hdg -> hdk and hde -> hdi, and vice versa.

I hope someone can help. Thanks in advance.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Linux cannot detect preexisting raid0 r_sherman Linux - Hardware 3 10-25-2005 05:06 PM
Moving files to new Redhat box urbman Linux - Software 1 03-30-2004 03:39 PM
boot to PATA raid0 on Promise 378 *and see* SATA raid0 ICH5 ? patr Linux - Hardware 2 02-25-2004 02:36 AM
Install REDHAT 9 on RAID0 (HPT370) anc Linux - Hardware 4 05-21-2003 06:33 PM
Moving from RedHat 8.0 to 7.0 newbie42 Linux - Distributions 3 12-06-2002 03:59 AM


All times are GMT -5. The time now is 02:17 AM.

Main Menu
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
identi.ca: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration