Well the reason why I didn't make it raid 0 because if one of the disks were to go bad then my whole raid would go down right? Can you give me the steps on how to make my md1 device a raid 0 without reformatting my whole system? I don't think I can stop md1 unless I stop all the md devices above it right? (md2/3/4/5) Thanks, - Jake On 10/9/07, Richard Creighton wrote:

Jake Conk wrote:

...

Hello,

I tried to setup a fresh install of 10.3 with 2 disks and I used YaST to mirror all my partitions. I had originally my raid swap partitions on an extended partition but that didn't work and I thought that was the problem so instead now I put the raid partitions on primary partitions, which still doesn't work.

Any suggestions would be appreciated because I'm completely stumped, I can't figure out for the life of mine why my swap partition won't work.

I think I would try making your swap partition RAID 0

I have MD0 /boot Raid1 MD1 swap Raid 0 MD2 / (root) Raid 1 MD3 /home Raid 5

and it works well with both 10.2 and 10.3. There is no advantage to using raid1 with swap anyhow, you really don't need it to be backed up or duplicated and it just slows the system down even if it did work.

...

FDISKS___________________________________________________________________

# fdisk -l /dev/sda

Disk /dev/sda: 250.0 GB, 250059350016 bytes 255 heads, 63 sectors/track, 30401 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x000a59ec

Device Boot Start End Blocks Id System /dev/sda1 * 1 13 104391 fd Linux raid autodetect /dev/sda2 14 275 2104515 fd Linux raid autodetect /dev/sda3 276 30401 241987095 f W95 Ext'd (LBA) /dev/sda5 276 928 5245191 fd Linux raid autodetect /dev/sda6 929 948 160618+ fd Linux raid autodetect /dev/sda7 949 30401 236581191 fd Linux raid autodetect

# fdisk -l /dev/sdb

Disk /dev/sdb: 250.0 GB, 250059350016 bytes 255 heads, 63 sectors/track, 30401 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x00026740

Device Boot Start End Blocks Id System /dev/sdb1 * 1 13 104391 fd Linux raid autodetect /dev/sdb2 14 275 2104515 fd Linux raid autodetect /dev/sdb3 276 30401 241987095 f W95 Ext'd (LBA) /dev/sdb5 276 928 5245191 fd Linux raid autodetect /dev/sdb6 929 948 160618+ fd Linux raid autodetect /dev/sdb7 949 30401 236581191 fd Linux raid autodetect

MOUNT___________________________________________________________________

# mount /dev/md2 on / type ext3 (rw,acl,user_xattr) proc on /proc type proc (rw) sysfs on /sys type sysfs (rw) debugfs on /sys/kernel/debug type debugfs (rw) udev on /dev type tmpfs (rw) devpts on /dev/pts type devpts (rw,mode=0620,gid=5) /dev/md0 on /boot type ext3 (rw,acl,user_xattr) /dev/md4 on /srv type ext3 (rw,acl,user_xattr) /dev/md5 on /srv/db type ext3 (rw,acl,user_xattr) /dev/md3 on /tmp type ext3 (rw,acl,user_xattr) securityfs on /sys/kernel/security type securityfs (rw)

MDADM DETAILS___________________________________________________________

# mdadm --detail /dev/md1 /dev/md1: Version : 01.00.03 Creation Time : Mon Oct 8 22:11:04 2007 Raid Level : raid1 Used Dev Size : 2104500 (2.01 GiB 2.16 GB) Raid Devices : 2 Total Devices : 2 Preferred Minor : 1 Persistence : Superblock is persistent

Intent Bitmap : Internal

Update Time : Mon Oct 8 22:14:55 2007 State : active, Not Started Active Devices : 2 Working Devices : 2 Failed Devices : 0 Spare Devices : 0

Name : 1 UUID : 397a7460:be70dee1:fc6c650c:7d900cf7 Events : 5

Number Major Minor RaidDevice State 0 8 2 0 active sync /dev/sda2 1 8 18 1 active sync /dev/sdb2

# cat /sys/block/md1/md/sync_action recover

FSTAB_________________________________________________________________

# cat /etc/fstab /dev/md2 / ext3 acl,user_xattr 1 1 /dev/md0 /boot ext3 acl,user_xattr 1 2 /dev/md4 /srv ext3 acl,user_xattr 1 2 /dev/md5 /srv/db ext3 acl,user_xattr 1 2 /dev/md3 /tmp ext3 acl,user_xattr 1 2 /dev/md1 swap swap defaults 0 0 proc /proc proc defaults 0 0 sysfs /sys sysfs noauto 0 0 debugfs /sys/kernel/debug debugfs noauto 0 0 usbfs /proc/bus/usb usbfs noauto 0 0 devpts /dev/pts devpts mode=0620,gid=5 0 0 /dev/fd0 /media/floppy auto noauto,user,sync 0 0

FREE__________________________________________________________________

# free -m total used free shared buffers cached Mem: 6089 57 6031 0 3 25 -/+ buffers/cache: 29 6060 Swap: 0 0 0

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Moby, THANKS!!!!!! That fixed my problem! Yes I don't know why people insist swap should be on raid 0 other that performance reasons but if your disk goes belly up then yeah your totally screwed. Anyways I'm glad I didn't have to go raid 0 in order to fix this problem and now it works my 10.3, very sweet thank you very much :) Regards, - Jake On 10/9/07, Moby wrote:

...

Well the reason why I didn't make it raid 0 because if one of the disks were to go bad then my whole raid would go down right? Can you give me the steps on how to make my md1 device a raid 0 without reformatting my whole system? I don't think I can stop md1 unless I stop all the md devices above it right? (md2/3/4/5)

Thanks, - Jake

On 10/9/07, Richard Creighton wrote:

...

Jake Conk wrote:

...

Hello,

I tried to setup a fresh install of 10.3 with 2 disks and I used YaST to mirror all my partitions. I had originally my raid swap partitions on an extended partition but that didn't work and I thought that was the problem so instead now I put the raid partitions on primary partitions, which still doesn't work.

Any suggestions would be appreciated because I'm completely stumped, I can't figure out for the life of mine why my swap partition won't work.

I think I would try making your swap partition RAID 0

I have MD0 /boot Raid1 MD1 swap Raid 0 MD2 / (root) Raid 1 MD3 /home Raid 5

and it works well with both 10.2 and 10.3. There is no advantage to using raid1 with swap anyhow, you really don't need it to be backed up or duplicated and it just slows the system down even if it did work.

...

FDISKS___________________________________________________________________

# fdisk -l /dev/sda

Disk /dev/sda: 250.0 GB, 250059350016 bytes 255 heads, 63 sectors/track, 30401 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x000a59ec

Device Boot Start End Blocks Id System /dev/sda1 * 1 13 104391 fd Linux raid autodetect /dev/sda2 14 275 2104515 fd Linux raid autodetect /dev/sda3 276 30401 241987095 f W95 Ext'd (LBA) /dev/sda5 276 928 5245191 fd Linux raid autodetect /dev/sda6 929 948 160618+ fd Linux raid autodetect /dev/sda7 949 30401 236581191 fd Linux raid autodetect

# fdisk -l /dev/sdb

Disk /dev/sdb: 250.0 GB, 250059350016 bytes 255 heads, 63 sectors/track, 30401 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x00026740

Device Boot Start End Blocks Id System /dev/sdb1 * 1 13 104391 fd Linux raid autodetect /dev/sdb2 14 275 2104515 fd Linux raid autodetect /dev/sdb3 276 30401 241987095 f W95 Ext'd (LBA) /dev/sdb5 276 928 5245191 fd Linux raid autodetect /dev/sdb6 929 948 160618+ fd Linux raid autodetect /dev/sdb7 949 30401 236581191 fd Linux raid autodetect

MOUNT___________________________________________________________________

# mount /dev/md2 on / type ext3 (rw,acl,user_xattr) proc on /proc type proc (rw) sysfs on /sys type sysfs (rw) debugfs on /sys/kernel/debug type debugfs (rw) udev on /dev type tmpfs (rw) devpts on /dev/pts type devpts (rw,mode=0620,gid=5) /dev/md0 on /boot type ext3 (rw,acl,user_xattr) /dev/md4 on /srv type ext3 (rw,acl,user_xattr) /dev/md5 on /srv/db type ext3 (rw,acl,user_xattr) /dev/md3 on /tmp type ext3 (rw,acl,user_xattr) securityfs on /sys/kernel/security type securityfs (rw)

MDADM DETAILS___________________________________________________________

# mdadm --detail /dev/md1 /dev/md1: Version : 01.00.03 Creation Time : Mon Oct 8 22:11:04 2007 Raid Level : raid1 Used Dev Size : 2104500 (2.01 GiB 2.16 GB) Raid Devices : 2 Total Devices : 2 Preferred Minor : 1 Persistence : Superblock is persistent

Intent Bitmap : Internal

Update Time : Mon Oct 8 22:14:55 2007 State : active, Not Started Active Devices : 2 Working Devices : 2 Failed Devices : 0 Spare Devices : 0

Name : 1 UUID : 397a7460:be70dee1:fc6c650c:7d900cf7 Events : 5

Number Major Minor RaidDevice State 0 8 2 0 active sync /dev/sda2 1 8 18 1 active sync /dev/sdb2

# cat /sys/block/md1/md/sync_action recover

FSTAB_________________________________________________________________

# cat /etc/fstab /dev/md2 / ext3 acl,user_xattr 1 1 /dev/md0 /boot ext3 acl,user_xattr 1 2 /dev/md4 /srv ext3 acl,user_xattr 1 2 /dev/md5 /srv/db ext3 acl,user_xattr 1 2 /dev/md3 /tmp ext3 acl,user_xattr 1 2 /dev/md1 swap swap defaults 0 0 proc /proc proc defaults 0 0 sysfs /sys sysfs noauto 0 0 debugfs /sys/kernel/debug debugfs noauto 0 0 usbfs /proc/bus/usb usbfs noauto 0 0 devpts /dev/pts devpts mode=0620,gid=5 0 0 /dev/fd0 /media/floppy auto noauto,user,sync 0 0

FREE__________________________________________________________________

# free -m total used free shared buffers cached Mem: 6089 57 6031 0 3 25 -/+ buffers/cache: 29 6060 Swap: 0 0 0

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

I still do not understand why people keep on insisting that swap should not be on raid - try keeping a system running when the disk holding swap goes South. As for the problem, try putting a noresume line in your menu.lst for grub - I have a feeling you have a resume= pointing to your raid1 set

Jake Conk wrote: that you are having issues with. I just got done fighting a similar problem here, and changing resume= to noresume fixed it.

-- --Moby

They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety. -- Benjamin Franklin

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Richard Creighton wrote:

Jake Conk wrote:

...

Moby,

THANKS!!!!!! That fixed my problem! Yes I don't know why people insist swap should be on raid 0 other that performance reasons but if your disk goes belly up then yeah your totally screwed. Anyways I'm glad I didn't have to go raid 0 in order to fix this problem and now it works my 10.3, very sweet thank you very much :)

Regards, - Jake

I can't say I insist that RAID 0 for swap is safer in case of malfunction, only that performance is improved and of course you get twice the space over raid 1 and with Linux's ability to mkswap in other partitions in an emergency, either when you run short or if the array poops, you aren't really as exposed as you might think, IMO. But, either way, I'm glad you got it going and I'll remember the 'resume' trick for the future...Thanks Moby... The way many of us divide up our physical drives to make up multiple raid arrays, if my swap partition were to fail, it is likely that I have also lost a drive, which has compromised my other raid arrays, so I have a lot of work to do, and worrying about loss of swap space is the least of my problems :) FWIW, I have 3 G of RAM and I rarely use any swap on disk, so my exposure is somewhat minimal I would expect. That and the general reliability of disk drives now days and I figure I have at least a few weeks of relatively uneventful computing due to hardware failure ahead....now, 10.3 GM is another story :)

Richard

...

On 10/9/07, Moby wrote:

...

Jake Conk wrote:

...

Well the reason why I didn't make it raid 0 because if one of the disks were to go bad then my whole raid would go down right? Can you give me the steps on how to make my md1 device a raid 0 without reformatting my whole system? I don't think I can stop md1 unless I stop all the md devices above it right? (md2/3/4/5)

Thanks, - Jake

On 10/9/07, Richard Creighton wrote:

...

Jake Conk wrote:

...

Hello,

I tried to setup a fresh install of 10.3 with 2 disks and I used YaST to mirror all my partitions. I had originally my raid swap partitions on an extended partition but that didn't work and I thought that was the problem so instead now I put the raid partitions on primary partitions, which still doesn't work.

Any suggestions would be appreciated because I'm completely stumped, I can't figure out for the life of mine why my swap partition won't work.

I think I would try making your swap partition RAID 0

I have MD0 /boot Raid1 MD1 swap Raid 0 MD2 / (root) Raid 1 MD3 /home Raid 5

and it works well with both 10.2 and 10.3. There is no advantage to using raid1 with swap anyhow, you really don't need it to be backed up or duplicated and it just slows the system down even if it did work.

Most welcome Jake and Richard. Richard, my apologies if I sounded a bit harsh - just my frustration showing on having fought the very same issue for too many hours today before finding the fix. As for swap on RAID1 - well, raid'ing any disk has pros and cons. The machine I was fighting the very same issue on today is almost guaranteed to swap once it goes live, and I need to have it able to run until I can get around to fixing it should a disk go bad - hence the decision to put swap (and everything else) on RAID. -- --Moby They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety. -- Benjamin Franklin -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On Oct 9 2007 21:04, Richard Creighton wrote:

I think I would try making your swap partition RAID 0

Swap on RAID0 is stupid, because swap is intelligent enough itself to stripe to multiple locations. All it takes is the pri= argument in fstab. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Wednesday 2007-10-10 at 06:12 -0400, Richard Creighton wrote:

Jan Engelhardt wrote:

...

Swap on RAID0 is stupid, because swap is intelligent enough itself to stripe to multiple locations. All it takes is the pri= argument in fstab.

That is one opinion. I don't happen to agree that it is stupid, it is only another method. Isn't it a wonderful thing to be able to exercise your rights? Even if the right you are exercising is the right to be wrong.

Swap on raid 0 is absurd and not recommended (although it works), you simply get better results using what Jan recommends. However, swap on raid 1 does have its merits (redundancy). - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (GNU/Linux) Comment: Made with pgp4pine 1.76 iD8DBQFHDK0RtTMYHG2NR9URAumHAJ4tI/oWe2T46J48ErJeo1tpIfvCNwCZAY9v 5EuKKfOYBiaJsMqcmusxu38= =dx2b -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Mohsen Rezayatmand wrote:

I agree, it is not a good idea to Raid swap space, it is also faster if you do not Raid the swap.

Given the main reason for RAID is fault tolerance, what happens if the drive holding SWAP craps out and it's not RAID? Would that not tend to cause problems for a running system to lose everything in SWAP? On a server I have at home, everything is on RAID 5, except /boot, which is RAID 1. -- Use OpenOffice.org http://www.openoffice.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On Wed, 2007-10-10 at 16:26 +0200, Carlos E. R. wrote:

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1

The Wednesday 2007-10-10 at 09:49 -0400, James Knott wrote:

...

Mohsen Rezayatmand wrote:

...

I agree, it is not a good idea to Raid swap space, it is also faster if you do not Raid the swap.

Given the main reason for RAID is fault tolerance, what happens if the drive holding SWAP craps out and it's not RAID? Would that not tend to cause problems for a running system to lose everything in SWAP? On a server I have at home, everything is on RAID 5, except /boot, which is RAID 1.

Yes, but it has been proposed to mount swap on a raid 0, and that doesn't have any fault tolerance, rather the contrary, and it is slower than two swap stripes. That is my point, that swap on raid 0 is not recomended.

Why not use filesystem swap (using a file on the mounted partition) as has been suggested a few times in the past? And it was also pointed out that filesystem swap is fast like a raw swap partition. This would effectively put swap on the raid set for the filesystem. -- Ken Schneider UNIX since 1989, linux since 1994, SuSE since 1998 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org