1753682
Members
5484
Online
108799
Solutions
SOLVED
A Proliant server with a smart array controller, 2 disks configured as a RAID 1, can allways boot from 1 disk alone.
Unless you do something wrong.
All smart array controllers are data compatible, between generation.
But they are not driver compatible between generations.
To avoid doing something wrong, you should know aboute the Smart Array controller (SA).
The SA, store the configuation, on all disks in the RIS area of the disks, as metadata.
It all information aboute, RAID level, status and posistion of the physical disks. And it's time stamped.
If you want to remove 1 disk from a pair - Shut down and power off the server.
If you want to put it in another server - Do maintain correct posistion.
If you have a failed or missing disk in a RAID 1. The SA will be in interim recovery mode, unless you select otherwise.
Wich mean, when you put in another good disk in the posistion of the failed/missing disk, it will start rebuild.
It will only rebuild to that physical posistion.
Example:
You've got a server with 2 disks.
Disk 0 and disk 1.
You want to boot 1 of the disks in another server.
shutdow/Powerdown, and remove disk 1.
Put it in bay 1 in the other server.
Leave Bay 0 empty.
Power on, and it will come up in interim recovery mode and boot just fine.
If you put disk 1 in bay 0, it will get confused.
Why didn't you leave the OK disk in the DL380, and replaced the failed disk?
BR
/jag
Unless you do something wrong.
All smart array controllers are data compatible, between generation.
But they are not driver compatible between generations.
To avoid doing something wrong, you should know aboute the Smart Array controller (SA).
The SA, store the configuation, on all disks in the RIS area of the disks, as metadata.
It all information aboute, RAID level, status and posistion of the physical disks. And it's time stamped.
If you want to remove 1 disk from a pair - Shut down and power off the server.
If you want to put it in another server - Do maintain correct posistion.
If you have a failed or missing disk in a RAID 1. The SA will be in interim recovery mode, unless you select otherwise.
Wich mean, when you put in another good disk in the posistion of the failed/missing disk, it will start rebuild.
It will only rebuild to that physical posistion.
Example:
You've got a server with 2 disks.
Disk 0 and disk 1.
You want to boot 1 of the disks in another server.
shutdow/Powerdown, and remove disk 1.
Put it in bay 1 in the other server.
Leave Bay 0 empty.
Power on, and it will come up in interim recovery mode and boot just fine.
If you put disk 1 in bay 0, it will get confused.
Why didn't you leave the OK disk in the DL380, and replaced the failed disk?
BR
/jag
Could be the hardware path to the logical drive that is different.
unix/linux use a hardware path to the disks.
pci bus, device no on the bus. cahannel on the scsi bus, and device ID.
It is likely to be different between the two server models.
So you will have to correct that somewhere in the oprating system.
BR
/jag
unix/linux use a hardware path to the disks.
pci bus, device no on the bus. cahannel on the scsi bus, and device ID.
It is likely to be different between the two server models.
So you will have to correct that somewhere in the oprating system.
BR
/jag
I don't understand, why you would have to re-create logical drive. All information is stored in the RIS area of the disks.
But it could be corrupted.
Good thing you got it to work.
BR
/jag
But it could be corrupted.
Good thing you got it to work.
BR
/jag