Koozali.org: home of the SME Server
Contribs.org Forums => General Discussion => Topic started by: jabell on April 09, 2007, 02:59:09 PM
-
Hi All!
I have been reading a number of posts here regarding Raid1 and rebuilding after failure, but need clarification of what needs to be done before proceeding so I don't blow away a pile of data that the users depend on.
I noticed a few references to a Menu item in the Server Manager for "Manage disk redundency ", but I can't see it there, and wonder why?
I have had disk 1, a 160gb SATA drive, fail, and so the server is now running on /dev/sda2.
Am I correct as one post stated, that all I do is shut down the server, replace disk1 (SDA1) , power the server back up and it will automatically rebuild? I'd hate to end up with 2 blank disks instead of the desired rebuilt mirrored pair.
Is it the situation that regardless which one fails, that replacing the failed drive in the same location will result in it rebuilding from the other one?
I have tested using MDADM and it shows disk 1 as failed.
Thanks to all!
cheers
john
-
the menu is in the console not the server-manager
mdadm knows which disk is good,
but you do have backups ...
just for fun try the following, see if smart thinks the drive is bad
edit
smartctl -t short /dev/sda or/and
smartctl -t short /dev/sdc
-
I have had disk 1, a 160gb SATA drive, fail, and so the server is now running on /dev/sda2.
Am I correct as one post stated, that all I do is shut down the server, replace disk1 (SDA1) , power the server back up and it will automatically rebuild?
sda1 is a partition, not a disk. sda1 and sda2 are two partitions on the same drive.