This is a discussion on SW Raid-1 - RH-9 questions within the Linux Administration forums, part of the Linux Forums category; I'm building up a webserver and am attempting to use software raid-1 with redhat 9.0 . I just ...
|
|||||||
| FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
|
|||
|
I'm building up a webserver and am attempting to use software raid-1
with redhat 9.0 . I just want to see if I'm on the right track here. I have set up a test box - 2 drives of different sizes (I will get two drives of the same size in final server) I get to the disk druid part of the installation and this is what I did: Created a raid partition on first drive then same size on other drive (\boot). Then I click the raid button and it lets me link the two together. I did this for boot, swap and the root \ partitions. Is this correct? At first I was under the impression you could not mirror the boot or swap partitions but then I figured out that that's exactly what you want to mirror as it defeats the whole purpose if the root partition is not mirrored and a drive goes bad? What happens if a part of the drive goes bad that contains the part of the os that runs the software raid - the code must be in memory right so if no one turns the power off it can just switch to the other disk or start to rebuild the damaged part. If the boot partition is bad how and the os didn't get time to rebuild it after a power loss what then - I presume a boot floppy is the only solution? I've heard that I can add a spare disk so if anything goes wrong this spare will come into play automatically. How will i know if a drive goes bad - is there a monitor script that can email me? TIA Richard (For Vervet) |
![]() |
| Thread Tools | |
| Display Modes | |
|
|