J
J
Jacob2011-08-13 20:32:14
FreeBSD
Jacob, 2011-08-13 20:32:14

Determining the RAID Build Scheme

So, there is a dead car with supposedly the sixth Freya on board. Of the guaranteed live - only memory, a raid controller and screws. The comrades from the data center who assembled the array did not name the exact scheme of the joint, but threw an approximate scheme on a napkin: the first screw on the second, the third on the fourth, the first and third are mirrors.

Judging by the description, it must be one of the RAID-1 configurations. But what exactly? Collective intelligence, please - tell me. I will also be glad to have useful links to mana and how to restore such things under Bzdey, because I have absolutely no experience in it.

Answer the question

In order to leave comments, you need to log in

4 answer(s)
S
SkyRZN, 2011-08-13
@SkyRZN

Regarding the recovery of RAID from under Windows (boots from the same LiveCD image), you can use R-Studio , it will also be able to try to determine the RAID.

E
Evgeny Yablokov, 2011-08-13
@Gular

Quite abstractly described the scheme to you. :) Maybe it's a degraded raid10?

J
Jacob, 2011-08-13
@ksenobayt

Probably. If I ever correctly understood the marking of circuits, then this is 1 + 0 - although then, logically, these should be the original, a mirror, and backup blocks for them. But if I remember correctly, there was once a dead screw in this array, and while they were looking for a replacement, the server continued to work, which means that it was hardly a zero.
Perhaps it was just two units?

0
009b, 2011-08-15
@009b

if there is a raid controller and it is really hardware, then it is enough for your recovery. He will write which screws are online, which failed. and in the same place in the configuration of virtual disks you will understand the scheme of their connection.
if the server was not dismantled and everything was plugged in exactly as it worked, then you can use information both from the controller’s memory and from disks. if everything was disassembled, then either you are lucky, or you will restore the markup from the disks.
my experience allowed me to recover from disks only when they were plugged in correctly (in working order), in this case the Intel controller took a copy of the config from the screws themselves and restored, replacing the blown out screws with the correct ones.
Let's clarify that such a recovery will not depend on the processor, memory (in case they are dead). If a dead mother, then I do not quite understand how the fact that the screw controller is alive was determined.
the whole process will take place before loading the operating system from the raid's own console after the initialization of the bios and before the bootloader of the operating system.
last of all, I would try to remove the disks marked by the hardware controller from the server and try to restore them with software tools by sticking the screws directly into the computer. Look for the same controller, for example, if this one is bent. there is experience of such actions, the experience is unsuccessful.

Didn't find what you were looking for?

Ask your question

Ask a Question

731 491 924 answers to any question