Big Problem with Raid on ML310 G5

[Odde23]

Hello,

we are running an Proliant ML310 G5 in our company. On this system with Windows 2003 Server is on of the HDDs faild on October 3. I changeed the faild hdd with a newone and start the rebuilding of the logical drive. But when 99% of the Rebuilding with the HP Storage Manager is done, the task stops automaticly with an Error that the running old disks has an Block error. So I can't rebuild the arry. I'm shocked at the moment, because I am not shure what I should do now.

I have nightly Backups from the drive. The drive contains the OS and the Server is the only Windows Domain Controller in our small company. The Images are taken with Acronis True Image Echo Server.

So I considre to change the running hdd with bad block also and configure a new array. Then I will rebuild the Acronis Image on this new Array. And the system should run again or am I wrong? If there will be somthing fail by rebuilding the acronis image, is ist possible to remove the both disk with new the array and insert the disc with defect blocks in the server so that I can boot them interim from this system again until I have a solution for the whole problem?

I hope someone can help me. I am very down because of this problem. Sorry for my bad english, english isn't my first language.

Greets
Stefan

[TTr]

You are not the only one. Bad blocks on the source drives will halt the rebuild of the raided disks.

Your course of action is the correct one. You need to replace all bad drives, recreate the arrays, re-image the OS and restore the data.

The step about re-imaging the OS depends on the method that you use. I can not tell you if the Acronis re-imaging will work because I have never used Acronis. Have you used and tested the re-imaging process before at your site?

[Odde23]

I have used Acronis in some other cases to reimage a system, but not on an HP Proliant ML310 G5.

Can you tell me, where the raid controller saves the raid information? Is it possible to reinsert the hdd with demaged blocks in the system, if the re-imaging will fail, so that we can use the system so long until we have found a solution for a success reimaging?

I ask there for, because I don't have tested the reimaging on this system :-(.

[TTr]

All the raid information is saved on each and every disk.

If you have used re-imaging on other servers it should work on this one. The logical drives of the raid controller appear as regular disks to the server so Acronis should see them.

[Odde23]

So when I understood you correct, it is possible to insert the hdd with the bad blocks in the system again, if the re-imaging would fail.

I aske therfor because the RAID contains also the OS W2003.

[TTr]

Yes the raid information is saved on the disk. If you pull that disk and save it for later, once you insert it on the server you should see the logical disks again.

If you have used Acronis successfully before, I think you will be able to re-image the server. I don't know if Acronis expects identical partitions for before and after re-imaging, if so ensure that the logical drives and partition sizes are identical when you create the disk arrays on the new disks.