Booting a ghosted SCO HDD

[aaronactive]

I have a SCO 5.0.5 server running on a Compaq Proliant 5500 with 5 logical drives setup on a Smart Array 3200 controller.

Tried to virtualise this machine on our ESX server awhile ago, and took images of each individual drive and then restored them to VM hard drives. Haven't been able to get that to boot so hasn't worked (it's been sitting on our ESX server now for 2 years)

Anyway in the physical server, the first HDD (0) had a couple of bad sectors pop up and SCO failed to boot.

I've taken the image from the VM server, restored it on another HDD, and then put that into the server in HDD(0)'s place and reconfigured the array so that its all OK from that side.

I believe when I first imaged the drive to transfer to the VM, I didn't do the drive, instead did the partition so missed the boot up info.

If I just try booting normally, with no boot disk - it just comes up with
----
not a directory

boot not found
Cannot open
Stage 1 boot failure: error loading hd(40)/boot
----

I have boot/root disks for this server, however I can't get past boot process (it dies JUST before INIT SINGLE USER MODE) saying PANIC: exit - cannot exec /etc/init (PID 1) 0x00000200

Using the bootdisk and then at Boot: typing hd(40)unix = That doesn't work - I'm assuming I need to LINK the BTLD drivers for the 3200 Smart array.

I have the BTLD for that and when I tried to link it =
defbootstr disable=ida link="ida cha" hd=ida0 Srom=wd(0,0,0)

It will ask for the IDA disk..that works fine, the when CHA comes up:
bootlocore : out of low (below 1Mb) memory!

Getting frustrating now and I have no idea what else I need to do.

If I try putting in the old HDD into slot 0, the array cracks it and says there is no array configuration. So I don't get anything for that. It wanted to do this so I could get the error..but no go.

Can anyone be of any help?

Thanks

 

[stanhubble]

moving openserver images (or even hard drives) around can be a real pain if you change the hd controller (ie raid -> vm WD)

the easiest is to do a new installation of 5.0.5 in the virtual machine and then restore a backup of your data/applications. do not restore /etc or /stand.

5.0.5 works just fine in a virtual machine.

 

[aaronactive]

Thanks for the tips, at the moment i'm trying to get the original server running...

problem is I don't have the original 5.0.5 media which doesn't help.

 

[aaronactive]

OK Have tried a number of things and failed miserably.

Although I have the 5.0.5 media now.

I have one idea now, i'd just like thoughts on whether it would actually work or not.

Raid array is setup as per OP.

If I do a sco reinstall onto the 1st disk, what are the chances of configuring it to read the rest of the disks in the array - that were setup on the old install?

I have another sco box built up now, went to restore a tape to only find that the backup is incomplete.

I want the data off the rest of the disks in the system.

 

[Annihilannic]

How many physical disks were in the original RAID array? How were the logical drives configured, as RAID1? RAID5? Across all of the disks?

Annihilannic.

 

[aaronactive]

Thanks for the replies, they was no raid setup..each physical drive was a logical drive

but its all OK now, we ended up doing a sco install onto a new box and then (luckily) found a complete backup tape and restored everything.

phew..