Link to home
Create AccountLog in
Avatar of Arnie Pavlik
Arnie PavlikFlag for United States of America

asked on

VMware

I had a power outage for longer than the UPS could help and there was a "bad" shutdown of our esxi host machine. I can't get it to reboot. My question is, can i reinstall esxi 5.5.0 without overwriting the existing VMs?
Avatar of Alex
Alex
Flag of United Kingdom of Great Britain and Northern Ireland image

Yes you can, you'd have to re-register them into ESXi.

I assume this is standalone?

https://kb.vmware.com/s/article/2147569
https://kb.vmware.com/s/article/1006160 is the official one but the one for 6.5 is also valid. They didn't change anything if you do it via the web client.
Avatar of Arnie Pavlik

ASKER

Yes, it is a standalone server. I'm at the "Welcome to the VMWare ESXi 5.5.0 Installation" screen. Just a tad nervous that I'll overwrite the VMs if I press (Enter) Continue.
It won't, when you install ESXi I assume you put it onto a separate drive right? Something like the internal SD card or a different partition.

It won't trash your other array as long as you don't manually trash it, it'll just overwrite the existing installation.

Regards,

Alex
I inherited this server, so I'm not sure how/where it was installed. Is it safe to assume that the installation will 'see' the previous version?
Normally yes, it'll say there is already a version installed and oyu'll get an option to either do a fresh installation or upgrade.

If however you have hardware issues due to the power down, I.E something broke and it's not coming back up, you could have lost your VM's already.

Have you checked the RAID array on it to ensure it's all still in one piece?

Alex
I'm looking into the RAID, etc., now.
No drives detected. What now?
Yeah that's bad, on a scale of 1 to 10, you're looking at a 15.

You've either lost the RAID card or all your drives failed which I think is unlikely. Do you have something like an MSA attached or is it using a SAN or something?

Also, do you have the drives populated in the server itself? I.E 2.5" SAS drives or something? If that's the case you could have blown the backplate of the disk ports, I've seen that before and it's a call to HP for an engineer to come out to replace it.

This is why it can't boot.
All the important data is backed up safely to an external HDD, but this is a DC. So, all AD items are inaccessible. I do have a Windows Image Backup as well as a System State Backup that's not too old.
Until you fix the hardaware issue, I.E why you can't see any of your drives we can't do anything else.

Your VM's are still going to be on the drives, you shouldn't need to reinstall ESXi purely from the fact that it's likely the backplate has broken.

Also, I need my other questions answered before we can continue.
What questions do you have?
Do you have something like an MSA attached or is it using a SAN or something?

Also, do you have the drives populated in the server itself? I.E 2.5" SAS drives?
Data is backed up to a SAN. There are 6 SATA hdds 2 that are 'populated'.
Right, so the data is backed up to the SAN which means your VM's are stored on the SAN?

Thanks
Alex
Also,

There are 6 SATA hdds 2 that are 'populated'.

You mean there are 6 available slots and 2 have hard drives in?

Is this server directly connected to the SAN?

Lastly, when you go into the installed RAID controller, can you see ANY drives? Also, verify that you're going into the correct RAID controller. Some servers have 2.
that 2 shouldn't have been there. There are 6 hdds. all are being used. This server is connected to the SAN via ethernet. RAID controller doesn't see any drives at all.
"HP Dynamce Smart Array B120i RAOD Controller 0 Logical Volume
1785 - Drive Array Not Configured
No Drives Detected"
I'm getting POST Errors 1785 & 1783
my question is: how could you know there was no corrupt data (in VM data area) caused by the bad shutdown?
Right,

Those two alerts are "Informational" only apparently, it's basically saying that the drives aren't configured. I suspect there is an issue with the RAID card or the back plate.

Since you can't confirm the previous configuration and you can't tell me what RAID set was on there previously or even if the VM's are residing on the local array, I can only come the conclusion that they were in fact in use, the array has failed and you're currently in this situation.

As such, since you previously had a RAID0 array as you described, I can only make the assumption that the backplate within the server has blown. You need an engineer to come out and take a look.

Your alternative is this, if you have another server, with the same RAID card, I.E A DL360 Gen 8 and another exact same server. Take the drives out, put them into the same slot you removed them from and see if it can see the array.

Regards
Alex
So you have a HPE server with 6 disks installed in the server, that are configured as some sort or RAID, and NOW the RAID has GONE!
Has there been any update on this one at all?
I am working on the assumption that the backplane failed. It's not in the budget to replace this, so I've purchased a new piece of hardware to replace this old server.
ASKER CERTIFIED SOLUTION
Avatar of Arnie Pavlik
Arnie Pavlik
Flag of United States of America image

Link to home
membership
Create an account to see this answer
Signing up is free. No credit card required.
Create Account