HP servers

[The_Librarian]

I have heard of somebody who've had an HP ProLiant ML370 server. The motherboard suffered a catastrophical failure.

Said person simply got another ML370 server, and transferred the hard drives in the same order over to the new server, which booted and continued to operate as if nothing happened.

Seems like the RAID arrays which HP used have a few tricks up their sleeves.

I used it to resuscitate a crashed HP Proliant DL380e Gen8 server whose iLO was shot.

There was no way to access the RAID controller in order to set up two new hard drives for RAID1 where the previous two was set up for RAID0.

Luckily we had another Proliant DL380 server (in working condition). I transferred the whole RAID over (8 drives), and started it up so that the RAID can get the config from the hard drives, which it duly did.

Then it was shutted down, and I replaced the first two hard drives with new ones, started it up, went to the RAID controller's configuration, added the two new drives as RAID1, created a new virtual boot drive, and from there it was easy.

The newly-created RAID and the drives was physically transferred back to the original server (the order of the drives was kept), the original server was started up, and I managed to install Server 2008 R2 on the primary (boot) partition without any issues.

So this is something to keep in mind, especially when dealing with older, recalcitrant servers.

Maybe this post will slip into obscurity... maybe not... but if it helps somebody, then it have done its work.

 

[hilt_ctn]

I have heard of somebody who've had an HP ProLiant ML370 server. The motherboard suffered a catastrophical failure.

Said person simply got another ML370 server, and transferred the hard drives in the same order over to the new server, which booted and continued to operate as if nothing happened.

Seems like the RAID arrays which HP used have a few tricks up their sleeves.

I used it to resuscitate a crashed HP Proliant DL380e Gen8 server whose iLO was shot.

There was no way to access the RAID controller in order to set up two new hard drives for RAID1 where the previous two was set up for RAID0.

Luckily we had another Proliant DL380 server (in working condition). I transferred the whole RAID over (8 drives), and started it up so that the RAID can get the config from the hard drives, which it duly did.

Then it was shutted down, and I replaced the first two hard drives with new ones, started it up, went to the RAID controller's configuration, added the two new drives as RAID1, created a new virtual boot drive, and from there it was easy.

The newly-created RAID and the drives was physically transferred back to the original server (the order of the drives was kept), the original server was started up, and I managed to install Server 2008 R2 on the primary (boot) partition without any issues.

So this is something to keep in mind, especially when dealing with older, recalcitrant servers.

Maybe this post will slip into obscurity... maybe not... but if it helps somebody, then it have done its work.

This is due to the array configuration metadata being written to both the disks and HP / HPE Smart Array controllers. So the configuration is maintained in 2 places. This enables the RAID disks stacks to be moved from server to server without data being lost, or Smart Array controllers to be replaced without data being lost. The configuration is simply imported