Setting physical drive status to offline with CLI.

[Jari Huuskonen]

Is there a utility from HP to set physical drive status to offline?
The main purpose would be to be able to break a raid1 mirror without unplugging the drive and also to be able to re-establish the mirror with a command.

Any help is appreciated!

[David Claypool]

ACU. It is part of the ProLiant Support Pack: http://www.hp.com/servers/swdrivers

[Jari Huuskonen]

Yes, i've used ACU, but my plan is next.
Suppose that we have OS installed on RAID 1, and i want to do next
onto command line, mark disk 0 to offline, patch servers and restart them. And wy to do this: "rapid OS "backups" before patching".
If someting fails turn to disk 0 online.

Can you do this with ACU?

[Jari Huuskonen]

Anyone, how can i mark in running server disk to offline with ACU?

[amhakassa]

HI Jari,

As far as I know there is no way to bring an online hard drive to offilne using ACU. However, if you are planning to do what you plan to do and keep on of the disk in the RAID 1 (Mirror) just for the sake of safty, I would recommend this.

1. Power Down Server
2. Take one of the hard drive out while server is powered down
3. Power server up and you will get a POST error message that one of the drive in the RAID 1 set is missing
4. Press F1 to continue (by pressing F1 you are acknowledging to the POST process that one drive is missing). Remember at this point you logical drive is degraded because you have a missing member in the raid
5. Do whatever patchs you wanted to do (Remember, If anything happened in the process you still have the other drive to boot from with the original data and OS so you can always start from that drive)

However, after the patch and any updates you ahve done with one of the drives only will leave the other drive that you pulled off unpatched. And if your patching and update is successfull

I would recommend to do this to make sure the patched drive is copied in to the one that was out while in the process
1. Power down server
2. Put the drive that was pulled of in the server
3. Power server on and if you got a post error message make sure you select the rebuild or reconstract option so that the drive data on the drive that was already patched will be copied to the drive that was out during the patch

But, I don't really why you chose this path to do this patch though.

Regards
Amha Kassa

[Jari Huuskonen]

Reason why i asked this is that we have over 200 proliant servers, and i'm tired to pull off the one disk before patching servers, and i was hoping that there is some mechanism to do this with software, yes we have decent backup mechanism, even its EVA to backup robots, thats not faster enough.
But your answer clearens what i too believe imbossible.

Thanks all for your time.

[Jari Huuskonen]

Thansk folks.