RAID 5 "OFFLINE" - please help

Are these all the original disks? Did you try swapping or inserting or reseating any of the disks?

See if you have this option:

On the VD MGMT screen, highlight Controller, hit F2, Foreign, IMPORT.

I think so!? I havent installed any disks myself.

I tried to re-seat the ones who where "missing" but that didnt do much.

Im gonna try to import the foreign config as soon as I get home in about 7 hours.

I bought a sata docking station a couple of days ago. But since the disks on the server are SAS i can't use the docking station to backup anything either.

Hello, good luck and let us know how it goes.

The "Foreign Config" is grayed out. (see attached image)Foreign Control

Hello @Guldstrand,

Sorry for the situation. I wish we could help you more. Please, always keep a backup of the data. In the current server condition, if some of the disk failed, data might be lost. For that reason, you always want to have a backup of the information.

So there is no foreign configuration present. Only one option is let, retagging the disks. When you clear the foreign configuration you are deleting the virtual disk. It only deletes the data that describes the virtual disk, so it may still be possible to get the virtual disk back by retagging. Retagging is the process of creating the same virtual disk as before. It will rewrite the data that describes the virtual disk.

If the data is important then you should contact a data recovery company. Not all controllers can be retagged, some controllers force a full initialization on virtual disk creation. A full initialization is a low level format, similar to writing zeros to a drive.

A retag requires that almost everything be the same. You must use the same disks (not failed ones). Predictive failure disks should still be functional, so you can use them to retag.

If you don't have a foreign configuration listed anymore then these are the steps for retagging:

1. Create a new array using the same stripe, RAID level, and member disks as before. The caching options and size don't matter.
2. Do NOT intialize the array when you create it. An initialization is a low level format that will wipe the data.
3. Boot up the server and check to see if the data is present.

Regards.

Thank you very much for your answer and for your consideration.
As I said, I can not afford to leave the server to any "rescue company".

Could there be a problem with the CMOS battery and / or the battery for the raid card?

And should I try re-tagging the disks even though there are four disks that are "missing / ready"?
Should I delete all current RAID configurations and then restart the server?
What happens if I put ONE disk "wrong"?

A link to a video i made during boot.

Guldstrand,

It wouldn't be an issue with the batteries, as the cmos battery would cause a loss of settings in the BIOS, and the raid battery is for saving cache data on case of failure.

With the retag, the drives would have to be visible to the controller, so if they are showing ready that woul be fine, but missing would be an issue.

If you are incorrect on the placement of the drives in the configuration when you do a retag, it can corrupt the data, as the data isn't in the location it is expected when it boots.

That is why we recommend taking the drives to a 3rd party data recovery company if the data is vital, and you have reservations of the retag.

Ok, i understand. (i think)
The four disks is shown as missing under "VD Mgmt>Physical Disks" but ready under "PD Mgmt", that´s why i asked.

Does this mean I should start re-taging, after i've cleared/removed any other configs?

Before you do that, I ran across this in the knowledge base article regarding the issue, you can find it here.

You may want to look at the steps there first.

I have PERC5 (not 6), but maybe that's ok?
Is there an easy way to check the status if the drives?  I mean IF one or two drives should be failing somehow? (All drives has green  lights and i can't see any error on the display, but still??

Yes... it should work. But, keep in mind PERC 5/i is a really old PERC (almost 20 years old). Remember: if the process fail, the data will be corrupt.

The only way to check the status of the drives is by checking the logs. You would need to boot to OMSA Live, then run a dset and with that it should pull a controller log. 

Check this post where my colleague @

Extacted from the post:

Run the dset in the OMSA live gui and then you can mount a USB and then offload it. Or extract it there and then you will go to dell/log/controller# open that and then look for a section in the log that looks like the example and then cut and paste it in your reply. We should be able to determain the configuration from that.

Example -

ARs configured: 2

T98: Ar 0: size=22dc0000, activeLd=1, numPd=2, seq=0, PDs: 08(0a/8) 09(0a/9)

T98: Ar 1: size=ae980000, activeLd=2, numPd=8, seq=0, PDs: 00(0a/0) 01(0a/1) ff(0a/2) 03(0a/3) 04(0a/4) 05(0a/5) 06(0a/6) 07(0a/7)

T98: LD  0: L=1  SS=128  Size=22dc0000  NL=3763  Status=3  DT=1655  BT=2606, prop=[ID=00,dcp=01,ccp=01,ap=0,dc=0,dbgi=0]

T98:        span 0: sBlk=00000000, nBlk=22dc0000, ar=0

I am sorry @Guldstrand, as you can see here, this is not an easy task.

Regards.

I tried to boot with OMSA but i could not run DSET for some unexplianed reason. 

I booted from an usb drive.

It was an old version of OMSA that supposed to work on PE 2900.

Maybe you could link to another version? 

Can you share some more info? Maybe some pictures. Sometimes the problem is in the media you are using to boot. Can you try again with a different media?

Regards.

I don't have this option "Managed Preserved Cache" in PERC 5 (it seems), from the post you linked to.

And which OMSA version should I use/test for PE2900?