Rebuild EVA4400 deleted configuration to original VDISKs access.

[mlapunzina]

Good morning,
I'm new to this community and I do not have a thorough knowledge of the EVA4400, so I apologize in advance if I will write something inappropriate, but I need your experience to understand if what I want to do is possible.

On our EVA4400 there was a DISKGROUP (with n.9 HDD in SINGLE protection level) that contained n.5 VDISKs configured in VRAID1.

Over time, n.3 HDD have failed (I do not know in what time and if any leveling has been successful), but as long as n.6 HDD good remained, everything worked fine (DISKGROUP=Active, VDISKs=Online): as soon as a fourth HDD has failed (n.5 HDD good remained), all VDISKs have become unavailable.

After several attempts to extract and reinsert the filed HDDs, we decided to send all HDDs to a data recovery company (Kroll Ontrack), which was not able to recover the files of our interest, or rather it has recovered some files which however are corrupt.

Having available all the screenshots of the old configuration, I thought to recreate a new DISKGROUP with the same name and the same properties as the original one (we have enough spare HDDs available), then create the same VDISKs with the same names and properties of the original ones (possibly setting the WORLD_WIDE_LUN_NAME's as the original ones, through SSSU tools).

At that point, imagining that the membership of a HDD to a specific DISKGROUP is based on the WWN in some EEPROM of its electronic card, I thought to turn off the SAN, extract the HDDs of the new DISKGROUP, remove their electronic card to replace the card of the original HDDs, then reinsert the original HDD's and turn on the SAN again.

I kindly ask you a feedback on this possible solution to recover the original state of my configuration (although faulty), before it was canceled, restarting from which I hope to be able to repair the last HDD that has failed (since until a moment before everything worked fine ) and resurrect the VDISKs to access their data.

Thanks.

Matteo.

[Torsten.]

I'm not really sure what your goal is ...

Lost data remains lost.

If you want to restore the configuration, consider to initialize the array and restore the configuration backup, what each admin certainly has.

[mlapunzina]

Thanks for the attention, but my goal is just to "try" to recover data, creating a new array with new HDDs and a configuration identical to the one that was deleted when the HDD with my data had already been extracted from the enclosures, and then trying to trick the SAN using the original HDDs after replacing their electronic card with that of the new HDDs.

Matteo.

[mlapunzina]

Talking to some HP consultants, I understood that there is the possibility that an EVA (possibly uninitialized or that no longer finds on board the HDDs with the configuration it knows) can read the configuration (I heard about RSS for the first time) stored inside a set of HDDs that even belonged to a DISKGROUP of another EVA4400: i did not understand if you need a special tool / firmware, but I realized that HP's direct support is necessary.

I also understood that it is an absolutely uncommon intervention, but I would like to know more about this possibility.

Thanks.

[vikashk]

Hi Matteo,

Greetings!

The failed drive information is updated at multiple places by array code to ensure it is not re-used by array.

When you insert new drives to create new DG, the corresponding WWPN would be updated instantly by array in its metadata.

Names are for users convenience. Every object is asssigned a UUID by array firmware. Hence, giving same name might not trick array.

With electronic card swapped (complex task), the EVA would detect the WWPN and discard the drive right away.

I would rather suggest opening a support case with original configuration to engage support on best effort basis to verify the hardware health and recovery is feasible. The support might need content of cache folders (CV directory) and any previous/later logs to review status.

**if not yet deleted, do not delete or make any further changes.

If drives are hard failed (mirrored pair failure) (SMART trips/mech errors etc) of drives, the recovery might not be feasible.

**I am HPE employee.

Regards,

Vikash

[mlapunzina]

Hello Vikash,

I have already opened a case to HP but they tell me that the "start deletion process" (via the CV button on the screen of the diskgroup failed), although executed after extracting the disks from the shelf, has deleted from the controllers (and also from the cache) information no longer recoverable in any way.

I would still like some more information on the community discussion "Replace existing EVA4400 with another EVA4400 and connect up disk shelves", where there was Torsten who in this discussion has already kindly replied to me, where I found the following expression by Sebastian Koehler still leave some hope: <From a technical point of view a "blank" EVA will accept your disks and boot up as long as the quorum disks are intact. Disk position is not relevant. The WWNN of the source system will also "move" to the new box since it is stored in the MLD. WWPNs will be the same as on the "source">

Is there perhaps the possibility to instruct a "clear" SAN (via command line ???) that recognizes at boot a set of HDDs with a previous valid configuration, simply to import it?

Thanks.

Matteo.

[vikashk]

Hi Matteo,

Greetings!

The community discussion referred is related to situations where drives are good and fault is soleley due to multiple drives minor faults.

Unfortunately, the hard faults (initial issue leading to DG meltdown) coupled with and data deletion process is not recoverable. The ways and special codes are used in situations where we could rely on faulted drives to come online; it is not an answer to deleted data or hard faults on drives.

We usually do not see data recovery vendors coming up with all data in scenarios such as this.

**I am HPE employee.

Regards,

Vikash