EMC is the world leader in storage technologies. In particular, the CLARiiON series is an example of a successful product, still in use at many data centers.
The CLARiiON series is famous for its ability to handle multiple raid disks, combining LUN and MATALUN. However, this makes data recovery more complex than other SAN storage.
Anyone wishing to learn more, we suggest the following LINK .
To us in particular an unfortunate customer has happened with the EMC CLARiiON CX3-20 system, configured with 15 disks of 500 GB each. He had already turned to several data recovery centers but unfortunately no one managed to face the job. Many stopped early, unable even to read the individual EMC discs.
In fact the storage disk is not readable by a normal PC due to the proprietary mode with which the machine writes the disks, with the aim of ensuring greater storage security and eliminating the so-called "silent raid corruption" .
Original storage configuration:
The individual disks were grouped into groups of 5 HDDs forming 3 RAID 5;
The LUN was generated by grouping the individual RAIDs;
The LUNs in turn were combined into a final METALUN of 5TB .
The system was used as data storage and the space offered to users appeared as a single network disk of the same size as METALUN.
Looking at the configuration scheme we see how a first RAID 5 composed of 5 HDDs was independent (LUN0). Two other 5-disk RAID 5s together formed a Component LUN via stripe - RAID 0 of LUN 1 and LUN2.
Essentially the two LUNs were merged into RAID 0 to form a RAID 50 .
The METALUN Finale LUN 0 FC was created by linking Component 0 and Component 1 (JBOD).
In fact the capacity of the disk seen by the users was 5502 GB, obtained with the concatenation of:
Component 0: (LUN0: 1st RAID5) of 1.834GB;
Component 1: RAID 0 (LUN1: 2nd RAID5 - LUN2: 3rd RAID5) of 2 * 1.834GB
Data recovery had obviously to take into account these peculiarities, therefore, after having solved the problems of the single disks (which caused the storage failure), we faced the phase of reconstruction of the raid. We have decided to work only through specialized software and to face reconstruction by hand.
Rebuilding a system that uses almost all known technologies (RAID 5, RAID 0, JBOD) is not easy. We worked a lot with the hex editor to check the raid parameters and the various offsets. The final result can be seen in the image taken by the data recovery software.
Let's look at the list of 15 HDDs, RAID A is the first RAID 5. RAID B and RAID C are two RAID 5 later merged into RAID 50.
JBOD BIG FINAL is the final raid achieved with the concatenation of RAID A and RAID 50, from which we extracted the data.
We are proud of this success, knowing that data recovery from EMC storage is out of reach for many of our competitors.
Comments