Data recovery in cases of service area failure:
All modern hard drives have service area (SA). It is an area of the drive which has microprogram (firmware) written on it.
Microprogram is written upon special negative cylinders (from -1 to -32), which have significantly lower media density, and which is started during disk boot up or initialization. This zone is not accessible through operating system and standard tools. It is marked as UBA (Upper Block Addressing) zone. Lower media density on that portion of hard drive actually makes data reading easier and is a prevention measure for one of the most common microcode failures – damaged modules (parts of the microcode).
Microcode itself has a lot of separate pieces (modules, control pages, system files, etc). Each module has its own function (S.M.A.R.T, defect lists, ATA passwords, SmartWare passwords, translation module, disk ID,…), and some of them are critical to HDD functioning.
Every module has several copies, but it is sometimes impossible to access them (depending on disk manufacturer).
In most cases, microprogram is segmented into two parts. One is written on PCB, and the other is written upon UBA zone. This approach brought unification of the code throughout a large number of families of a drive manufacturer. In other words, a large portion of the program is identical, and only unique parts (adaptive parameters) are put on PCB. That is exactly the reason why data recovery business becomes additionally complicated, or downright impossible, if original PCB is lost or destroyed during “professional” intervention.
Adaptive parameters are created during hard drive manufacturing. In them are unique parameters for that specific hard drive (number of and head map, servo markers, heads distance…)
Service area malfunction are most commonly manifested as wrong disk ID in system BIOS. The most usual way is for the drives to give factory aliases as ID.
- Here are some examples:
- Maxtor: N40P, ATHENA, CALYPSO, ROMULUS, SABRE
- Seagate: Disk registers with system BIOS, but blocks the system or reports disk size as 0GB.
- Western Digital: WD-ROM-0GB.
Among the many malfunctions and manifestations, it is not possible to find unified way to detect service area malfunction. Our technicians are primarily relying on experience and equipment, as well as our own experiments with service area modifications, so they can recover data for the client regardless of SA damage.
Most of the time, service malfunctions come from poor media quality. During normal operation, hdd is generating new bad blocks, which are then marked and their location is written in a growing list of defects. Once marked as BAD, that sector will not be used anymore, and the defect list continues to grow. Constant writing into this list brings about several different types of damage to the list itself, and so to one of the most common malfunctions – translator damage. Translator is a name used for a group of several modules whose primary function is translation of logical to physical address, so the drive can allocate every sector.
The other common malfunction is SMART module damage. Even though their function is additional information, which should not affect the way a drive works, these modules evolved during the years. Modern drives have sophisticated tests which can test that a drive is fully operable, as well as each part individually. Also, the sheer number of attributes that can be checked is enormous, so the is the length and “weight” of that part of the microcode.
Any error during reading of critical modules (ID, Configuration, Translator…) will make it impossible for the drive to correctly identifies to system BIOS, and that will bring about data loss.
In these cases there are different principles of reparation from using backup copies (physically present on other platters) to using donor modules, or even more advance, to boot up disk with the help of a working donor drive. All these procedures of data recovery are safe and bring about complete data recovery.
Unfortunately, and it happens more and more frequently, IT people are using cheap Chinese tools and attempting to “automatically” repair disk firmware, which results in serious damage to vital data. When those disks come to our laboratory, they need at least double the amount of time to repair the damages, but the outcome is usually positive.