Jump out of hard drive misunderstanding

Several common sense questions that have been misleading Gao Peng for many years are:
1. Hard Disk logical bad sectors can be repaired, but physical bad sectors cannot be repaired. The actual situation is that bad channels are not divided into logical and physical bad channels. I don't know who invented these two concepts. The technical materials provided by the manufacturers do not have such a concept, it is divided into bad sectors recorded by logical address and bad sectors recorded by physical address.
2. There is no bad track when the hard disk is released. If a bad track is detected, the hard disk enters a dangerous state. The actual situation is that each Hard Disk records a certain number of bad sectors before leaving the factory, and some even reach thousands or tens of thousands of bad sectors. In contrast, how dangerous do users find one or two bad sectors?
3. Hard disks cannot be recovered if they are not recognized. If the 0 channels are damaged, you can use the partitioning method to solve the problem. The actual situation is that a considerable number of hard disks that are not recognized can also be repaired, but it is difficult to partition when the 0 channels are bad.
This is misleading. For example, if you have not collected and studied foreign-language materials for a long time, you cannot say that you have used them as your creed for a long time. There are many professional hard drive repair forums in foreign countries, where you can find that hard drive repair technologies in some countries have reached a high level. I'm sure some of their technologies will be a headache for many hard drive manufacturers. Communicate with many professional hard drive repair experts in the world, making Gao Peng more interested. Over the past three years, Gao Peng has resigned as a teacher and specializes in hard disk repair. He has already repaired more than hard disks.
To sum up, Gao Peng has three sources of technology:
1. Collect foreign technical information and exchange with foreign professionals;
2. Purchase professional tools and software (support for synchronized technical updates );
3. Your own practical experience.
Unfortunately, I didn't find a teacher who taught me how to fix the hard disk, and I don't think that this textbook is too helpful for me to repair the hard disk.
Several Basic concepts that hard drive repair personnel need to understand
Some basic concepts will be involved in hard disk repair and repair using professional software. Here, based on his own research and practical experience, Gao Peng tries to summarize and explain some concepts related to "Hard Disk defects" and communicate with readers.

Bad sector (bad sector)
The sectors that cannot be normally accessed in the hard disk or read or written correctly are called bad sector. One slice can store bytes of data. If any byte in a slice cannot be correctly read or written, the slice is bad sector. In addition to bytes, each slice has dozens of bytes, including the ID, check value, and other information. Any byte error in this information will cause the sector to change to "bad ". For example, in the low-level formatting process, each slice is assigned a number, which is written in the ID. If an error occurs in the ID part, this slice cannot be accessed. This slice is a bad sector. Some bad sector can rewrite this information through low-level formatting.
Bad Cluster)
After you partition a hard disk and perform advanced formatting, a file allocation table (FAT) is created for each partition ). Fat records the usage and links of all clusters in the region. If you find that the slice used by a cluster includes a bad slice during the advanced formatting (or tool software scanning) process, the fat Column records the cluster as a bad cluster, the cluster will not be used for storing files in the future to avoid data loss. Sometimes viruses or malware may mark normal clusters without bad sectors as bad clusters in fat, so normal clusters cannot be used. It should be emphasized that each cluster contains several sectors. As long as one sector exists, the rest of the sectors in the entire cluster will not be used together.
Defect (defect)
All defective parts in the hard disk are called defect. If a head is in bad state, the head is defect head. If a track (track) on the disk cannot be normally accessed, the track is defect track. if a sector cannot be normally accessed or data cannot be correctly recorded, this sector is also called the defect sector. it can be considered that bad sector is equivalent to defect sector. in general, a hard disk is called defect hard disk as long as it has some defects.
P-List (permanent defect table)
Nowadays, the hard disk density is getting higher and higher, and the data volume stored on a single disk exceeds 40 gbytes. hard Disk manufacturers are extremely sophisticated in the production process, but it is also extremely difficult to achieve 100% perfection, hard disk surface more or less there are some defects. Before the hard disk leaves the factory, the manufacturer uses low-level formatting for all hard disks. During the low-level formatting process, all defect track and defect sector are automatically located and recorded in the p-list. In addition, during the numbering of all the channels and sectors, skip (skip) these defects so that users can never use them. In this way, it is difficult for users to find problems when partitioning, formatting, and checking the new hard disk they just purchased. A general hard disk records a certain number of defect records in the p-list, with hundreds of records and tens of thousands of records. If it is a SCSI hard disk, you can find a variety of general software to view the p-list, because various brands of SCSI hard disks use compatible SCSI instruction sets. Different brands of different models of IDE hard drives use different instruction sets and want to view their P-list using targeted professional software.
G-List (growth defect table)

When using the hard disk, you may find some new defect sector. According to the "three packages" rule, as long as a defect sector appears, the merchant should change or repair the user. The probability of a defect sector appearing on a large-capacity hard disk is very high. In this case, hard disk merchants will be busy with after-sales services. As a result, hard drive manufacturers have designed an automatic repair mechanism called automatic reallcation. Most hard disks have the following features: If a defect sector is found during hard disk read/write, a backup sector is automatically allocated to replace this sector, and record the sector and its replacement in G-list. In this way, a small number of defect sector does not have a great impact on your use.
There are also some hard disk auto-repair mechanisms that require strict excitation conditions. Some software is required to determine the defect sector and call the automatic repair mechanism through a port (which is said to be 50 h. For example, lformat, ADM, zero fill in DM, wipeinfo in Norton, wddiag in the western data kit, and erase in the ibm dft. The reason why these tools can eliminate some "Bad channels" after they run is the automation reallcation (of course there are other reasons ), rather than simply summarizing what "bad track" is, "logical bad track" or "false bad track ". If a misleading and poisoned reader does not believe this fact, he will know after finding a professional tool that can view G-list. After these tools are run, how many records will be added to G-list! Is it necessary to record "logical bad track" or "false bad track" in G-list and replace it with other sectors?
Of course, the records of G-list are not limited, and all hard disks are limited to a certain number. For example, the maximum number of fireball series is 500, the limit of the second generation of the US diamond is 636, and the limit of the West BB is 508. If the limit is exceeded, the automatic reallcation will no longer work. This is why a small number of "bad channels" can be repaired using the above tools (some people will say that "logical bad channels" can be fixed ), however, when there are many bad channels, they cannot be repaired using these tools (some people say that "Physical Bad channels" cannot be repaired ).
Bad Track)
This concept originated from a small-capacity hard disk (less than MB) more than a decade ago. At that time, the hard disk was attached to a small table on the shell, which listed the defective Track location of the hard disk (as well as the new hard disk ). During Low-Level formatting of the hard disk (such as using tools such as ADM or DM 5.0, or low-level tools in the motherboard), you need to fill in the location of these bad tracks, this allows you to skip these tracks during the low-case process. The structure of the current large-capacity hard disk is very different from those of small-capacity hard disks. This concept is far-fetched for large-capacity hard disks.

Readers may also find that many domestic publications and online articles have the following concepts: Physical Bad Sectors, logical Bad Sectors, real bad sectors, false Bad Sectors, hard bad sectors, soft Bad Sectors, etc. Gao Peng did not find the corresponding English concept in the hard drive technical materials outside China. Maybe the Chinese people have summarized it themselves, right? Since so many people can accept these concepts, some experts may give some reasonable explanations. Gao Peng is not used to these concepts and does not want to explain them far-fetched. Readers should ask who said what they said.
Learn more about hard disk Parameters

Under normal circumstances, after the hard disk is powered on, it must be initialized (also known as "self-check "). At this time, the self-check sound will be sent for a while. The length and regularity of these sounds vary according to different brands of hard disks, but the self-check sound of normal hard disks of the same model is the same. Experienced people know that these self-check sound is triggered by the seek and retrieval actions of the head inside the hard disk. Why are so many actions required when the hard disk is powered on? In short, it is the initialization parameter of the record read by the hard disk in the disk.
Generally, people familiar with hard disks know that hard disks have a series of basic parameters, including brands, models, capacity, number of cylinders, number of cores, number of sectors per track, serial number, cache size, speed, and S. m.a. r. T value. Some of the parameters are written on the hard disk label, and some are tested by software. However, Gao Peng tells you that these parameters are only a small part of the initialization parameters. There are dozens or even hundreds of initialization parameters recorded in the disk! After the hard disk CPU is powered on, it automatically finds the Startup Program in bios, and then reads the corresponding parameters at the specified position in the disk according to the startup program requirements. If an important parameter cannot be found or an error occurs, the startup program cannot complete the startup process, and the hard disk enters the protection mode. In protection mode, you may not be able to see parameters such as the hard disk model and capacity, or you may not be able to access any read/write operations. Recently, some series of hard disks have similar common faults Due to this reason. For example, the self-check sound of the Fuji MPG series is normal but the disk is not recognized. The maxtor series cannot recognize the correct model and stop switching after self-check, the wd bb eb series can recognize disks but reject read/write operations.
Different brands of hard disks of different models have different initialization parameter sets. Taking the familiar Fujitsu hard disk as an example, Gao Peng briefly explained some of the parameters so that readers can understand the principle of internal initialization parameters.
Use a dedicated program to control the cpu Of the hard disk, read the initialization parameter set in sequence according to the needs of the BIOS program, and store 69 different files by module, the file name is also consistent with the parameter name called in the BIOS program. Some parameter modules are described as follows:
Basic Management Program in DM Hard Disk
-Pl permanent defect table
-Ts defect track table
-Actual number of physical magnetic heads and sequence of HS
-Sm highest-level encryption status and password
-Su user-level encryption status and password
-Ci hardware information, including the CPU model, BIOS version, head type, and disk Disk Type
-Fi manufacturer information
-We write an error log table
-Re read error log table
-Si capacity setting, which specifies the maximum capacity (max LBA) that can be used by users and converts it to the number of external logical magnetic heads (usually 16) and the number of logic sectors per track (usually 63)
-ZP region allocation information: divide each disc into fifteen regions, and allocate different slices in each region to calculate the maximum physical capacity.
These parameters are generally stored in locations that ordinary users cannot access. Some parameters are considered to be in the position of the negative track before the physical zero track. Each parameter may occupy one module, or several parameters may occupy the same module. The module size is different. Some modules have only one byte, while others have 64 K bytes. These parameters are not stored consecutively, but have their own fixed positions.
After reading the internal initialization parameter table, you can analyze whether each module is in normal state. Of course, you can also modify these parameters and re-write them back to the specified position in the disk. In this way, some hard disks that cannot be normally used due to incorrect parameters can be "fixed" back to normal.
If you are interested in further research, you may wish to remove the ROM chip from the hard disk board and read the BIOS program from the decoder. You can find the parameter names listed above in the program segment.
Low-level formatting for hard disk repair

People familiar with hard disks know that they need to perform "low-level formatting" (hereinafter referred to as "low-level") on hard disks when necessary "). There are also a variety of tools used for low-level: low-level tools used by specialized equipment manufacturers, low-level software tools provided by manufacturers, low-level tools used by DM, low-level tools used in the motherboard BIOS, low-level debugging tools, and low-level professional software ......
Different tools have different effects on hard disks. Some people think that low cells can repair some hard disks, while others think that low cells are very dangerous and will seriously damage the hard disk. Gao Peng has used a variety of low-cell tools and believes that low-cell is an effective way to repair hard disks. The following summarizes some low-level views and communicates with the majority of users.
One of your concerns is: "What operations have been performed on the hard disk during the low-case process ?" Practice shows that the following work may be performed in the low-case process. The low-case process of Different Hard Disks varies greatly, and the low-case process of different software varies greatly.
A. Check the slice clearing and rewriting.
In the low-level process, all the bytes in each sector are set to zero, and the check value of each sector is also written back to the initial value. This can correct some defects. For example, because the sector data does not correspond to the validation value of this sector, an ECC error is usually reported ). If it is not due to magnetic medium damage, it is very likely that the sector data will be re-matched with the validation value of the sector after being cleared, to achieve the effect of "repairing" the sector. This is the most basic operation content for each type of low-level tool and each type of hard disk, and it is also the basic reason why the low-level tool can "repair a large number of bad channels. In addition, the zero fill (zeroing) operation in DM has the same effect as the erase operation in the ibm dft tool.

B. Rewrite the ID information of the slice.
Old hard disks used many years ago (such as hard disks using the st506 Interface) need to re-write the ID information of each sector and some other information that retains the track in the low-case process, at that time, non-standard tools must have such features. However, the current hard disk structure is quite different. If you use the tool for low-level storage many years ago, it will lead to many painful accidents. It's no wonder that people often shout in pain: "dangerous! Do not uncompress the hard drive! My hard drive has been broken down !"
C. Perform read/write checks on the slice and try to replace the defective slice
Some low-level tools perform read/write checks on each slice. If an error occurs during the reading or writing process, the slice is considered as a defective slice. Then, call the general automatic sector replacement command to try to replace the sector, which can also achieve the "Repair" effect.
D. Re-number all physical sectors
The number is based on the record and segment allocation parameters in the p-List (this parameter determines the number of sectors of each track, each slice is allocated with a specific ID ). When the number is specified, the system automatically skips the defective sectors recorded in the p-list so that users cannot access those defective sectors (users do not have to care about the quality of the defective sectors that will never be used ). If this process is abandoned halfway, some or even all sectors may be reported as being identified incorrectly (Sector id not found, idnf ). Note that this numbering process is based on real physical parameters. If some low-CASE tools perform low-case calculation based on logical parameters (16 Heads 63sector is the most typical, it is impossible to perform such operations.
E. Write the track servo information and re-number all tracks
Some hard disks allow you to overwrite the servo information of each track and re-assign a number to the track. The number skips the defect track based on the p-list or TS record, making it inaccessible to the user (that is, it never needs to be used. This operation is based on real physical parameters.
F. Write status parameters and modify specific parameters

Some hard disks have a status parameter that records whether the low-level process ends normally. If the low-level process is not completed normally, the entire hard disk will reject read/write operations, this parameter is typical of Fujitsu IDE hard disk and Seagate SCSI hard disk. Some hard disks may also modify some parameters based on records in the low-case process.
Next let's take a look at some of the operations that some low-level tools have done:
1. Low Level format in DM
Operations A and B are performed. It is faster and rarely damages the hard disk, but the repair effect is not obvious.
2. lformat
Operations A, B, and C are performed. Because the read/write check is performed at the same time, the operation speed is slow and some defective sectors can be replaced. However, it uses logical parameters, so it is impossible to perform operations on D, E, and F. When an idnf error or a servo error occurs, it is difficult to pass through and the half path will be interrupted.
3. Low-cell tools in the scsicard
Because most SCSI hard disks have a general instruction set, this tool can perform a, B, c, d, and f operations on some SCSI hard disks, and has significant effect on some SCSI hard disks (such as Seagate. The faulty track fails. In addition, because of the automatic replacement function, when the number of defects detected exceeds the G-list limit, the system will end halfway and the hard disk will enter the read/write rejection status.