How it is possible that the health of the hard disk is low, the text description shows the bad sectors but the tests do not show any problems? When the hard disk finds a bad sector, it tries to save (reallocate) the contents from the sector to the spare area. When this completes, the original sector location is marked and the disk does not use it any longer: all further read and write operations are redirected to the spare sector. This way the tests (which does not only mean tests of this software) can only access this modified surface. In an optimal situation, the surface is really perfect – the problems are fixed, bad areas are no longer used. The problem can be caused only if the bad area is relatively large (thus the health is low) and more and more problems (for example weak and bad sectors) born. To fix these issues, it is recommended to use the tests built into Hard Disk Sentinel because they can verify and fix the problems under a controlled environment. For more information about problems, please check the text description section. All tests are designed to reveal disk problems. The health is calculated based on the problems found with the hard disk, the test may cause a decrease in health. The health could increase only if the problems would become “hidden” (but not fixed!). This is not possible – except if the drive had weak sectors before the test and their count can decrease (even back to zero). The count of other problems detected (and fixed) by the hard disk cannot be decreased and would not be good, but by using the “offset” column on the S.M.A.R.T. page these problems can be ignored if the drive status is stable. For more information, please check the text description and the temperature calibration sections. Why does the test find more or less bad sectors than displayed in the text description? The bad sectors that appear in the text description are already reallocated by the hard disk, which is no longer used. The bad sectors revealed by the test are still on the active surface so these can potentially cause data loss. Our goal is to force the hard disk to reallocate these sectors also to prevent their future use. Usually, the disk works better (even its health value may decrease because of the reallocation). Can health be improved? Can it be restored to 100 %? If the hard disk had weak sectors, its status can be improved during the tests. This may result in increased health value – even to 100 % if there were no other problems found with the drive. In all other cases, the health may decrease if the status of the drive is not stable and newer problems are revealed. If the drive is stable: the health does not change even after many passes of the write tests, the previous problems may be cleared on the S.M.A.R.T. page by using the Offset column. This way the health can restore to 100 %. Warranty replacement is possible only at 0 % health. If the health is higher (but still low) it is recommended to start the tests (even many times) to reveal further, currently not detected problems. Why it is good to use the disk tests of Hard Disk Sentinel? Why it is not enough to simply copy files (to force reallocation)? During the tests in the software, the hard disks are under better control and monitoring compared to plain file operations. The file system is unmounted to prevent the operating system and other software from accessing the disk and the status of the file system (for example fragmentation) does not affect the test. The performance of the blocks, retry count and disk health and temperatures are all factors that are continuously monitored during the tests. It is even possible to inspect the sector contents at any time. Like surgery is only possible in hospital – hard disk testing and repair has its own controlled environment. Types Of Tests Generally, two different tests are available: hardware self-test, and software testing (both are available in Hard Disk Sentinel). The latter category contains the random seek time test also, which is a special test as it does not verifies the surface, just the seeking capabilities of the drive. The hardware tests can be better than software methods because they can examine the disk system and components (for example head, servo, electric components and so) in a special way which is not possible by software methods. It does not affect the stored information so they can be used any time (on IDE and S-ATA hard disks which support them), without any special preparation or backup. The use of hardware tests is recommended periodically, especially on perfect (or near-perfect) disks to reveal any hidden problems. In some cases, these hardware tests (Disk -> Short self-test, Disk -> Extended self-test) are not available, not supported or result in an error quickly even in a relatively low number of problems. No further information is returned about the result, for example, it is not possible to list the sector(s) which are damaged. In such a case, an appropriate software testing method is required. When we speak about software testing, it is important to know that the WRITE operation affecting the disk surface is the most important: this forces the hard disk to verify the integrity of the data sector, reallocate it if required, and repair the weak sector. Because of this, all READ-type tests (for example, ScanDisk but the format also) cannot be used to repair disk problems (and in many cases, they even can’t detect such problems). Scandisk verifies and fixes logical errors only (file system-related errors) which are not really connected to physical disk status, errors (this is why they usually vanish after complete re-partition and re-format of the disk). The WRITE tests are usually destructive as they need to overwriting the stored data. In all cases be very careful before starting any of such disk tests, it is important to backup all data! The software tests are recommended to use less frequently compared to hardware tests. Software tests are useful when the new empty hard disk is installed in the system hardware self-tests complete with error the hard disk is suspicious (for example, making weird noises) and/or its health changes it is required to refresh the stored data on hard disk or SSD/flash device (without erasing data) which is recommended to use periodically to prevent the cells to “forget” the stored information. Its frequency is based on the actual usage of the device. In such cases, it is recommended to use the reading test first which does not affect stored information. The reading test cannot fix any errors but can show which areas of the drive are not perfect, damaged, and/or slowly accessible and shows the temperature and performance characteristics of the drive. The real verify and fix of errors is possible later by a suitable write test. In case of more problems, it is recommended to back up the data and use the reinitialize disk surface test (even with a level 3 or higher). Other software may (incorrectly) call a simpler version of this test a low-level format. IMPORTANT: The developer offers no warranty on the stored data on any disks! Under all circumstances, it is recommended to backup ALL DATA and read the description for the tests carefully before starting any of the tests.

Hardware Self Tests

TR.rh { background-color: #ffffff; height=18 } TR.rw { background-color: #f0f0f0; height=18 } TD.fe { font: 80% arial; text-align: left; }

Hardware Self Tests
The hardware self-tests built-in IDE / Serial ATA / SCSI / SAS hard disks are the perfect solution to test the hard disk status quickly or with very high detail. Usually, the short and extended self-tests are supported by the hard disks but some models support other (conveyance) self-tests also. The list of all supported self-tests and the estimated time of such tests are displayed on the Information page of the hard disk.
The self-tests are non-destructive, they do not affect the data stored on the hard disk. During the test, the hard disk is still usable but may be slower (and vice versa: hard disk usage may make the test much longer. It may be important for a system hard disk where the op. system or other software may cause much read/write operations).
The Short self-test verifies the major components of the hard disk (read/write heads, servo, electronics, internal memory, etc). This test takes only some minutes.
The Extended self-test performs the above and it has a complete surface scan which reveals the problematic areas (if any) and forces the bad sector reallocation. It is recommended to periodically use this test to verify the disk health – especially on a hard disk with less than 100% health.
The Conveyance self-test (if supported) performs manufacturer-specific test steps. This usually verifies the mechanical parts of the hard disk to ensure that no handling damage occurred.
The date and results (successful completion or the failing hard disk element) of the last self-tests are displayed on the Information page. On this page, all supported hard disk tests can be started manually – or it is possible to configure automatic (scheduled) hard disk tests on the Operations page (Professional version only).

Random Seek Test

By this test it is possible to verify the noise level, track-to-track seek time, and temperature levels of the hard disk, especially under excessive disk usage.
The test is especially useful after changing the acoustic level to verify how the modified setting affects the noise and seek time. This verification can be done by a short (quick) test.
The test can be used to verify the stability and temperature increase of the disk during a longer period of high usage. It is possible that during the test the temperature can increase by 10-15 Celsius degrees (compared to the idle temperature) so if the hard disk was too hot before starting the test, it is recommended to use the test only under control and cancel it upon reaching a critical temperature level.
Recommended use of the test: after installation of a new hard disk. A longer test (8 hours or longer) can be used as stress-testing, to verify the stability of the disk.
The test can be used even if the hardware self-test stops with an error or if it is not supported by the hard disk. The test is safe, it does not affect the stored data, just positions the read-write head of the disk.
Note: the unregistered version allows only the shortest test (2 minutes) and only once on one disk.

Surface Test

Surface Test
By using this test, the surface of the hard disk can be examined with different methods. Some of these methods are safe for the stored data but others are destructive: the stored information is permanently erased during the test operation.

First, it is required to select the hard disk and the test type to be started. The detailed description of the test is displayed upon selection, and the software displays how the test affects stored data. It is strongly recommended to read this information carefully before starting the test.

The following tests are possible:
READ Test – Reads the disk surface to verify if all sectors are accessible and readable.

The test is safe, it does not affect stored data.

Disk problems may remain hidden and revealed disk issues are NOT fixed (like with ScanDisk).
WRITE test – Overwrites the disk surface with the configurable pattern. Forces the analysis of any weak sectors and verifies any hidden problems and fixes them by reallocation of bad sectors (this is drive regeneration).

Enables better use of the disk as after the test the spare area will be used instead of any problematic sectors.
WRITE + read test (Pro version only) – Overwrites the disk surface with configurable pattern and then reads back sector contents, to verify if they are accessible and consistent. Forces the analysis of any weak sectors and verifies any hidden problems and fixes them by reallocation of bad sectors (this is drive regeneration).

Enables better use of the disk as after the test the spare area will be used instead of any problematic sectors.
Read + WRITE + read test (refresh data area) (Pro version only) – Reads stored data from each block, writes back the contents, and finally reads the information and compares with original contents. By the extensive test, an additional write cycle is used before writing back the contents to improve the efficiency of the error correction (drive regeneration).

The operation is usually safe for the stored information but data loss may occur if the system is not stable and/or upon power failure, overclocking memory/power supply/cable problems, and other factors.

The test can be used to refresh the data area of the storage device, without the need for a complete erase but it is still recommended to backup important data before this test.
Reinitialize disk surface (Pro version only) – Overwrites the disk surface with a special initialization pattern to restore the sectors to default (empty) status and reads back sector contents, to verify if they are accessible and consistent. Forces the analysis of any weak sectors and verifies any hidden problems and fixes them by reallocation of bad sectors (this is drive regeneration).

Enables better use of the disk as after the test the spare area will be used instead of any problematic sectors.
Very intensive and time-consuming test, especially if the level (the number of overwrite cycles) is set to a higher value.

The simple version of this test is usually (incorrectly) called a low-level format by other tools. It can be effectively used to permanently destruct stored information (Navso P-5239-26 standard data destruction).
Each test has different configuration options:

In general, it is possible to configure the order of blocks to be tested and optionally the repeat count can be specified (so the software runs the same test on the disk two or more times) and it is possible to unmount all logical volumes (partitions) during the test. The latter is strongly recommended to prevent other software and the operating system from accessing the disk during the test. System drive (containing the operating system, the root folder of the user, or the swapfile of Windows) cannot be locked this way. Because of the actual system running, only read test (which does not affect stored data) can be used on a system drive.
Note: some tests (for example the Reinitialise disk surface and Refresh data area tests) and test options are available only in the Professional version of the software.
Note: to prevent unauthorized people from using the tests (especially the write tests) the software asks for the configured password before opening the test window.

Test Details
The software divides the data area into 10000 blocks, this is the surface map of the disk. When any block is clicked by the mouse, the contents of the sector under the selected block are displayed. This way it is possible to inspect the stored data and/or the written data after performing a writing test. It is possible to examine the sector contents (and load back the results of a previously saved test) without starting a test by clicking the View button. This is safe for the stored data on the disk.
During the tests, Hard Disk Sentinel verifies the integrity of each sector under the displayed blocks. If the block is accessible and can be processed without errors, they are shown by green color. Darker green color means that the sectors are slower and harder to process. The software filters out the effect of slower transfer times at the end of the disks. If sectors are harder to access and can be processed only after one or more retries, then the blocks are damaged but repaired, the software shows the corresponding blocks with yellow. Red color means that the block of sectors cannot be processed (read or written – according to the selected test). Further WRITE test may be required to force the drive to reallocate the problematic sector(s) in the block.
In case of damaged or bad blocks found, it is recommended to examine the whole data surface with one or more WRITE test(s). During write tests, the software forces the hard disk to verify the status of the sectors on the surface and re-allocate them to the spare area. This way the original (bad) sector will not be used and all further read and write operations are redirected to the spare area. This is called regeneration. The resulting hard disk functions better, it is even possible that on following tests it shows only good (green) blocks because the bad sectors are no longer used. The side effect of regeneration is that the health of the drive may decrease because more and more problems are revealed. So regeneration usually does not improve health (except if weak sectors were found and their status if stabilized).
During the tests, if bad block(s) are found, it is possible that the software and/or the operating system are not responding during the re-allocation. This can take some minutes or even hours – depending on the size of the affected surface area. It is not recommended to cancel the operation and avoid reset or power off the computer. It is possible that the disk controller or its driver may disconnect the not responding disk so it disappears from the system. This is more likely possible with a USB connection. If this happens, it is recommended to connect the hard disk directly to the motherboard or a different controller and start the test again.
Recommended use of the tests: start an extensive write test (for example with “reinitialize disk surface”) to verify if the disk surface is usable and force the verify and reallocation of any sectors at this time. Like the random seek test, the software measures temperature levels during these tests, so the highest possible temperature can be checked during this extreme load. The tests can be started on another type of storage devices (not only hard disks), for example on SSD devices, FLASH memory cards, pen drives also can be tested – and the “read-write – read” test can be effectively used to prevent data loss caused by the “forget” effect of the memory cells.
The tests can be useful even if the hardware tests stop with an error, not supported, or not available
Note: the unregistered version does not allow the starting of any test(s). It is only possible to view sector contents with unregistered versions.
Note: during the tests, software and the operating system running in the background may affect the performance (thus the resulting color) of each block. To minimize this effect, the software prevents the screensaver from starting, prevents the computer from entering any standby or lower power management mode, and using a higher priority. Anyway, it is recommended to close all other background operations during the test and repeat the test if the performance decrease affects a larger area (not only 1-1 blocks). The software has the option to repeat the test to minimize this momentary performance decrease.

Examples

The end of the usable data area is bad. The bad blocks form continuous areas, suspecting a serious surface problem. Notice the yellow damaged block near the lower-left corner: it is neighbor of a bad block and will probably turn to bad at later use. Even non-damaged blocks are very slowly accessible. It is recommended to check the location of the first problematic block with the mouse pointer and create a partition(s) that do not use the following area.

The hard disk showed many problems. It stopped during reading and writing operations and its health was relatively low (30 %). There were different bad, damaged, and slowly accessible blocks.

The same disk after many “Reinitialize disk surface” tests. The weak sectors are fixed and the bad sectors are reallocated (the health decreased to 20%) but the overall surface is improved and the disk can still be used for some time (but not recommended to store important data because of the problems).

Typical low health (22 %) hard disk. Many write tests (especially the re-initialize disk surface test) can improve the status but it is not recommended to use the first half of this drive (by creating partition(s) only at the second half).