S. M. A. R. T. technology allows you to read the information stored in the service area of the hard disk, necessary to assess its condition. The interpretation of the term is as follows: Self – itself, Monitoring – control, Analysis – analysis, Reporting Technology – reporting technology. How and why to use S. M. A. R. T. is discussed in detail in this article. Check microphone sound online.
For SMART analysis we use the CristalDiskInfo program.
Lines 05, C5, C6 show basic information.
Line c5 – Unstable sectors (Current_Pending_Sector).
Line 05 – Reallocated Sectors(Reallocated_Sector_Ct )
Line c6 – Uncorrectable sector errors (Offline_Uncorrectable)
Switch the language to English!
How to read SMART. Not for SSD.
198 Offline_Uncorrectable 0x0008 100 100 000 Old_age Offline – 0
Indicates the number of bad sectors from which an attempt was made to read.
If a non-zero number appears at the end, the hard disk is scrapped.
5 Reallocated_Sector_Ct 0x0033 100 100 005 Pre-fail Always – 0
197 Current_Pending_Sector 0x0022 100 100 000 Old_age Always – 0
The number of sectors suspected of failure. After some time, they turn into Offline_Uncorrectable or Reallocated_Sector_Ct. If the value is 0 and the number of Reallocated_Sector_Ct does not grow, then the broken sectors described in paragraph No. 2 can be scored.
9 Power_On_Hours 0x0012 093 093 000 Old_age Always – 50172
The number of hours that the hard drive has been running.
4 Start_Stop_Count 0x0012 100 100 000 Old_age Always – 143
The number of times the hard disk is turned on/off.
The allowable amount is calculated according to the formula Start_Stop_Count / Power_On_Hours. (smartctl code 9) If the value is greater than 20 then:
– you have a laptop, you should not pay attention to it;
– disk firmware is wrong, so the hard disk is constantly parked, which threatens with an early failure without the possibility of data recovery;
– costs windows vasyan edition;
– the dude set the power management parameters to the letter M.
In general, if you do not have a laptop and a value of more than 20, but a hard one for more than a year, then get ready for a possible sudden failure.
191 G-sense error rate
The number of physical hits hard. If the value is greater than 10, it means someone was shaking it while working.
Running smartctl on linux: smartctl -a /dev/sda (sda is the drive name.)
The command is executed with administrator rights. (Disk names can be found in the gparted utility)
Attribute value S.
For each of the attributes, the test program will display the following information (depending on the application, they may differ slightly from the list here):
Here you should pay attention to the colors of the attribute indicators. Green color indicates that the corresponding indicator is normal. If some attributes fell into the yellow zone, the situation should be regarded as alarming. If the indicator turns red, the state of the hard drive is critical, and it can completely break down at any time.
Consider each of the S. M. A. R. T. attributes of a hard drive.
This indicator is used to determine the number of errors that occur when reading data from the hard drive. Its values may be interpreted differently depending on the device model. For some manufacturers, a zero value is considered ideal, for others, the more, the better.
02 Throughput Performance
Displays the average performance of the drive. There are no strict rules for him. For diagnosing the HDD is practically useless.
03 Spin-Up Time
Allows you to set the time required for the hard drive to spin up. By itself, this parameter means little. It should only be judged against the stated specifications of the specific hard drive.
04 Number of Spin-Up Times (Start/Stop Count)
Shows the number of times the hard disk has been powered on during its lifetime. Can be used to get an indirect estimate of the duration and intensity of device use.
05 Reallocated Sector Count
One of the most important attributes that allows you to determine the physical condition of the hard drive. Shows the number of bad sectors replaced by healthy ones from the spare area. This replacement is called a remap. Remap is performed automatically if reading information from any part of the disk is very difficult or impossible. In this case, the bad sector is marked as bad, so that the operating system will no longer try to use it.
It should be understood that the reserve area is not unlimited, and when the backup possibilities are exhausted, the hard disk will be irreversibly destroyed. The number of spare sectors for different models of hard drives is different. But their maximum number does not exceed several thousand (most often not more than a thousand).
Displays data that can be used to determine the frequency of occurrence of failures during positioning of the magnetic head unit. Similar in many ways to the Raw Read Error Rate attribute. The difference is that only zero is considered normal for Hitachi drives. On Seagate drives, Samsung SpinPoint F1 and its newer models, as well as Fujitsu 2.5 ”, this indicator should not be taken into account at all.
08 Seek Time Performance
Shows the average performance of disk head positioning operations. There are no limit values for it.
09 Power On Hours Count (Power-on Time)
With this parameter, we can find out how many hours the hard drive has been running since it was first used.
10 (0A) Spin Retry Count
Allows you to determine how many times the spindle has been restarted since the first unsuccessful attempt to start it. However, the growth of this indicator does not always mean a physical malfunction of the hard drive. In most cases, the problem is due to poor contact between the HDD and the power supply or insufficient power received by the device. If the attribute value does not exceed 2, then everything is in order. Otherwise, you should check the power supply and its contact with the hard drive.
11 (0B) Calibration Retry Count (Recalibration Retries)
This displays the number of retries to reset the media (this procedure sets the magnetic heads to track zero) after the first failed attempt was registered. If the attribute value is zero, there are no problems; if not, the device is most likely faulty.
12 (0C) Power Cycle Count
The total number of cycles “on-off” of the hard drive is noted.
This parameter stores how many SATA downgrade attempts failed. The fact is that when certain errors are detected, the HDD may try to switch to work in a mode with a lower speed. Such switching will fail if the controller for some reason refuses to execute the incoming command. But in any case, this has nothing to do with the health of the drive.
It makes it possible to estimate how many errors occurred in the process of transferring information through the hard disk cache during the entire period of its use. Any non-zero value can indicate a problem with the device.
Indicates the number of sectors to be reassigned soon. Sometimes a sector may be repeatedly identified as a candidate for reassignment, which also causes the attribute value to increase. If this line is not zero (especially when attribute 197 is also not zero), destructive changes began to occur with the hard drive.
188 (BC) Command Timeout
Stores data on how many operations had to be interrupted due to exceeding the maximum allowable waiting period. Any value greater than zero indicates the presence of such failures. But this is not always due to a hard drive failure. The problem can also occur when using low-quality cables, bad adapters, damaged contacts, incompatibility with the SATA / PATA controller on the motherboard. In Windows, such an error can be manifested by the appearance of a “blue screen of death”.
189 (BD) High Fly Writes
Shows how many media writes were recorded when the head speed exceeded the calculated value. The main reason for this phenomenon is external influence (shocks, shocks, vibration). However, there are no standards for this item.
190 (BE) Airflow Temperature
Displays the hard disk temperature at the time of testing. Heating above +55 – + 60ºC negatively affects the operation of the device. In this case, it will be useful to install additional cooling.
191 (BF) G-Sensor Shock Count (Mechanical Shock)
This parameter can be used to determine the number of critical accelerations of the HDD head. The reasons for their appearance may be the fall of the drive or blows to its case. But even if such accelerations were registered by the device’s sensors, this does not mean that it was damaged. HDD status should be evaluated taking into account the values of other attributes. It should also be noted that this parameter can be omitted for Samsung hard drives, since its sensors can almost react to air movement.
192 (C0) Power Off Retract Count (Emergency Retry Count)
The information displayed in the corresponding line depends on the model of the device. Here you can display either the total number of magnetic head parking operations performed in case of emergencies, or the number of on / off cycles of the device for the entire time of its operation.
193 (C1) Load/Unload Cycle Count
Shows the total number of parking and unparking cycles of the magnetic heads of the drive. With this parameter, we can find out if automatic HDD parking is enabled. If the value of attribute 192 is greater than the value of attribute 09, then automatic parking is enabled and in use.
194 (C2) Temperature (HDA Temperature, HDD Temperature)
Displays the temperature of the hard drive at the time of reading information from S. M. A. R. T. It may also contain information about the minimum and maximum temperatures of the device recorded during its operation. You need to make sure that the hard drive does not overheat (the maximum allowable temperature is +55ºC).
195 (C3) Hardware ECC Recovered
Allows you to determine the total number of errors handled by the ECC HDD hardware. It is analogous to attributes 01 and 07.
196 (C4) Reallocated Event Count
One of the most significant attributes for determining the real state of the hard drive. The higher its value, the worse things are. But in order to give an objective assessment of the state of the device, the values of other parameters should also be taken into account.
This indicator is closely related to attribute 05. If one of them began to deteriorate, negative changes usually begin to occur with the other. If the changes affect only attribute 196, this means that during the remap it turned out that the problems with the sector were due to a violation of the logical structure, and not a physical malfunction, and were eliminated by means of the hard disk.
Sometimes a situation arises when the value of attribute 05 is greater than that of attribute 196. In this case, several sectors were remapped simultaneously.
197 (C5) Current Pending Sector Count
Displays information about the number of sectors to be reallocated. But they do not always have a physical malfunction. Only candidates that have received the bad status are redistributed, and sectors with the soft status (logical error) after their correction become usable again.
198 (C6) Offline Uncorrectable Sector Count (Uncorrectable Sector Count)
Similar to attribute 197 in many ways. The main difference is that attribute 198 shows a fixed number of remap candidates identified during offline testing (it starts during downtime).
This indicator allows you to determine how many errors occurred during the execution of information transfer operations over the interface cable, carried out in UltraDMA mode. If there is a tendency for the parameter to increase, this may indicate a poor-quality or damaged data transfer cable, the operation of the PCI / PCI-E buses in overclocking mode, or a poor connection of the SATA cable to the corresponding connector on the motherboard or hard drive.
When such errors occur, the HDD may automatically switch to PIO mode, resulting in a noticeable decrease in its performance. In most cases, the problem is solved by reconnecting the interface cable or replacing it with a new one.
This parameter is responsible for the number of errors registered when writing to the media. If their number is steadily increasing, the hard drive can no longer be considered a reliable device. First of all, this applies to WD drives. For them, high attribute values of 200 may mean that the writing head will fail soon.
Shows how many errors occur while reading information.
High values of this indicator indicate problems that arise during the operation of the hard drive.
203 (CB) Run Out Cancel
The number of ECC errors is fixed here.
220 (DC) Disk Shift
Allows you to know the value of the shift of the plates in relation to the axis of the accumulator spindle.
240 (F0) Head Flying Hours
The attribute can be used to estimate the time it takes to position the head. Allows you to monitor the status of the block of magnetic heads.
254 (FE) Free Fall Event Count
Registers hard disk failures and provides an opportunity to determine their number. If this value is not zero, this is cause for concern, since physical damage to the HDD cannot be ruled out in this case.
⇡#About powering hard drives
Will a normal 220V light bulb work on 230V? Of course it will be. And from 240 V? Same. The question is how long will it last? It is clear that less or significantly less – it depends on the particular light bulb. She is destined for a bright but short life.
Approximately the same situation with hard drives. Naive manufacturers designed them relying on standard +5 V and +12 V. However, in a typical computer power supply (PSU), only the 5 V line stabilizes. What does this lead to?
With a high load on the processor (and modern “stones” consume a lot) and insufficient PSU power, the 5 V line sags, and the stabilization system works out this matter, increasing the voltage to the nominal value. At the same time, the voltage of 12 V also rises (due to the lack of stabilization on it). As a result, the HDD, which is already unstable to heat, also operates at increased voltage, which is supplied to the most heated components – the engine control chip (in the jargon of repairmen – “twist”) and the drive heads (the so-called “voice coil”). Bottom line – see the argument about the light bulb.
A burnt “twist” on the board as a result of increased voltage and poor cooling. Often, the microcircuit literally burns out, with pyrotechnic effects and burnout of tracks on the board. This is beyond repair
Hence the advice on the power supply. The greater its power, the better (within reasonable limits: a reserve of more than 30-35% in relation to real consumption reduces the efficiency of the unit, so you will heat the room). A less powerful, but branded PSU is better than a more powerful, but rootless Chinese one. Remember – not only processors are overclocked. As a first approximation, 420 “Chinese” watts are equivalent to 300 “correct”.
In a good way, one should also take into account the age of the PSU: after 2-3 years of operation, its real power decreases noticeably, and the output voltages drift. Of course, in low-quality products that work on an honest Chinese word, the aging processes are much more pronounced. It’s good if such a block dies quietly on its own, and does not drag away half of the system block in agony!
The maximum allowed is 12.6V (+5% of nominal). However, with increasing voltage, many disks exhibit a non-linearly sharp heating of the above-mentioned nodes – “twists” and “coils”. Therefore, I recommend monitoring the PSU more strictly with an external voltmeter (the sensors on the motherboard that measure the voltage for the BIOS and programs like AIDA can be quite inaccurate).
Capacitors (2200 uF, 25 V) soldered to the HDD power circuit (yellow wire – +12 V, red – +5 V, black – ground). In this case, they reduce the voltage ripple, from which the power supply emits an annoying high-frequency squeak
If the voltage on the 12 V line is too high, and you are not afraid of a soldering iron and are able to distinguish a transistor from a diode, then you can turn the latter into a HDD power cut (I remind you that the 12 V line corresponds to a yellow wire). The diode will play the role of a limiter – “extra” 0.2-0.7 V will drop at its p-n junction (depending on the type of diode), and the disk will feel better. Only the diode must be taken powerful enough to withstand the starting current of 2-3 A.
And without fanaticism: the resulting voltage should not fall below 11.7 V. Otherwise, unstable operation of the disk (multiple restarts) and even data corruption is possible. And some models (in particular, Seagate 7200.10 and 7200.11) may not start at all.
Prediction in Windows PowerShell
Windows PowerShell is a built-in extensible automation tool provided by Microsoft. To predict possible problems with it, you need to perform the following steps:
What to do with S mistakes.
The answer to this question depends on the nature of the problems with the hard drive and the degree of its malfunction.
Stop using the failed HDD
If bad sectors have already appeared on the hard disk, this indicates its significant wear. In fact, it has already begun to crumble, and it is impossible to stop this process. Continued use of such HDD may result in data loss. Since the reason for this is a physical malfunction of the device, it is most likely impossible to restore them.
Recover deleted disk data
Information from the media may also disappear due to logical errors (they may occur when the file system is damaged. In this case, the data lost as a result of a failure must be restored (if they were not overwritten by other data), since there is no physical damage on the hard disk. They can be restored, for example, using the R-Studio program, which allows you to save information even from deleted or formatted partitions
Scan the disk for bad sectors
Scanning can also be performed by a number of third-party applications. The Victoria program is perfect for this. To fully check the entire hard drive for bad sectors, you should select the HDD on the Standard tab, and then go to the Tests tab and click the Start button there. The number of bad sectors found will be displayed during the scan to the right of the blue rectangle labeled “Err”. The numbers next to the red and orange rectangles are still working sectors, but the speed of access to them is very low (a small number of them can even be on a new hard drive). A full scan can take several hours.
Reduce disc temperature
Overheating of a hard drive can have a negative effect on the operation of its mechanical components and electronics. Therefore, when its temperature rises to 55ºC and above, it needs additional cooling. An additional fan can be installed in the computer case to reduce the temperature of the device. There are also special fans designed to cool hard drives. Finally, the temperature of the drive can be slightly lowered by turning off heat-producing devices installed in the PC case that you can do without for some time (for example, a second HDD or a video card if there is an integrated video card on the motherboard).
Defragment your hard drive
The slowdown in the speed of reading and writing to the disk is often due to the high degree of fragmentation of the files stored on it. Strong fragmentation of the file system can contribute to the accelerated wear of the block of magnetic heads. This will lead to additional problems associated with a deterioration in their positioning performance, as well as an increase in the temperature of the drive (since files are broken into fragments, often located at a considerable distance from each other, magnetic heads have to perform additional movements, which increases heat generation).
SSD drives do not need to be defragmented, because they do not have moving platters and heads, unlike HDDs.
To prevent these problems, you should defragment the disk. To do this, go to its properties (by calling the context menu), go to the “Service” tab and click on the “Optimize” button (in Windows 10). Then place the cursor on the desired disk or partition and click on the “Optimize” button in this window. Typically, the optimization procedure takes several minutes.
Get a new hard drive
If the number of bad sectors exceeds the reserve for their reassignment, approaches this indicator or steadily increases, you should take care of buying a new hard drive. After the purchase, you need to install the operating system on it as soon as possible and copy all the information while reading it is still possible.
How to reset S. T error and is it worth it?
Information recorded in S. M. A. R. T. H DD can in principle be deleted. Once all drive data has been reset, its S.M.A.R.T. will look like a brand new drive that hasn’t been used yet. Of course, physical problems will not disappear from this. But it is sometimes useful to take advantage of this opportunity (and not only to unscrupulous sellers of used hard drives), if, for example, the sectors designated as candidates for remap turned out to be physically serviceable, and they received such a status as a result of logical problems with the file system.
This operation can be performed using special applications. One of such programs is DRevitalize (it does not work with some models of hard drives). After launching this utility, select the HDD to be reset and click on the “Start” button. Next, select the item “Features menu and firmware data”, then click on the line “Clear defect reassign list” and confirm the operation. After a few seconds, you can go to “SMART Reset Attribute Values” and click “OK”. If after these manipulations the S. M. A. R. T. update does not occur, you should restart the computer.
⇡#HDD Barriers
The mechanics has long been the Achilles’ heel of HDDs, and not so much because of the sensitivity to shock and vibration (this can still be compensated), but because of the slowness. The fastest “twitching” by a block of magnetic heads (2-3 ms for the best server models) is thousands of times inferior to the speed of electronics.
And you can’t really improve anything here. There is nowhere to raise the rotation speed of the disk pack, 15000 rpm is already the limit. The Japanese a few years ago approached 20,000 rpm (quite a gyroscopic speed), but in the end they refused – the materials do not withstand, the design turns out to be too expensive and poorly suitable for mass production. In small batches, hard drives will come out gold, no one will buy these – these are not gyroscopes, which there is nothing to replace.
It turns out that we hit a barrier. You can’t go around mechanics on a crooked goat. The only way out is to increase the recording density, transverse and longitudinal. Longitudinal density (along the track) affects the performance of the drive, i.e. on the data flow to the rest of the computer nodes. But still, even the achieved 100-130 MB / s is too little for today’s computers. For example, ordinary random access memory (DRAM) has a real performance of about 3 GB / s, and the processor cache is even more. The difference is orders of magnitude, and it greatly affects the overall performance. Of course, no one expects the same performance from a non-volatile storage device with hundreds of times the capacity of DRAM. But even a simple doubling would be noticeable to any user.
The transverse recording density is the density of tracks on a platter, in modern HDDs it exceeds 10,000 per 1 millimeter. It turns out that the track itself has a width of less than 100 nm (by the way, nanotechnology in its purest form). This allows you to dramatically increase the capacity per surface, and also speeds up positioning due to sophisticated algorithms (their development would pull on a couple of doctoral theses).
As a result, HDD capacity and performance have grown significantly in recent years. All this became possible thanks to the technology of perpendicular recording, which has existed for more than 20 years, but only in 2007 matured to mass implementation. Moreover, the capacity then grew even more than required: the first terabyte disks met with a sluggish response from users. The people simply did not understand where to adapt such monsters, especially since at first they were built on five plates, they were capricious, noisy and hot (we are talking about the then Hitachi flagships).
Then, of course, people figured it out, torrents started working at full capacity, and the number of plates decreased. At the same time, the recording density has grown to 500-750 GB per platter (meaning desktop segment drives with a 3.5″ form factor). Terabyte platters are about to go into mass production, which will make it possible to produce hard drives with a capacity of up to 4 TB (more than four platters in a standard 26.1 mm high case cannot fit; Hitachi’s five-platter firstborns have not received much development).
Three-terabyte drive WD Caviar Green WD30EZRX, the most capacious one today. It has a four-plate design and has been produced for exactly one year (since October 20, 2010). As expected, it became cheaper in spring and summer, but in recent days it has risen sharply due to flooding in Thailand (WD assembly plants are located there, and the elements blocked the delivery of components)
By the way, “fast” discs are easy to distinguish even by weight – they are noticeably heavier than “slow” ones. A full-sized jar with thickened walls that contributes to geometric stability and vibration suppression, a high-speed spindle motor, powerful positioner magnets, a two-layer lid of increased rigidity – all this adds tens and hundreds of grams to such a drive. The gap is even greater in server models at 15,000 rpm, where the reduced plates are surrounded by an impressive volume of cast aluminum, and the total weight of the “hard” reaches a kilogram.
WD Raptor high performance 10,000 RPM blade. With a capacity of 150 GB, it weighs 740 g (mass models of the same capacity – 400-500 g). Pay attention to the size of the magnets and the wall thickness
With the cheapening of solid-state SSDs, which are used primarily for the operating system, the need for high-performance HDDs began to decrease, and they themselves gradually stand out in a special market segment (such, for example, is the “black” series from WD). Professional workstations with resource-intensive applications that are critical to access speed are equipped with such disks. Ordinary users are not in a hurry to take rather expensive drives, preferring the amount of performance.
At the other end of the spectrum are popular “green” models with deliberately slow spindle rotation (5400-5900 rpm instead of 7200) and slow positioning of the heads. Cheap, quiet, cool and fairly reliable, these drives are ideal for storing multimedia data in home computers, external cases and network storage. On our shelves, all these Green and LP have strongly pressed other lines, so sometimes you won’t find anything else in small “points”.
Analysis in System Monitor app
Unlike the previously considered applications pre-installed in the system, System Monitor does not work in console mode, but in graphical mode. To assess the state of the hard drive, the user will need:
⇡#Migration from flash
NAND Flash memory appeared much later than HDD, and adopted a number of its technologies – take at least ECC codes. Further, both directions developed in parallel and relatively independently. But recently there has also been a reverse process: the migration of technologies from flash memory to hard drives. Specifically, we are talking about wear leveling.
As you know, any flash chip has a limited resource in terms of the number of erasures-writes in one cell. At some point, it is no longer possible to erase it, and it freezes forever with the last recorded value. Therefore, the controller counts the number of entries in each page and, if exceeded, copies it to a less worn place. In the future, all work is carried out with the new site (this is managed by the translator), and the old page remains as it is and is not used. This technology is called Wear Leveling. So, there is wear and tear in hard drives, but there it is mechanical and thermal. If the magnetic head hangs over one track all the time (for example, this or that file is constantly changing), then the probability of damage to the track increases due to accidental shocks or vibration of the disk (for example, from neighboring drives in the basket). The head can touch the plate and damage the magnetic layer with all the ensuing unfortunate consequences. Even if there is no harmful contact, the fixed head heats up locally and, albeit reversibly, degrades. Writing to this place is less reliable, the probability of subsequent unstable reading increases (and with today’s huge recording densities, any deviation of the parameters is fatal).
These considerations are quite obvious, and the firmware of server disks with SCSI/SAS interface (and they are quite hot) has long since learned to move heads in idle mode so that they do not overheat. But it’s even better to “transfer” information along the platter together with the head – in this case, the described effects are suppressed to the maximum, and the reliability of the drive increases. Here Western Digital has introduced a similar mechanism in the new VelociRaptor models. These are expensive high-performance discs with a 10,000 rpm spindle speed and a five-year warranty, so Wear Leveling is appropriate there.
VelociRaptor outside and inside. A powerful radiator attracts attention. The plates, on the other hand, have a reduced diameter – this is typical for modern high-speed discs.
In addition, the entire VelociRaptor line is aimed at use in highly loaded systems, primarily servers, where disk writes are very intensive and often to the same files (transaction logs are a typical example). Mass “consumer” disks are not threatened by high loads, they also heat up moderately, so such a refinement is unlikely to appear there. However, let’s wait and see.
⇡#The extravagance of magnetic recording
The magnetization of the domains of a hard disk, as in the middle of the twentieth century, is changed using a magnetic head, the field of which is excited by an alternating electric current and acts on the magnetic layer through the gap. Also, this technology requires fast rotation of the plates, precise control of the position of the head, etc. The motor and positioner of the hard drive, as well as the electronics that control them, consume noticeable power, and they cost a lot. But the main thing is that a lot of energy is spent on the very excitation of the magnetic field.
The wastefulness of the standard magnetic recording method is difficult to appreciate when working on a personal computer. Mass-produced hard drives consume less than 10 W even during active operation, which is almost imperceptible against the background of other components (100 W or more). But your views will immediately change after visiting the server room of some large bank, and to get impressions of a lifetime, just walk up to the disk rack of a supercomputer. In the noise of hundreds and thousands of hard drives, blowing fans and precision air conditioners, it becomes clear how much energy is spent on a global scale for such work.
It is not for nothing that energy efficiency in the list of characteristics comes to the fore for data storage systems. Google is already moving its data centers to barges at sea (that’s where the real offshores are!). It turns out that cooling storage systems with outboard water radically reduces operating costs, primarily due to savings on air conditioners.
Viewer S.
A number of HDD manufacturers also release their own utilities designed to receive information from S. M. A. R. T. They are maximally adapted to work with media of certain models. But such software is not designed for all hard drives, and sometimes its capabilities are not enough for a comprehensive assessment of the state of the drive.
Alternatively, you can use one of the many third party software products. Next, we will look at a few well-established applications that provide access to S. M. A. R. T.
CrystalDiskInfo
CrystalDiskInfo is a free application for viewing S. M. A. R. T. parameters and assessing their trend. The utility interface is fully Russified (the language can be switched using the menu). The temperature of the hard drive or solid state drive is shown in the system tray (bottom right of the screen). The program allows you to build a graph that will clearly display the changes that have occurred over the past month with the storage medium. If necessary, the application can be launched with a delay. Using CrystalDiskInfo, it is convenient for the user to change the mode of operation of the hard disk: set the maximum possible speed or turn on the power saving mode (this will also reduce the noise emitted by the HDD). In addition, the developers have implemented support for external HDDs and pockets, as well as Intel RAID arrays.
AIDA64
AIDA64 is a shareware application with a free trial period of 30 days. To be able to work with it further, you need to buy a license.
Victoria
Victoria is one of the best utilities for diagnosing and repairing hard drive problems. There are 2 versions of the program: to run from bootable media and to work directly in the Windows environment. In the latter case, for the application to work correctly, it must be run as an administrator (the corresponding command can be selected from its context menu by right-clicking on the icon). To boot from external media, you must first create a bootable USB drive or CD (DVD) and burn the application image to it.
After Victoria starts, on the “Standard” tab in the right half of the window at the top, select the HDD to be tested and click on the “Passport” button to update information about it. At the very bottom of the window, information about the hard drive model, its capacity in tracks and serial number will be displayed. Then you can go to the “SMART” tab. To read the data, click on the “Get SMART” button in the right section of the window at the top.
With all its other advantages, the program is free. It should also be noted that its latest versions support working with S. M. A. R. T. data from SSD drives.
HDDScan
A distinctive feature of the utility is its extreme ease of use. It is enough to select a hard drive from the “Select Drive” list and press the “S. M.A.R.T.”, and a new window with similar information about the hard disk will appear on the screen. The developers provide the ability to change some of these parameters (AAM, APM, etc.). And you don’t have to pay anything for all this.
Speccy
Using the free Speccy application with Russian language support, you can get information about the components installed in your computer and their technical characteristics. It is possible to save this information as a detailed report.
From the menu on the left side of the screen, select “Data storage”, and in the right side of the application window, information about all hard drives installed on the user’s machine will appear at once. If the information is not immediately displayed on the screen, you must wait a few seconds until the reading is completed.
Predicting disk failure on the command line
There are two ways to check the hard drive for faults using the command line. This is determining the status of the disk and obtaining information about its predicted failure.
Disc status detection
In order to check the S. M. A. R. T. of a hard drive using the command line, you should follow the following sequence of actions:
Predictive failure
In order to predict the probable failure of the hard drive in advance, the user should adhere to the following algorithm:
⇡#Advanced Format and its application
For over 20 years, all hard drives have had the same physical sector size: 512 bytes. This is the minimum amount of writing to disk, which allows you to flexibly manage the distribution of disk space. However, as the volume of HDDs grew, the disadvantages of this approach became more and more apparent – first of all, inefficient use of the capacity of the magnetic plate, as well as high overhead costs when organizing a data stream.
Therefore, high-capacity disks (terabytes and higher) began to be produced using Advanced Format technology, which operates with “long” physical sectors of 4096 bytes. The marking of magnetic platters for AF is very beneficial for the manufacturer: fewer inter-sector gaps, higher usable capacity of the track and the entire platter (and this, along with magnetic heads, is the most expensive HDD component). It was Advanced Format that made it possible to launch inexpensive hard drives on the market, which are now so popular among consumers of audio and video content. AF-disks with a capacity of 1-3 TB are equipped not only with computers, but also with a lot of external drives, network storages and media players.
One of the first 3.5″ discs with Advanced Format released in 2009
But nothing is given for free, new discs are already being brought in for repairs. Looks like the reliability has gone down. After all, a single disk failure or surface defect now corrupts 8 times more user data than usual. With a physical sector of 4 KB and emulation of “short” sectors of 512 bytes, from 1 to 8 sectors will not be read. The operating system reacts to this clearly as: an accident, everything is gone! As a result, a minor problem on the plates grows for the user into a freeze or something even worse.
I don’t think OS, application programs and databases with many small files should be kept on AF disks. So far, their lot is multimedia data that is not critical to dropouts.
⇡#What is worth reading about hard drives
First of all, I recommend checking out the HARDW.net forum. Its “Information Storage” section is visited by many professional repairmen and enthusiasts (almost 40,000 participants). There you can find answers on almost any topic related to the HDD, with the exception of the newest “undiscovered” models. Start with the “Sandbox” subsection: simple (in the understanding of professionals) questions are answered there in detail and meaningfully, and not rejected, as in other places – “take it to the repairman”.
Even more information, however, in English, can be found on the HDDGURU portal. In addition to repair and diagnostic software and articles on specific issues (for example, how to change disk heads), there is an international forum for repairmen, as well as a huge archive of HDD resources (firmware, documentation, photos, etc.). The portal instills a broad view of things, it will be of interest to prepared and motivated people. In any case, in closed conferences of repairmen, references to it run constantly.
I will also refer to my article “How to extend the life of hard drives” in three parts. It gives initial information on handling HDDs, and although it was written more than three years ago, it is not much out of date – drives have not fundamentally changed during this time, except that they have become even less reliable due to fierce savings. Manufacturers, caught in the global crisis, reduced their costs in all areas, which caused a number of high-profile failures in 2008-2009. One of them will be discussed in the continuation of this material, which will be released in the near future.
If you notice an error, select it with the mouse and press CTRL+ENTER.
Scanning difficulties
As a rule, there are no problems when checking hard drives. Scanning is not possible only for old models of hard drives that do not support S. M. A. R. T. technology, or whose self-test is disabled. But there’s nothing to be done here.
What is this technology for
All modern hard drives are equipped with S. M. A. R. T.-unit responsible for tracking and storing information about their main parameters: hard drive heating during operation, rotation speed, positioning time of magnetic heads intended for writing and reading data. Failures that occur during the operation of the drive are also monitored. Instructions on how to do a webcam test.
If bad sectors are found on the disk surface, they are replaced with spare blocks. The use of this technology allows you to predict the failure of the hard drive in a timely manner and take care of replacing it with a working disk device in advance. The user can, without waiting for the final failure of the hard drive, create a backup copy of all files stored on it. In this case, the loss of information can no longer be feared.
