0xF8 by Extended Self-Test

[registus]

Hi,

While doing Extended Self-Test I alwayse receive 0xF8 error. I already read the discussion what it means. The steps I already did to solve the issue:

1. Reinitialise disk surface - at least 3 times - sequential test, sequential backward test with surface initialization level 5, etc.
2. ReadWriteRead test
3. Read only tests (also backwards)

Firstly there was 98 % health with 2 weaks sectors, then it dropped to 95%, after sequential backward test with level 5 it came to 100 %, but Extended Self-Test showed 0xF8 error after 20 minutes. Now I tried to do the Read test again and it showed one weak sector and one bad. In previous Read (Read Write Read) tests I usually received 2 bad sectors (now first time I see one weak and one bad). It dropped again to 98 %.

Hard Disk Model ID . . . . . . . . . . . . . . . : WDC WD3200BEVT-80A0RT1
Firmware Revision . . . . . . . . . . . . . . . : 01.01A01
Total Size . . . . . . . . . . . . . . . . . . . : 305242 MB
Power State . . . . . . . . . . . . . . . . . . : Active
Logical Drive(s) . . . . . . . . . . . . . . . . : G: []
Current Temperature . . . . . . . . . . . . . . : 40 °C
Maximum Temperature (Ever Measured) . . . . . . : 43 °C, 27. 4. 2015 16:10:27
Minimum Temperature (Ever Measured) . . . . . . : 23 °C, 29. 4. 2015 8:20:22
Maximum Temperature (During Entire Lifespan) . . : 52 °C
Power On Time . . . . . . . . . . . . . . . . . : 178 days, 7 hours
Estimated Remaining Lifetime . . . . . . . . . . : more than 1000 days
Health . . . . . . . . . . . . . . . . . . . . . : #################### 100 % (Excellent)
Performance . . . . . . . . . . . . . . . . . . : #################### 100 % (Excellent)

The hard disk status is PERFECT. Problematic or weak sectors were not found and there are no spin up or data transfer errors.
No actions needed.

After running last Read test the status changed to:

There are 3 weak sectors found on the disk surface. They may be remapped any time in the later use of the disk.
At this point, warranty replacement of the disk is not yet possible, only if the health drops further.
It is recommended to examine the log of the disk regularly. All new problems found will be logged there.

No actions needed.

Read test showed 1 damaged sector and 1 bad sector, but the status is about 3 weak sectors.


ATA Information
-----------------
Hard Disk Cylinders . . . . . . . . . . . . . . : 5814021
Hard Disk Heads . . . . . . . . . . . . . . . . : 16
Hard Disk Sectors . . . . . . . . . . . . . . . : 63
ATA Revision . . . . . . . . . . . . . . . . . . : ATA8-ACS
Transport Version . . . . . . . . . . . . . . . : SATA Rev 2.6
Total Sectors . . . . . . . . . . . . . . . . . : 732566646
Bytes Per Sector . . . . . . . . . . . . . . . . : 4096 [Advanced Format]
Multiple Sectors . . . . . . . . . . . . . . . . : 16
Error Correction Bytes . . . . . . . . . . . . . : 0
Unformatted Capacity . . . . . . . . . . . . . . : 2861588 MB
Maximum PIO Mode . . . . . . . . . . . . . . . . : 4
Maximum Multiword DMA Mode . . . . . . . . . . . : 2
Maximum UDMA Mode . . . . . . . . . . . . . . . : 6 Gbps (6)
Active UDMA Mode . . . . . . . . . . . . . . . . : 6 Gbps (5)
Minimum Multiword DMA Transfer Time . . . . . . : 120 ns
Recommended Multiword DMA Transfer Time . . . . : 120 ns
Minimum PIO Transfer Time Without IORDY . . . . : 120 ns
Minimum PIO Transfer Time With IORDY . . . . . . : 120 ns
ATA Control Byte . . . . . . . . . . . . . . . . : Valid
ATA Checksum Value . . . . . . . . . . . . . . . : Valid


S.M.A.R.T. Details
--------------------
Off-line Data Collection Status . . . . . . . . : Never Started
Self Test Execution Status . . . . . . . . . . . : Test Failed By Read Element
Total Time To Complete Off-line Data Collection : 9900 seconds
Execute Off-line Immediate . . . . . . . . . . . : Supported
Abort/restart Off-line By Host . . . . . . . . . : Not supported
Off-line Read Scanning . . . . . . . . . . . . . : Supported
Short Self-test . . . . . . . . . . . . . . . . : Supported
Extended Self-test . . . . . . . . . . . . . . . : Supported
Conveyance Self-test . . . . . . . . . . . . . . : Supported
Selective Self-Test . . . . . . . . . . . . . . : Supported
Save Data Before/After Power Saving Mode . . . . : Supported
Enable/Disable Attribute Autosave . . . . . . . : Supported
Error Logging Capability . . . . . . . . . . . . : Supported
Short Self-test Estimated Time . . . . . . . . . : 2 minutes
Extended Self-test Estimated Time . . . . . . . : 117 minutes
Conveyance Self-test Estimated Time . . . . . . : 5 minutes
Last Short Self-test Result . . . . . . . . . . : Test Failed By Read Element
Last Short Self-test Date . . . . . . . . . . . : 30. 4. 2015 15:56:13
Last Short Self-test Duration . . . . . . . . . : 2 minutes
Last Extended Self-test Result . . . . . . . . . : Test Failed By Read Element
Last Extended Self-test Date . . . . . . . . . . : 30. 4. 2015 16:18:02
Last Extended Self-test Duration . . . . . . . . : 20 minutes
Last Conveyance Self-test Result . . . . . . . . : Never Started
Last Conveyance Self-test Date . . . . . . . . . : Never Started

As you can see, both Short and Extended Self-Test failed.

How can I get the Self-test passed? I took a look at http://www.hdsentinel.com/hard_disk_cas ... ectors.php

According to the article it seems that Reinitialise disk surface could help to solve it, but it didn't in my case even if I tried to launch it in different ways.

chkdsk showed 0 bad sectors even if I ran full chkdsk:

CHKDSK is verifying files (stage 1 of 3)...
256 file records processed.
File verification completed.
0 large file records processed.
0 bad file records processed.
0 EA records processed.
0 reparse records processed.
CHKDSK is verifying indexes (stage 2 of 3)...
276 index entries processed.
Index verification completed.
0 unindexed files scanned.
0 unindexed files recovered.
CHKDSK is verifying security descriptors (stage 3 of 3)...
256 file SDs/SIDs processed.
Security descriptor verification completed.
10 data files processed.
Windows has checked the file system and found no problems.

312568831 KB total disk space.
3093568 KB in 6 files.
12 KB in 12 indexes.
0 KB in bad sectors.
75747 KB in use by the system.
65536 KB occupied by the log file.
309399504 KB available on disk.

4096 bytes in each allocation unit.
78142207 total allocation units on disk.
77349876 allocation units available on disk.


The drive is empty actually.

I have registered Pro version of HD Sentinel.

Thanks for any helpful information .

Peter

[hdsentinel]

> While doing Extended Self-Test I alwayse receive 0xF8 error.

I must tell that response: 0xF8 is not error, this is completely normal if you see the self test window like this:



This is exactly why th test window says "Running, no problems found". Generally, the response code (returned by the hard disk drive) can show
- 00: if the test is successfully completed
- F9...F1: if the test is currently running and the digit after the F shows progress (remaining of the test) in 10% steps. For example 0xF8 simply shows that the test is in progress and 80% of the test is still remaining.
- other values are interpreted by Hard Disk Sentinel and shows the failed element which caused the real error.


> Firstly there was 98 % health with 2 weaks sectors, then it dropped to 95%, after sequential backward test with level 5 it came to 100 %,

Yes, this is completely expected: the FAQ page about the "weak sectors" ( http://www.hdsentinel.com/hard_disk_cas ... ectors.php ) shows that the Disk menu -> Surface test -> Reinitialise disk surface test should correct such problems and even increase the health back to 100% - if there are no other issues.

However, such weak sectors are usually not really related to hard disk problems but caused by some other factors, like
- improper shut down / power off
- power loss / power failure / sudden reset
- insufficient power provided by the power supply
- data or power cable issues / connection problems
- general overheating (not only the hard disk but the chipset or hard disk controller)
- overclocking, RAM issues
- general instability of the system

as these all can cause problems in the operating of the hard disk (the weak sector is a very common result of these).
And if these not fixed, then (even Hard Disk Sentinel fixed the problems and improved the situation) weak sectors will appear again and again.


> but Extended Self-Test showed 0xF8 error after 20 minutes.

Yes, it is completely normal: the Extended Self-Test may need very long time (even hours or in worst case, if the hard disk is heavily used during the test, then days), so it is normal that after 20 minutes the progress changes from 0xF9 to 0xF8.

But it is not error, if there is an error, Hard Disk Sentinel displays in the test window with red, see the above mentioned page about weak sectors ( http://www.hdsentinel.com/hard_disk_cas ... ectors.php ) which clearly shows the example: usually when at least one weak sector present, the Short self test / Extended self test functions usually fail - but after the status is fixed by Hard Disk Sentinel (by the Reinitialise Disk Surface test), then usually the Extended self test completes with no problems (just it may take long time).


> Now I tried to do the Read test again and it showed one weak sector and one bad.
> In previous Read (Read Write Read) tests I usually received 2 bad sectors (now first time I see one weak and one bad). It dropped again to 98 %.

Yes of course. If the original cause of the weak sector is not yet fixed, then new and new such problems will be reported.
If you (by the small floppy button in the upper right corner of the disk surface test window) save the disk surface map after the Disk menu -> Read test (which showed the problematic red/yellow blocks) and then, after fixing (when the Reinitialise Disk Surface test restored the status to perfect) - and after new problems - you compare the new disk surface map with the previous one, you may notice that the problem is on a completely different location of the hard disk surface map. This indicates that there is no problem with a particular sector, but the issue may happen at random position (which was just read/written by the hard disk and turned to weak).

> After running last Read test the status changed to:
> There are 3 weak sectors found on the disk surface.
> Read test showed 1 damaged sector and 1 bad sector, but the status is about 3 weak sectors.

Yes, this is what I mean: if the original cause of the weak sectors (eg. cables / connections / power etc...) not fixed, then new and new problems detected with time (usually quickly).

Note that each blocks in the test contains several sectors, so it is possible that many sectors damaged/bad/inaccessible in the same block.
The log on the bottom (just below the surface map) contains detailed information about each problematic sectors, so there it is possible to check the real position/location of such sectors.


> As you can see, both Short and Extended Self-Test failed.

Yes: if the status is not perfect (even on 1 weak sector - or even if there is something which causes weak sectors) these tests may fail, usually quickly, but it is possible that only after longer time.

> How can I get the Self-test passed? I took a look at http://www.hdsentinel.com/hard_disk_cas ... ectors.php

That page describes what are the weak sectors in general, how they may appear and how they can be fixed.
But in the "Not stable hard disk status" section, it shows that
"... It is recommended to check the items listed above, replace data / power cables, verify the power supply. If this does not happen, we may expect further weak sectors to be detected during the later use of the hard disk. ..."

> According to the article it seems that Reinitialise disk surface could help to solve it, but it didn't in my case even if I tried to launch it in different ways.

The Reinitialise disk surface helped even in this situation - as fixed / stabilized (and improved) the status as should.
But this may be not enough alone.


> chkdsk showed 0 bad sectors even if I ran full chkdsk:

Excuse me, but that "weak sectors" page clearly shows that chkdsk is absolutely not usable to diagnose / repair such problems, as it may report problems - but does not fix - even making more problems by marking the sectors as bad instead of real fixing.

Chkdsk is designed to diagnose and fix the logical partition (volume) only. Even if this sounds surprising, issues related to the logical volume is completely independent from the problems related to the physical hard disk, as chkdsk may
- show problems (related to the logical volume) on a perfect hard disk
- show no problems even on an almost dead hard disk

Personally I'd try to connect the hard disk to a competely different operating environment, for example to a completely different PC (different cables, motherboard, power supply) or try with an external enclosure / adapter / docking station (which gets power from different source, not from this computer).

I'd try the tests as described at: http://www.hdsentinel.com/faq.php#tests ( Hard disk health is low or recently changed or I just installed a new (used) hard disk. How can I perform a deep analysis? ) to diagnose the situation. If the first short self test / extended self test functions fail (if the status is not yet fixed, there is high chance for that) then I'd try the Disk menu -> Surface test -> Read test, save the resulting disk image (especially if one or more blocks showed with yellow/red colors, indicating problems there) and then I'd try Disk menu -> Surface test -> Reinitialise Disk Surface test to fix it, and then re-start the others, to verify if the status is now better (stable) but if not, confirm if the problematic sectors are in the same area of the hard disk - which will suggest that a particular hard disk area is really damaged and unstable.

Personally I'd be more than happy to follow how the status changes, so if you can use Report menu -> Send test report to developer option (even many fimes, for example before / after the test steps) and send the saved disk surface map images, I can also check and verify with more details, track how the status changes.

[registus]

I received these 2 errors - Test Failed by Read Element:

Short

short.jpg (39.96 KiB) Viewed 18490 times

Extended

extended.jpg (40.32 KiB) Viewed 18490 times

The "best" result was - Short test with 0xF1. The "best" extended was stopped at 0xF8.

Now I turned the computer back on after 3 days. I don't expect influence on the result, but there are slightly different when it was last time, because extended was stopped after 20 minutes 0xF8 with the same error message (Test Failed by Read Element).

I use actual SATA cable for this disk also for different drives during last weeks (I'm just doing health care procedure of my drives ) and by other drives I always get 100 %, but I will try to change the cable and if I will have time different computer to test, too). By the way the computer is in a good condition and all other drives connected work properly even all other hardware staff. But there still may be some hidden errors I don't see yet.


That drive was not used at all, it is completely empty drive and I don't need to work with it during the test, but it dropped by Extended always after max. 20 minutes (today it was only 30 seconds).


> usually when at least one weak sector present, the Short self test / Extended self test functions usually fail - but after the status is fixed by Hard Disk Sentinel (by the Reinitialise Disk Surface test), then usually the Extended self test completes with no problems (just it may take long time).

This is what I can't get for. I tried Short & Extended - I did more Initialize disk procedures, get back to Short & Extended and received "Test Failed by Read Element" error again.



I will try to save the position of weak sectors and do more tests also with noting weak sectors on the bottom bar. It was maximum of 3 weak / damaged sectors at the time I got with this disk. Optically it always seemed to by on the same position. I know that one block can mean different sector, but optically it looked to be at the same place.


I will follow your recommended steps and also create and send test reports (for this time with the same hardware). Hopefully it will help you to see the issue.

Thanks for your answer. I appreciate that.

Peter

[hdsentinel]

Hi Peter,

Thanks for the reports and the information !

Yes, from the reports I saw that the short/extended self tests failed (when there were weak sectors reported) - and yes, then I saw that after the Reinitialise Disk Surface test, the weak sectors repaired and the health improved back.

Then you may try to perform a Disk menu -> Surface test -> Read test as well - and inspect if that will also show any problem (red or yellow block) or at least minor performance degradation (dark green block) on the same place where you previously saw issue?

If so, then yes, maybe one (or more) sectors in that area are not really stable and even the repair done by the Reiniialise Disk Surface test fixed only temporarily. Not sure about this location, but if it is near the beginning or end of the hard disk surface, it may be a good idea to make data partition where this block will not used at all (eg. before or after this problematic area). Then (with smaller capacity) the hard disk may be used, for example for secondary storage.

Also if possible, I'd try the tests in different environment, eg. USB adapter/docking station or in a different computer, just to be sure - to verify if the same happens there.

[registus]

Thanks for your post.

I was testing the disk several times and the result (with the same computer) is as follow

There were 5 sectors (some of them marked 3 times) marked with Error 12 (cyclic redandancy check)

After running Reinitialise Disk Surface test for about 3-4 times 4 of them disappeared. 1 sector (marked 3 times) is still there and another 4 Reinitialise Disk Surface tests did not help to solve the issue. This is probably the cause why are Short and Extended Self Test failing.

The health is being increased after Reinitialise Disk Surface test to 100%, but after Read test as bad sector is found health is being dropped to 99%

Is it possible to deny this block with HD Sentinel, so it can't be used anymore? It is in sector 2502, app. the quarter of the drive.

I will try to do tests in different environment, I did not have time to test it so far.

Thanks

[hdsentinel]

Thanks for the mails with the details about the tests and the results.

I investigated and yes, I see that by the Reinitialise Disk Surface test, the hard disk repaired, all sectors corrected - but some sectors in block 2502 again reports problems: 156461358, 156462744, 156464122

Seems the repeated Reinitialise Disk Surface tests corrected the first two of them, but the last one (even corrected by the Reinitialise Disk Surface test) always report problems on the following read test (or when the short / extended self test used).


> Is it possible to deny this block with HD Sentinel, so it can't be used anymore?

Generally, the Reinitialise Disk Surface is designed exactly for this purpose: to force the hard disk to examine the status of the sectors and reallocate if required (to prevent further use by redirecting all further reads/writes to the spare area) or repair the sectors.
Seems in this case that one sector always repaired, but not reallocated (not replaced with a spare sector), so that sector always re-used.

The best would be to partition this hard disk in a way to make sure that the corresponding sector (and neighbour sectors, to be sure) should never be used for storing data. So I'd suggest to

- move the mouse pointer over the disk surface, on block 2502 which had the problematic sector
- inspect the MB: xxxxx displayed on the bottom which shows the position of that block (in MBytes) on the hard disk surface.
- in Windows Disk Management, create a partition which starts on the beginning of this hard disk and lasts until that MByte position (or maybe 100 MBytes smaller).
- Then create a second small partition (eg. 200 MBytes or so) which should cover the problematic area.
- Then create a 3rd partition on the rest of the hard disk.

Then the second (small) parition can be deleted and the other two partitions (covering the areas before and after the problematic area) can be used for data storage, as then for normal reads/writes the sector will be never used.

This may reduce the usable capacity a bit - but you can be sure that the sectors previously reported with problems will not contain any data and those will not be read/written by the hard disk during normal operation.

[registus]

Hi,

Thanks for your advices.

I already did the tests with completely different PC environment (hardware + software) and results are as follow:

1. First test showed 23 Data error in one sector 156462040 and three 23 Data errors in sector 156464040 (both are blocks 2502)

2. After Reinitialise Disk Surface test I did another Read test with three 23 Data errors in sector 156464040 .. Bad sector 156462040 disappeared.

3. After four days another Read test the result was the same like it was in the first test

4. Reinitialise Disk Surface fixed 156462040 again

5. After another 5 days new Read test again, but now with three 23 Data errors in sector 155856840, three 23 Data errors in sector 156462040 and three 23 Data errors in sector 156464040

6. Reinitialise Disk Surface now fixed 156464040 and new Read test show only three 23 Data errors for 156462040.


Interesting is, that HDS showed different sectors in different PC environment. It was the same block, but the sectors were different. Is there any answer why?



The most bad sector in first PC environment was 156464122 (block 2502)

The other were 155856840 (block 2492) 156461358 (block 2502) 156462744 (block 2502)



So in the first PC environment sector no. 156464122 was the bad one, which was not fixed after Reinitialise Disk Surface test.
In second PC environment sector no. 156464040 was not fixed after Reinitialise Disk Surface test.

I would tell that after Reinitialise Disku Surface some bad sectors were completely fixed, some were fixed, but after a while appeared again. I wasn't do anything with that drive (copying files or something like that), only do the tests during last 2 months.

I will have to format the drive and miss the problematic parts according to your suggestion.

If you want, I can do the tests in new evnironment again and send results to you after each test.

Thanks.

Peter

[hdsentinel]

Thanks for the information about the results.

Yes, it is interesting that the different tests in the different environments showed problems with different sectors.

Did you use the same Hard Disk Sentinel version in different hardware + software environments (just to be sure)?
Did you use USB / eSATA enclosure, special RAID controller or so in one of these environments? These may translate the sector addresses differently which can cause that the actual problem on the disk surface reported in diffeent sectors.

Generally, if the sector would be reallocated (replaced by the spare area), then the original sector number (which showed the problem previously) should not show problems - as then the spare area is used instead of the original. Then yes, it is possible that a different sector (usually neighbour sector or a sector near the original problem) will fail, depending on the original problem and health status.

In some cases even we expect reallocation (even if using the Reinitialise Disk Surface test in Hard Disk Sentinel) the hard disk does not perform reallocation, so the original sector will be re-used. If this happens, the number of "bad sectors" in the text description (if any reported) do not change, as no new reallocation happens.
If this happens but the sector is not perfect, then yes, it may fail again with time, for example in a new read test started. Usually not immediately (otherwise the Reinitialise Disk Surface test would show it) just after some time, for example some days or weeks.

Yes, in this situation, the best way is to make partition(s) to not use that problematic sector(s) / area.
Then these sector(s) will be no longer used, the remaining hard disk capacity can still usable.

[registus]

Hi,

Sorry for late reply, but at least did testings few months later. At first, after Read test it showed 2 bad blocks - 2493 and 2502 and 3 bad sectors (each sector displayed 3 times), after Reinitialise Disk Surface test it seemed, to be ok, so I did Read Write Read test again and it showed block 2492 and 2493 as bad (2 bad sectors). Then I tried to do Read test again (so no Reinitialise test anymore) from block 2480 to 2520 and it showed 3 blocks as bad - 2492, 2493 and 2502 (3 bad sectors). The same sectors as in first two tests are involved, but completely different like it was few months before (but still same blocks)

Answers to your questions:

I used the same HDS version by both environments.
I used SATA connection directly to mobo.

I realized your suggestions about formatting. Thank you for the advice and for quick responses and for your software. Keep it up.

All the best.

Peter, SK