Screen garbage turned out to be a bad DVI cable . Red herring.

darry wrote on 2020-04-19, 19:47:

- After running a few tests, Windows bluescreens with disk write error . - I press the reset button. PC POSTs but SIL3114 contro […]
Show full quote

- After running a few tests, Windows bluescreens with disk write error .
- I press the reset button. PC POSTs but SIL3114 control does not detect the hard disk .
- I power down , wait 30 seconds and power up, but hard is still not detected
- I test the drive in another PC : SMART status OK, disk detectable, SMART history has logged 39 write errors . I get no read errors while running Acronis backup of disk .

I guess there's not much detective work to do here. The hard disk failed writing data, so better replace it before it dies completely. Note that if SMART logged the errors, you can rule out all the rest. SMART is running on the hard disk itself.

derSammler wrote on 2020-04-19, 20:21:

darry wrote on 2020-04-19, 19:47:

- After running a few tests, Windows bluescreens with disk write error . - I press the reset button. PC POSTs but SIL3114 contro […]
Show full quote

- After running a few tests, Windows bluescreens with disk write error .
- I press the reset button. PC POSTs but SIL3114 control does not detect the hard disk .
- I power down , wait 30 seconds and power up, but hard is still not detected
- I test the drive in another PC : SMART status OK, disk detectable, SMART history has logged 39 write errors . I get no read errors while running Acronis backup of disk .

I guess there's not much detective work to do here. The hard disk failed writing data, so better replace it before it dies completely. Note that if SMART logged the errors, you can rule out all the rest. SMART is running on the hard disk itself.

Alas, I feel it is not quite that simple, as the disk is fine in another machine read wise and now apparently fine read/write-wise in its original machine . I will definitely run a SMART extended test to know 100 % and as I have the "bad" LBA addresses in the SMART, I can try dd-ing the block under Linux .
What really gets me is the undetectability of the disk for a while . A (likely) media related write error would not cause that, normally, AFAIK, especially after a hard power cycle . Also shouldn't the pending reallocation count have incremented (which was not the case here )?
Maybe the drive is just being intermittent, but I wish I could rule out a power issue, just in case .

In my experience, when several parts of the PC start acting up at about the same time, the power supply is usually the culprit. Try using a different one when doing your testing. Also, having a backup of the hard drive would be good, just in case 😀

aha2940 wrote on 2020-04-19, 20:55:

In my experience, when several parts of the PC start acting up at about the same time, the power supply is usually the culprit. Try using a different one when doing your testing. Also, having a backup of the hard drive would be good, just in case 😀

Good advice . I already had a backup, but a fresher one never hurts, that's why I imaged it with Acronis TrueImage .

My video card issue was due to flaky cable, so it looks like it's "only" a disk issue . I am running an extended SMART self-test now .
I will probably end up changing the SATA cable, though, because upon closer inspection, they might be interface errors (Gsmartctl calls them Interface CRC errors (ICRC in the SMART output)) :

Complete error log:

SMART Error Log Version: 1
ATA Error Count: 39 (device log contains only the most recent five errors)
CR = Command Register [HEX]
FR = Features Register [HEX]
SC = Sector Count Register [HEX]
SN = Sector Number Register [HEX]
CL = Cylinder Low Register [HEX]
CH = Cylinder High Register [HEX]
DH = Device/Head Register [HEX]
DC = Device Command Register [HEX]
ER = Error register [HEX]
ST = Status register [HEX]
Powered_Up_Time is measured from power on, and printed as
DDd+hh🇲🇲SS.sss where DD=days, hh=hours, mm=minutes,
SS=sec, and sss=millisec. It "wraps" after 49.710 days.

Error 39 occurred at disk power-on lifetime: 7704 hours (321 days + 0 hours)
When the command that caused the error occurred, the device was active or idle.

After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
84 51 01 2e 6c 15 01 Error: ICRC, ABRT 1 sectors at LBA = 0x01156c2e = 18181166

Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
c8 00 01 2e 6c 15 41 00 00:09:49.391 READ DMA
c8 00 01 2e 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2e 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA

Error 38 occurred at disk power-on lifetime: 7704 hours (321 days + 0 hours)
When the command that caused the error occurred, the device was active or idle.

After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
84 51 01 2e 6c 15 01 Error: ICRC, ABRT 1 sectors at LBA = 0x01156c2e = 18181166

Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
c8 00 01 2e 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2e 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA

Error 37 occurred at disk power-on lifetime: 7704 hours (321 days + 0 hours)
When the command that caused the error occurred, the device was active or idle.

After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
84 51 01 2e 6c 15 01 Error: ICRC, ABRT 1 sectors at LBA = 0x01156c2e = 18181166

Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
c8 00 01 2e 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA

Error 36 occurred at disk power-on lifetime: 7704 hours (321 days + 0 hours)
When the command that caused the error occurred, the device was active or idle.

After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
84 51 01 2d 6c 15 01 Error: ICRC, ABRT 1 sectors at LBA = 0x01156c2d = 18181165

Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
c8 00 01 2d 6c 15 41 00 00:09:49.390 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA

Error 35 occurred at disk power-on lifetime: 7704 hours (321 days + 0 hours)
When the command that caused the error occurred, the device was active or idle.

After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
84 51 01 2d 6c 15 01 Error: ICRC, ABRT 1 sectors at LBA = 0x01156c2d = 18181165

Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.389 READ DMA
c8 00 01 2d 6c 15 41 00 00:09:49.388 READ DMA

When SMART attributes shows errors and failures of the hard drive, time to pull the plug on hard drive. Get another hard drive is a requirement.
The SMART monitors the health of the disk surfaces and all parts of the hard drive. Bad power supply rarely affects this but for safety sake, replace that hard drive!

Reliability and data are important.

Cheers,

pentiumspeed wrote on 2020-04-19, 21:51:

When SMART attributes shows errors and failures of the hard drive, time to pull the plug on hard drive. Get another hard drive […]
Show full quote

When SMART attributes shows errors and failures of the hard drive, time to pull the plug on hard drive. Get another hard drive is a requirement.
The SMART monitors the health of the disk surfaces and all parts of the hard drive. Bad power supply rarely affects this but for safety sake, replace that hard drive!

Reliability and data are important.

Cheers,

I agree that reliability and data are important . Just to be clear, I am not clinging for dear life to that 500GB hard drive and am ready to chuck it . What worries me is that the drive may not be the (sole) culprit .

The extended self test ran without a hitch , the "bad" blocks are perfectly readable and writable with dd under Linux and the only "bad" thing in the SMART variables is a reallocated sector count of 5 (which , if memory serves, is the same value as when this drive was pulled from a set-top box PVR put into service in its current home; it is my only current drive with a non zero reallocated sector count) . I would have been really happy if something had gone wrong during testing as that would have confirmed without the shadow of a doubt that the drive is going bad .

To be one the safe side, I will put this drive in my unreliable/for parts bin, put in a new drive and change the SATA cable .

I will also keep an eye out for other issues and replace the PSU if something suspicious crops up that points towards it . It will be fun finding a decent modern replacement that has enough amps on the 5V rail, all Japanese caps (primary in my Enermax is a Hitachi, have not checked the secondaries) and a reasonable price (at least I don't need -5v) .

1smartctl 5.43 2012-06-30 r3573 [i686-w64-mingw32-win8(64)] (sf-5.43-1)
2Copyright (C) 2002-12 by Bruce Allen, http://smartmontools.sourceforge.net
3
4=== START OF INFORMATION SECTION ===
5Device Model:     TOSHIBA DT01ABA050V
6Serial Number:    84F8U7BAS
7LU WWN Device Id: 5 000039 ffcee4838
8Firmware Version: MU1OA720
9User Capacity:    499,570,991,104 bytes [499 GB]
10Sector Sizes:     512 bytes logical, 4096 bytes physical
11Device is:        Not in smartctl database [for details use: -P showall]
12ATA Version is:   8
13ATA Standard is:  ATA-8-ACS revision 4
14Local Time is:    Sun Apr 19 20:06:36 2020 EDT
15SMART support is: Available - device has SMART capability.
16SMART support is: Enabled
17
18=== START OF READ SMART DATA SECTION ===
19SMART overall-health self-assessment test result: PASSED
20
21General SMART Values:
22Offline data collection status:  (0x00)	Offline data collection activity
23					was never started.
24					Auto Offline Data Collection: Disabled.
25Self-test execution status:      (   0)	The previous self-test routine completed
26					without error or no self-test has ever 
27					been run.
28Total time to complete Offline 
29data collection: 		( 4417) seconds.
30Offline data collection
31capabilities: 			 (0x5b) SMART execute Offline immediate.
32					Auto Offline data collection on/off support.
33					Suspend Offline collection upon new
34					command.
35					Offline surface scan supported.
36					Self-test supported.
37					No Conveyance Self-test supported.
38					Selective Self-test supported.
39SMART capabilities:            (0x0003)	Saves SMART data before entering
40					power-saving mode.
41					Supports SMART auto save timer.
42Error logging capability:        (0x01)	Error logging supported.
43					General Purpose Logging supported.
44Short self-test routine 
45recommended polling time: 	 (   1) minutes.
46Extended self-test routine
47recommended polling time: 	 (  74) minutes.
48SCT capabilities: 	       (0x003d)	SCT Status supported.
49					SCT Error Recovery Control supported.
50					SCT Feature Control supported.
51					SCT Data Table supported.
52
53SMART Attributes Data Structure revision number: 16
54Vendor Specific SMART Attributes with Thresholds:
55ID# ATTRIBUTE_NAME          FLAG     VALUE WORST THRESH TYPE      UPDATED  WHEN_FAILED RAW_VALUE
56  1 Raw_Read_Error_Rate     0x000b   100   100   016    Pre-fail  Always       -       0
57  2 Throughput_Performance  0x0005   100   100   054    Pre-fail  Offline      -       0
58  3 Spin_Up_Time            0x0007   132   132   024    Pre-fail  Always       -       157 (Average 167)
59  4 Start_Stop_Count        0x0012   100   100   000    Old_age   Always       -       700
60  5 Reallocated_Sector_Ct   0x0033   100   100   005    Pre-fail  Always       -       0

61  7 Seek_Error_Rate         0x000b   100   100   067    Pre-fail  Always       -       0
62  8 Seek_Time_Performance   0x0005   100   100   020    Pre-fail  Offline      -       0
63  9 Power_On_Hours          0x0012   099   099   000    Old_age   Always       -       7709
64 10 Spin_Retry_Count        0x0013   100   100   060    Pre-fail  Always       -       0
65 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       669
66192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       700
67193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       700
68194 Temperature_Celsius     0x0002   166   166   000    Old_age   Always       -       36 (Min/Max 6/46)
69196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       0
70197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       0
71198 Offline_Uncorrectable   0x0008   100   100   000    Old_age   Offline      -       0
72199 UDMA_CRC_Error_Count    0x000a   200   200   000    Old_age   Always       -       39

Restored backup and replaced drive (new one is 7200 rpm , yay!) and SATA cable and am running 3d Mark 2001 SE in a 20-iteration loop to test video card/CPU, so far so good .

I will keep on using and will post back if anything craps out . Hopefully, it was just a hard drive with a slightly odd intermittent failure mode .

Thanks to all for your valued feedback.