I always trusted drive S.M.A.R.T. attributes and used for years a S.M.A.R.T monitoring tool to tell me if something could go wrong with the HDD before it gets worse.
Until this weekend. My S.M.A.R.T monitoring tool said everything OK but my HDD failed to read a file with a “Could not read from drive” error.
Manual inspection of S.M.A.R.T attributes showed nothing. Cables are well plugged and I did not change anything to the hardware lately.
WD Data Lifeguard quick check succeeds. No errors.
WD Data Lifeguard extended test FAILS !!! with “Too many bad sectors” error !
How is this possible ? If the HDD really has so many bad sectors should S.M.A.R.T not be able to report it ?
Many thanks for helping me understand how S.M.A.R.T can miss this.
Yes good idea, I can test it on another machine if I take it with me to work.
Here are my smart values, do you see something wrong ?
HD Tune: WDC WD10EZEX-08RKKA0 Health
ID Current Worst Threshold Data Status
(01) Raw Read Error Rate 200 200 51 7920 Ok
(03) Spin Up Time 177 173 21 2141 Ok
(04) Start/Stop Count 88 88 0 12071 Ok
(05) Reallocated Sector Count 193 193 140 301 Ok
(07) Seek Error Rate 100 253 0 0 Ok
(09) Power On Hours Count 84 84 0 11849 Ok
(0A) Spin Retry Count 100 100 0 0 Ok
(0B) Calibration Retry Count 100 100 0 0 Ok
(0C) Power Cycle Count 89 89 0 11348 Ok
(C0) Power Off Retract Count 200 200 0 296 Ok
(C1) Load Cycle Count 197 197 0 11774 Ok
(C2) Temperature 109 81 0 34 Ok
(C4) Reallocated Event Count 148 148 0 52 Ok
(C5) Current Pending Sector 200 200 0 131 Ok
(C6) Offline Uncorrectable 200 200 0 0 Ok
(C7) Ultra DMA CRC Error Count 200 200 0 0 Ok
(C8) Write Error Rate 200 200 0 0 Ok
Power On Time : 11849
Health Status : Ok
The drive is old, I doubt WD support can help me.
Anyway I allready buyed another WD blue who should arrive today, so I’ll begin moving the data to the new drive this evening.
What’s worrying is that the drive is not allways reporting errors. I have copied 3 movies to my TV BOX drive. 1 failed to copy, but the two that succeded have errors when playing the video.
I wonder why S.M.A.R.T technology is unable to warn me. The WD blue are so cheap I could have buyed a new one far earlyer if warned.
That’s why I’ll do your test, because one explanation could be that the MB controller is failing and not the drive. But the chance is low because I also have a SSD in that machine and he is performing well without errors.
I did the test on another machine, with same sector errors.
But when I checked the the S.M.A.R.T attributes before the test, and they were updated and showed the bad sectors rellocations and errors ! It was just delayed !
It’s like if S.M.A.R.T is not realtime, but instead the drive takes his time to check itself before updating them, even when he allready encounted read errors during normal work .
This is not what I expected from S.M.A.R;T , because if the drive is failing fast I’ll have to wait for hours for S.M.A.R.T full checkup to terminate and get a warning.
It was also misleading, because when the read error occured I automaticlly considered the drive as safe because S.M;A.R.T was still saying it was Ok and I started searching problems elsewhere with the OS, antivirus, etc … instead of backuping asap.
Some tools (HDDguardian) even have features tu enable “SMART every 4 hours offline scans” … I wonder what they mean by “offline” ?
My conclusion is that S.M.A.R.T is not a realtime warning technology, and more an asynchronous scheduled drive autotest.
