Home / ubuntu / Questions / 78600
Accepted
bioShark
Asked: 2011-11-13 11:21:55 +0800 CST

Disk Utility reports slower initial spinup time via S.M.A.R.T.

  • 772

I am not sure this is a real question or a bug I should report Ubuntu.

Using: Ubuntu 11.10, on a Intel Q6600, Samsung Spinpoint F4 2TB.

I have set my PC on Suspend and after I came back, pressed Enter and after logging in everything was back to normal. However, I had a message from Disk Utility that one disk reports errors. I entered Disk Utility, and my Samsung 2TB disk, the one on which my Ubuntu is installed, had the SMART Status turned red, with error message on it. The error was:

Spinup time failed Value 21, Threshold value was 25 (so the error was reported because 21 < 25) I restarted and booted up in Windows to see what HD Tune is reporting. Unfortunately it was exactly the same 21/25.

After reading up on Wiki about SMART and the errors, I discovered that Spinup time is the time required for the disk to reach full spinning speed in milliseconds.

Then it hit me that, in Ubuntu I had Suspended the system, making essentially all my hardware stop. And when I rebooted to Windows, the hardware doesn't really stop, so SMART's reading of the Spinup time was still from Ubuntu's suspension. So I did a full PC stop and then booted up again, both in Ubuntu and Windows to see if there are different readings. Both reported successful Spinup time, 68 (a little better then 21 :) ), although in Disk Utility I have a nice message: Failed in the Past

So now I am pretty sure that Ubuntu didn't handle the Suspend correctly, but then again should I worry about Imminent hardware failure ? Am I missing some drivers? Should I report this as a bug to Ubuntu?

Sorry if this was a bad place to ask this question.

11.10

2 Answers

  1. Best Answer

    These gui tools are neat and all but the only thing that really matters is what the output of the short/extended run tests from the smart tests say. To put it into perspective, a SATA disk from cold plug takes about 14-20 secs to become available, including spinup. I bet that parameter measures the time it takes for the platter to reach a minimum speed so spinup can start. Like you said, an additional 4ms slow down was detected, that's not earth shattering.

    Run the self-tests and see what they say.

    sudo -s
apt-get install libatasmart-bin
sktest /dev/sdX short
watch -d 'skdump /dev/sdX | head -n 20'

    When it's 100% just run skdump again without the stdout throttling. Then do the same thing with the extended test when the short one finishes, that one takes a while.

    I'm not a huge fan of these gui tools, they don't have any heuristics so all the user sees is "failure" with no context of severity.

    For reference: Wikipedia S.M.A.R.T. entry

    03      0x03    Spin-Up Time    
Average time of spindle spin up (from zero RPM to fully operational [millisecs]).
    • 3

  2. 2018 here and I'm still rocking a bunch of these Korean-made Samsung HD204UI 2TB drives, one of the finest drives ever made IMHO. I have about 20 in various computers and RAID arrays, and a good friend has at least 10. We've never had a problem with any of them, and have a few have well over 60K hours on them, and 1000s of power up cycles with no bad or pending sectors.

    Anyway, I recently suspended my Linux Mint 19 machine (coughs Ubuntu), something extremely rare for me. Upon power up/resume, 2 of my 4 installed drives now report SMART errors regarding Spin-Up-Time. (both are drives with very low hours <9000).

    Checking smartctl -a /dev/sdX:

    3 Spin_Up_Time            0x0023   068   021   025    Pre-fail  Always   In_the_past 9889
3 Spin_Up_Time            0x0023   067   021   025    Pre-fail  Always   In_the_past 10272

    Disk utility reported both drives as "Imminent Failure in next 24 hours" or something to that effect, with all 4 drives reporting the polished value of 9-10 seconds for spin up on time.

    Interestingly, I ran a Short Offline Test on both 'failed' drives which completed OK. Checking the SMART log shows the: Short offline -- Completed without error. Interestingly, Disk Utility reports one drive as Disk is OK, one attribute failed in the past while the other drive still shows SELF-TEST FAILED. Repeating the test from the gnome-disk-utility aka 'Disks' seems to have reverted the first

    This is probably a firmware bug with the Samsung F4EG (aka HD204UI), and not likely to ever be fixed, but it's also possible that Ubuntu/Debian/Mint is misusing some sort of low-level SATA interface as has been suggested elsewhere on the nets.

    • 0