Skip to content

Workaround for aborted smart test for Seagate disk

March 15, 2010

For 2 days, I’ve been trying to run smart tests on my Seagate USB disk. The good news is that smart tests can be run through USB. This was not always the case.

Unfortunately, the long test never finished: smart indicated them as “aborted”:


SMART Self-test log structure revision number 1
Num Test_Description  Status                  Remaining
# 1 Extended offline  Aborted by host         90%
# 2 Short offline     Completed without error 00%

Why aborted ? There was no indication of a problem in the logs, and I certainly did not aborted the tests. Why did the short test work and not the long ? I’ve serarched the net for a solution to no avail.

Then, I connected the dots with the power save feature of this disk. Could the abort be triggered by the power save ?

To check that, I’ve mounted the disk and run this small script during before starting the long tests:

while true ; do echo "foo" >> test.txt; sleep 60; done

And about 4 hours later, voilà:


Num Test_Description Status                  Remaining
# 1 Extended offline Completed without error 00%
# 2 Extended offline Aborted by host         90%
# 3 Short offline    Completed without error 00%

The long test did complete.

This looks like a bug in Seagate firmware. I hope this blog will help people that face a similar problem.

Dod

NB: I’ve slightly edited smart output for clarify

Advertisements

From → computer

7 Comments
  1. thanks for the tip, had the same problem in my NAS and was driving me crazy.

  2. David permalink

    Same problem with a USB Hitachi drive. Except I am trying dd if=/dev/sdb of=/dev/null bs=1K count=1 since I removed the drive from the original system (CentOS has ancient libraries and can’t do smartctrl from a USB disk) and I don’t want to mount it.

  3. Patrick Cronin permalink

    I’ve made it to 80% remaining so far! It looks like its working–thanks for figuring this out!

  4. Greg permalink

    This trick got my seagate drive to finish the extended smart test. Twist: it is a ST4000DM000 – 1F2168 which was removed from a USB 3.0 external case, but is now running as an internal drive. It initially aborted the test by itself after 10%. I wonder if the problem is with seagate drive firmware, some Windows power scheme issue, or something else.

    • On a Linux system, the standby timeout can be tuned with the hdparm command. I can only guess that this timeout is setup on your drive. I’d suggest to use systemrescuecd
      to boot your system on Linux and use the hdparm command from there. Be sure to carefully read the documentation to find out the proper hdparm option and find the correct drive on your system.

      • I got a similar “aborted by host” error under OS X when “Put hard disks to sleep when possible” is checked in the “Energy Saver” options in system preferences. Sounds similar to the sorts of things you’d be changing in hdparm—i.e. the HDD is getting turned off when idle and hence the disk is aborting the surface scan. Turning the above option off and leaving it to run overnight ran the scan correctly for me.\\

        Hope that helps!

  5. G. Trowsdale permalink

    Most discussions of this issue don’t mention or explain why this happens with new USB disks. Thank you for posting the explanation and workaround.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: