[smartmontools-support] I can't complete the long test

[Carlos E. R.]

On 07/05/2021 07.10, Nathan Stratton Treadway wrote:
> On Fri, Apr 23, 2021 at 11:56:34 +0200, Carlos E. R. wrote:
>> I have another identical disk and it suffers the same problem. But
>> other two external disk, half the size, don't exhibit the problem.
>>
>> It says "aborted by host". How, why? What can I do to impede this
>> abortion? The machine runs 24/7 and that disk is currently not mounted.
> 
> Note that in this context (a message worded from the SATA drive's point
> of view), "host" is not your computer but rather the SATA controller
> built in to the USB enclosure.

Ah.

> 
> Some enclosures have an automatic inactivity timeout hard-coded into
> their firmware, separate from any Linux-level timeout -- and since no
> traffic is passing between the computer and the drive while the
> self-test is underway, that period of time gets considered "inactive" by
> the firmware and it proceeds to suspend the drive, mid-test.

Ah... Matches.

> 
> You may be able to find more detailed information on your particular
> enclosures' behaviors based on their USB ids (e.g. the output of
> "lsusb").
> 
> In the past I have indeed been successful working around this problem by
> using a loop to periodically do "activity" on the drive while the
> self-test was running.  Did you ever get that to work on your drives?

Yes, it works. I forgot to post the analysis I did on this. Will do this 
morning.

-- 
Cheers / Saludos,

		Carlos E. R.
		(from 15.2 x86_64 at Telcontar)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 203 bytes
Desc: OpenPGP digital signature
URL: <https://listi.jpberlin.de/pipermail/smartmontools-support/attachments/20210507/2ef6ca10/attachment-0001.sig>