Custom Query (1565 matches)

smartctl -l selftest does not show running selftest status with Toshiba MK2002TSKB HDD

To reproduce: 1) Launch selftest: smartctl -t long /dev/sdX 2) Try to check the status after some time: smartctl -l selftest /dev/sdX result: there is no running selftest logged, selftest will appear in log only after completition.

Workaround: Lauch another selftest:

smartctl -t long

Smartctl will show the status of current running selftest:

Can't start self-test without aborting current test (20% remaining)

Platform: CentOS 6.3, 6.4 smartmontools 6.1

SMART Selective self-test log data structure revision number 1

SPAN MIN_LBA MAX_LBA CURRENT_TEST_STATUS

1 314572801 629145601 Self_test_in_progress [10% left] (373644952-373710487)

Says 10% left when it's not even close (314572801-629145601 but it's still at 373710487). Is this some bug in smartd or just the HDD (Western Digital Green 2TB / 3TB, they all show the same behavior) reporting it wrong? Couldn't smartd calculate the correct percentage given it seems to know the current position?

I'm trying to use select tests in smartd.conf as a full extended test just takes too long in one go. I wish I could specify the size directly there so I don't have to worry about it getting confused as to which size/address to use (which seems to happen a lot and doesn't raise an error - I'm already using savestates). Like when I do select tests outside of smartd for other reasons (when there was an I/O error at a specific address). It shouldn't use that as a reference for its own tests...

Seagate is requesting that the failure notice on attribute 190 be removed or somehow be tagged with an 'information only' descriptor. The attribute isn't warrantable and isn't an indicator of drive failure. During customer qualifications, our drives are being failed because they see that this attribute has 'failed in the past'.