Bug#670333: st0: Sense Key : Medium Error [deferred]

[Jonathan Nieder <jrnieder@gmail.com>]

tags 670333 + unreproducible
quit

Hi Giuseppe,

Giuseppe Sacco wrote:
>> Giuseppe Sacco wrote:

>>> today our backup did not complete successfully because of this error I
>>> found with dmesg:
>>> 
>>> ------------------------------------------------------------------------------
>>> [16186645.419007] st0: Sense Key : Medium Error [deferred] 
>>> [16186645.419014] Info fld=0x2800
>>> [16186645.419016] st0: Add. Sense: Write error
>>> [16186783.801594] INFO: task tar:2574 blocked for more than 120 seconds.
[...]
> A second try completed without errors. So, without ejecting the tape
> after the backup failed, the tape has been rewinded and the backup
> started again and completed without errors.

I can see a few things wrong here:

 - saying "Add. Sense" for the additional sense code is a bit cryptic

 - "Medium Error" (the official name for SCSI sense #3) is also
   cryptic.  My favorite quick reference, Wikipedia, tells me it

	Indicates the command terminated with a non-recovered error
	condition, probably caused by a flaw in the medium or an error
	in the recorded data.

   Perhaps it would be useful to provide documentation somewhere
   memorable fleshing out what that "probably" means.

 - write error during exit trips "task blocked" message.  Avoidable?

Let's keep this open to track that third question.  Any ideas for the
documentation part?

Thanks,
Jonathan