Dud tapes

Don Lewis truckman at FreeBSD.org
Fri May 15 22:12:58 UTC 2015


On 15 May, Dan Langille wrote:
> 
>> On May 15, 2015, at 5:07 PM, Don Lewis <truckman at FreeBSD.org> wrote:
>> 
>> On 12 May, Dan Langille wrote:
>> 
>>> The problem occurred today on another tape after writing just 313 MB:
>>> 
>>> May 12 08:28:21 knew kernel: (sa0:sym0:0:1:0): WRITE FILEMARKS(6). CDB: 10 00 00 00 01 00
>>> May 12 08:28:21 knew kernel: (sa0:sym0:0:1:0): CAM status: SCSI Status Error
>>> May 12 08:28:21 knew kernel: (sa0:sym0:0:1:0): SCSI status: Check Condition
>>> May 12 08:28:21 knew kernel: (sa0:sym0:0:1:0): SCSI sense: MEDIUM ERROR asc:c,0 (Write error)
>>> May 12 08:28:21 knew kernel: (sa0:sym0:0:1:0): Command Specific Info: 0x1228
>>> May 12 08:28:21 knew kernel: (sa0:sym0:0:1:0): Error 5, Retries exhausted
>>> May 12 08:28:22 knew kernel: (sa0:sym0:0:1:0): WRITE FILEMARKS(6). CDB: 10 00 00 00 02 00
>>> May 12 08:28:22 knew kernel: (sa0:sym0:0:1:0): CAM status: SCSI Status Error
>>> May 12 08:28:22 knew kernel: (sa0:sym0:0:1:0): SCSI status: Check Condition
>>> May 12 08:28:22 knew kernel: (sa0:sym0:0:1:0): SCSI sense: MEDIUM ERROR asc:c,0 (Write error)
>>> May 12 08:28:22 knew kernel: (sa0:sym0:0:1:0): Command Specific Info: 0x1228
>>> May 12 08:28:22 knew kernel: (sa0:sym0:0:1:0): Error 5, Retries exhausted
>>> 
>>> The tape before this took 111 GB, the tape after, took 791 MB (and then the backup completed).
>>> 
>>> What do these errors mean? Does it indicate a dead tape?
>> 
>> Bad tape, dirty or defective head, some sort of mechanical tape handling
>> problem that affects the tape tension or movement, or less likely some
>> other drive problem.  MEDIUM ERROR means that the drive couldn't write
>> the bits to the tape and read them back correctly.
> 
> I have a bunch of new-to-me tapes.  So far, 5 out of 15 have resulted in that error.

Try some from another source, preferably a different brand.  If you see
the same, then it is likely a drive problem.  At a previous job, we used
HP DDS drives for backups. We got refurbed drives whenever we sent one
in for a warranty repair. The would work for a while (a few months) and
then cr*p out like what you are seeing.

Another test is to write data repeatedly to the same tape.  If it cr*ps
out at about the same place every time, then it is likely to be a bad
spot on the tape.




More information about the freebsd-scsi mailing list