2

When I copy to a tape (LTO-6 for instance) the drive eject and retract the tape multiple times during the copy process. After a while it finishes and the files validate when I do a checksum test after the copy process has finished. But because of this kerfuffle, it takes a lot longer to copy.

I'm using Windows 11, TeraCopy (have tried native copy as well) and I'm using regular HP LTO-6 tapes. I mount the tapes using LTFS-configurator provided by HP. I have also tried to run a cleaning tape through the drive a couple of times as well, but this does not seem to make any difference.

What is causing this and is this normal behavior. Should I be concerned? HPE support is not the best and fastest responder either.

1 Answers1

1

As the tape system seems to have its age (driver support ended 2016), there might be a mechanical problem:

Tape drives probably measures the tension of the tape, i.e. to recognize the end of tape. I could imagine, the grease meanwhile got muddey and stiff, dust in the corners and hairballs wrapped around the axes of moving parts. So the measured tension is not all from tape but from the resistance in the mechanics.

Your normal operation mode might be just at the edge of the tension tolerance and a small deviation might trigger the event "End of Tape", that probably will trigger the eject.

Just running a cleaning tape will clean the magnetic heads, perhaps also eliminate dust in the path, but because of the age that is not enough.

I think, the tape drive has to be opened and get an inspection (cleaning + re-greaceing).

But beware to just take some grease laying around quite handy!
If you're lucky you find some technical service details. Else I would suggest

  • use a grease (no oil)
  • the grease should not be a creeping grease
  • fit for high-speed movement (search i.e. for fast moving ball bearings)
  • have constant lubrication and low resistance for normal and operation temperature.
  • be aware for lifetime lubrication

Or you find a professional technical service.


As a first an easy step, I would try to carefully eliminate any visible dust in the mechanics, especially searching for hair wrapped around the moving parts.

This already might get the necessary drift back to the valid tolerances.
Simply a wrapped hairball might suddenly get hooked during its movement so suddenly raising the resistance and triggering the "tape end, so do eject"-event.

dodrg
  • 1,570