On Fri, Mar 05, 2010 at 07:44:32PM -0500, rory_f wrote:
>
>
> Jon LaBadie wrote:
> > On Fri, Mar 05, 2010 at 02:57:01PM -0500, rory_f wrote:
> >
> > >
> > > Hey guys. Thanks
> > >
> > > I checked my library GUI and for the tape run i was just talking about,
> > > where it was stopping around 230gb, it seems compression WAS on. I turned
> > > it off via mt and now the gui says "off" (after stopping the run
> > > obviously)
> > >
> > > I was *sure* i turned it off with mt before. Im doing another amtapetype
> > > now.
> > >
> > > And the comment regarding labeling: thanks. I'll most certainly relabel
> > > all the tapes after this tapetype is done.
> > >
> > > Thanks again guys.
> > >
> > >
> >
> > Recalling your data in the original posting I think an amtapetype run was
> > estimated
> > to take about 8 hours. I think LTO drives are capable of writing a full
> > tape in
> > about 1.5 hrs, amtapetype should do two full writes, thus about 3+ hours
> > total.
> >
> > As you system seems to be writing at less than 1/2 full speed, I wonder if
> > it
> > is failing to stream. If an LTO drive is not streaming, is its capacity
> > affected
> > by having large gaps caused by the start and stopping action?
> >
>
>
> ok, so here is the latest one.
>
> [ama...@backup tor]$ amtapetype -o -f /dev/nst0 -e 400G
> Writing 1024 Mbyte compresseable data: 32 sec
> Writing 1024 Mbyte uncompresseable data: 32 sec
> Estimated time to write 2 * 409600 Mbyte: 25600 sec = 7 h 6 min
> wrote 6815744 32 Kb blocks in 52 files in 5733 seconds (No space left on
> device)
> wrote 6815744 32 Kb blocks in 104 files in 5959 seconds (No space left on
> device)
> define tapetype unknown-tapetype {
> comment "just produced by tapetype prog (hardware compression off)"
> length 212992 mbytes
> filemark 0 kbytes
> speed 37322 kps
> }
>
> compression was set OFF in the gui for the library interfacing with the drive.
>
> so, it seems, unless these tapes i have here are all pooched, the drive is
> giving me problems. ugh.
>
> i'd try it on our other drive but ... that's out being replaced ;-). maybe
> it's time to get this replaced too, while i still have the maintenance
> contract.
>
> thanks for the help.. any further input on this is greatly appreciated!
When the other drive was sent out for maintenance, perhaps the cableing was
changed or disturbed.
With SCSI it is always a good idea to check hardware, card, cables,
terminators, etc.
Try reseating, replacing, changing order, ...
jl
--
Jon H. LaBadie [email protected]
JG Computing
12027 Creekbend Drive (703) 787-0884
Reston, VA 20194 (703) 787-0922 (fax)
