Jochen wrote:
> On  2 Feb, NetBSD list wrote:
> 
> > I've discovered that on both the pmax and vax ports, running on
> > a DS3100 and Vax 4000/60, respectively, HP 35480A DDS DAT drives
> > aren't probed.
> Maybe your drive needs some time to setle after the SCSI bus reset. Try
> to rescan the scsi bus / device with scsictl(8) and / or increase
> SCSI_DELAY in your kernel config and build a new kernel. 

I've just tried setting SCSI_DELAY to 15, as well as a scsi probe
with scsictl(8).  Neither effect recognition.  It looks like it
must simply be a probe ID recognition.  Does anyone know if an
unrecognised ident string produces messages otherwise?

> > st0 at scsibus0 target 7 lun 0: <, , > SCSI1 1/sequential removable
> > st0: rogue, drive empty
> The SCSI version of the TK50 is not very SCSI compatible. This drive
> was never designed to work on a real SCSI bus. It was designed for the
> VS2000 "tape port". This "tape port" is SCSI, you can connect and use
> disks on it, you can even boot from it when you patch the boot ROMs, but
> DEC never intended it for usage as a general purpose SCSI bus.

On the VaxStation 4000/60 (at least), I recall the external connector 
is merely an extension of the internal SCSI cable.
Regardless, the TK50 on this machine does indeed work wonderfully.

On to hacking st.c for identifying the HP drive then.

Thanks,
-scott