On Wed, 2002-01-16 at 17:53, Ethan Dicks wrote:
>
> --- Michael Schneider <ms_at_silke.rt.schwaben.de> wrote:
> > See below...
>
> > > Do you have access to another VAX? I used to cut console tapes all the
> > > time for our 11/730 (k-panda when it was in the UUCP maps) - I still
> > > have it (and the tapes).
> >
> > Yes, but only newer ones, nothing with a tu58.
>
> OK... if you had a newer VAX with the right version on it, you could,
> theoretically, get the files you need from that new VAX and run a
> serial line to that emulator PC (at a slower baud rate, probably) and
> "write" the emulated tape image to your emulator box from the working
> VAX, then move the cable, up the speed to 38400, and have the 11/730
> feed off of that. The rub is that I don't recall how much of the console
> data files are available in the regular SYS$SYSTEM directory structure
> and how many I may or may not have pulled from a working tape. The
> boot files are just text, short and easy to recreate, even with EDT.
> The binary files are the problem.
Especially the file(s) with the CPU's microcode. They are definitively
not part of a standard VMS installation. The VMS tape, i mean.
Otherwise, your method would work, i think.
>
> The reason for using the other VAX is to use it to easily create the
> filesystem on the emulated tape. If you had a way to read the tape
> that you do have, you could just put the image on the PC, run the
> TU-58 emulator and go.
How true! You see the Catch-22 i'm in? If i had a working tape, i could
make me a working tape-image, so i would not need a working tape
anymore.
>
> > > The tape from DEC was sub-optimized...
> >
[snip]
>
> > an was cut Jun21/84.
>
> Somewhat old - probably the very original tape that shipped with
> the CPU.
I think so. The machine comes from a measurement installation of one of
our local gouvernment agencies. They validated beer-keg-volumes
(obviously a very important thing to do in Germany 8-).
So, it was just set up, fired up and left running 'till it has been
replaced in the '90s.
>
> > Unfortunately i do not know the exact VMS version installed on my
> > machine, because the only time got it booting i was so exited that
> > i spoiled my notes. Very un-scientific, i know.
>
> Ah, well.
>
> > But i think it is VMS 3.2, IIRC. Or 3.6, that's the label on my
> > "standalon backup"-tape.
>
> Hmm... my memory of those days is hazy... I played with 3.6, but that
> was as a novice user on an 11/750. We did get an early model 11/730
> (order the week they were announced) - that's the one I have in storage.
> By the time I touched it, it was running VMS 4.x. I installed Ultrix 1.1
> on it at one point (and hooked it to Usenet before the Great Renaming,
> but that's another story), and it spent its last years as a linker box
> for our product, running 5.x (we fed the objects and asm source to it
> via 56K DECnet link, let it chew on the files, then transferred the
> binaries back to our VMS 4.6 general-use 11/750 and put them in the tape
> build area for making distribution tapes)
Well, the bean-counters obviously never upgraded VMS on this one, so, my
guess is that the version of VMS installed is the one that has been
deliverede back then.
>
> I think all I am going to have for it is VMS 4.x and VMS 5.x console
> media. I am fairly certain I had my customized tape for booting VMS 4.4
> in my hands on Sunday. I know exactly where it is. If that will work
> for you, I may be able to help. I honestly do not recall how sensitive
> the machine is to console tape version vs VMS version mismatch. I know
> we didn't have to change the tapes on minor revs, but I do not recall
> if we were forced to change for major revs.
This, i don't know either. I would have to try this.
[snip]
regards
ms
--
Michael Schneider email: ms_at_vaxcluster.de
Germany http://www.vaxcluster.de
People disagree with me. I just ignore them.
(Linus Torvalds)
Received on Wed Jan 16 2002 - 15:29:35 GMT