On 2013-05-17 21:13, Cory Smelosky wrote:
On Fri, 17 May 2013, Bob Armstrong wrote:
Really? Not from tape? Interesting design choice...
Old VAXes booted standalone BACKUP from the console media (TU58 or
RX01).
SA BACKUP then could talk to tape drives just fine. It's all you need to
install VMS, so that was all they did.
Ah right. Makes more sense how.
How were the VAX BSDs installed then? Standalone bootloader on a floppy?
The one time I did it, I actually hacked the boot loader into memory from the console, and started from there. The boot loader was listed in the manual.
But (as I mentioned in another mail), this came from MtXinu. Not sure how other BSDs did it, as I never had an original distribution from any other.
But once you had the system installed, you also created a console media which could boot the thing, as VMB was not able to boot a Unix system anyway, even from disk.
Johnny
On Fri, 17 May 2013, Johnny Billquist wrote:
On 2013-05-17 21:00, Cory Smelosky wrote:
On Fri, 17 May 2013, Johnny Billquist wrote:
Unibus VAXen basically means VAX-11 machines. They booted either from
VMB on console media, or (for the 11/750) from a boot block. No network
capabilities there. They could not even boot from tape.
Johnny
Really? Not from tape? Interesting design choice...
That's one way of putting it. Another would be "annoying".
Yeah. VMS distribution did come on tape, but you also got a floppy or
TU58 from which you booted the standalone backup on the VAX, and then
did the installation using that.
Now, if you instead wanted to install something like Unix from a tape,
you had to enter an initial bootstrap by hand into memory. At least the
MtXinu distribution actually have the bootstrap codes for all VAX-11
systems listed in the manual.
(Quite helpful when I was installing 4.3 Reno from tape, as it turned
out those bootstraps worked for my Reno tape as well, on an VAX 8650.)
That does sound rather helpful. I like useful manuals. ;)
Johnny
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
On Fri, 17 May 2013, Bob Armstrong wrote:
Unibus VAXen basically means VAX-11 machines.
They booted either from VMB on console media, or
(for the 11/750) from a boot block.
TO be fair, the limitation on booting was mostly just that nobody bothered
to implement support for any other device in VMB. It wasn't any kind of
hardware limitation.
Later versions of VMB (e.g. in a MicroVAX) could boot directly from
various tape devices.
Really? Not from tape? Interesting design choice...
Old VAXes booted standalone BACKUP from the console media (TU58 or RX01).
SA BACKUP then could talk to tape drives just fine. It's all you need to
install VMS, so that was all they did.
Ah right. Makes more sense how.
How were the VAX BSDs installed then? Standalone bootloader on a floppy?
Bob
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
On 2013-05-17 21:08, Bob Armstrong wrote:
Unibus VAXen basically means VAX-11 machines.
They booted either from VMB on console media, or
(for the 11/750) from a boot block.
TO be fair, the limitation on booting was mostly just that nobody bothered
to implement support for any other device in VMB. It wasn't any kind of
hardware limitation.
True. As long as you actually even used VMB to boot, which only VMS did back then. VMB did eventually add the capability of booting Unix as well, but that was later. So Unix booted without VMB.
Also, the VAX-11/750 do not even use VMB...
Later versions of VMB (e.g. in a MicroVAX) could boot directly from
various tape devices.
Right.
Really? Not from tape? Interesting design choice...
Old VAXes booted standalone BACKUP from the console media (TU58 or RX01).
SA BACKUP then could talk to tape drives just fine. It's all you need to
install VMS, so that was all they did.
Yes. And of course, nobody ever wanted to boot anything but VMS... ;-)
Johnny
On 2013-05-17 21:00, Cory Smelosky wrote:
On Fri, 17 May 2013, Johnny Billquist wrote:
Unibus VAXen basically means VAX-11 machines. They booted either from
VMB on console media, or (for the 11/750) from a boot block. No network
capabilities there. They could not even boot from tape.
Johnny
Really? Not from tape? Interesting design choice...
That's one way of putting it. Another would be "annoying".
Yeah. VMS distribution did come on tape, but you also got a floppy or TU58 from which you booted the standalone backup on the VAX, and then did the installation using that.
Now, if you instead wanted to install something like Unix from a tape, you had to enter an initial bootstrap by hand into memory. At least the MtXinu distribution actually have the bootstrap codes for all VAX-11 systems listed in the manual.
(Quite helpful when I was installing 4.3 Reno from tape, as it turned out those bootstraps worked for my Reno tape as well, on an VAX 8650.)
Johnny
Unibus VAXen basically means VAX-11 machines.
They booted either from VMB on console media, or
(for the 11/750) from a boot block.
TO be fair, the limitation on booting was mostly just that nobody bothered
to implement support for any other device in VMB. It wasn't any kind of
hardware limitation.
Later versions of VMB (e.g. in a MicroVAX) could boot directly from
various tape devices.
Really? Not from tape? Interesting design choice...
Old VAXes booted standalone BACKUP from the console media (TU58 or RX01).
SA BACKUP then could talk to tape drives just fine. It's all you need to
install VMS, so that was all they did.
Bob
On Fri, 17 May 2013, Johnny Billquist wrote:
On 2013-05-17 19:18, Cory Smelosky wrote:
On Fri, 17 May 2013, Bob Armstrong wrote:
John Wilson wrote:
this is a ridiculous security risk
so that's why MOP had a magic number
Yep, even the DMR has a boot "password", which is just an 8 bit value
set
by dip switches on the board, that has to match up with the MOP message.
They DMA a small program into memory
Ah, I was wondering about that, because there aren't any separate boot
ROMs on the DEUNA. I bet the "small program" is just part of the T11
code.
And I'm assuming that this scheme didn't work on any UNIBUS VAXes - only
PDPs.
I'd assume so as well. On a semi-related note, wasn't it not possible
to netboot a UNIBUS VAX with a DEUNA? I seem to recall that from the
manual. It required an intermediate bootloader or something, right?
Unibus VAXen basically means VAX-11 machines. They booted either from
VMB on console media, or (for the 11/750) from a boot block. No network
capabilities there. They could not even boot from tape.
Johnny
Really? Not from tape? Interesting design choice...
--
Cory Smelosky
http://gewt.net/ Personal stuff
http://gimme-sympathy.org Experiments
On 2013-05-17 20:51, Clem Cole wrote:
On Fri, May 17, 2013 at 1:57 PM, Johnny Billquist <bqt at softjar.se
<mailto:bqt at softjar.se>> wrote:
Unibus VAXen basically means VAX-11 machines. They booted either
from VMB on console media, or (for the 11/750) from a boot block. No
network capabilities there. They could not even boot from tape.
Amen, and seemingly hard to believe. It seems so primitive by today's
standards. But it actually makes sense. Disks in those days were a
huge expense within the total system price, but definitely part of it.
A system in the Vax class really needed to be self-supporting. So the
concept of it not have local storage would have been strange and frankly
not able to be sold.
Let's also not forget that in those days Ethernet HW was not
particularly cheap either. The 3Com stinger taps cost about $500 each,
and that did not include the $~2-3K for the 3Cxxx for the Unibus.
Well, in all honesty. We did just cover network booted PDP-11s. Not to mention that PDP-11s could boot from tape pretty much from day 1.
On the other hand, VAXen (as well as PDP-11) came out on the market before Ethernet existed.
However, later VAXen definitely supported network booting, as well as totally being diskless. But those were not using the Unibus. :-)
Johnny
On Fri, May 17, 2013 at 1:57 PM, Johnny Billquist <bqt at softjar.se> wrote:
Unibus VAXen basically means VAX-11 machines. They booted either from VMB on console media, or (for the 11/750) from a boot block. No network capabilities there. They could not even boot from tape.
Amen, and seemingly hard to believe. It seems so primitive by today's standards. But it actually makes sense. Disks in those days were a huge expense within the total system price, but definitely part of it. A system in the Vax class really needed to be self-supporting. So the concept of it not have local storage would have been strange and frankly not able to be sold.
Let's also not forget that in those days Ethernet HW was not particularly cheap either. The 3Com stinger taps cost about $500 each, and that did not include the $~2-3K for the 3Cxxx for the Unibus.
I remember when Apollo announced the "Twins" machines were 2 nodes with a shared single disk in ~1984/85 - which actually did work reasonably well. Sun did the "diskless" Sun-3 in response, and they did not. I do not think DEC even tried.
The funny part is that Sun's answer was an accidental marketing genius - because it became the worlds best add in disk upgrade business for them (diskless Sun's were known as having the lack of male anatomy).
I was leading the networking group at Masscomp at the time, and my team refused to do diskless support - because thought it was a stupid product (there is a infamous email I sent to all of the company with a dyslexic typo in it - which I wish I still had). I was technically 100% right. A WS-500 cost $1.5K less that and equivalent Sun 3. But, end users could buy a diskless Sun3 for $2K less than the WS-500. -- only to discover the performance sucked. So would have to go back to Sun at $5K a crack to get the disk subsystem.
The genius was the sales got the original sale, and you wer not going to through out the Sun3 and get the cheaper system. You would spend the $5K later and make it better - sigh.
Clem
On 2013-05-17 19:18, Cory Smelosky wrote:
On Fri, 17 May 2013, Bob Armstrong wrote:
John Wilson wrote:
this is a ridiculous security risk
so that's why MOP had a magic number
Yep, even the DMR has a boot "password", which is just an 8 bit value
set
by dip switches on the board, that has to match up with the MOP message.
They DMA a small program into memory
Ah, I was wondering about that, because there aren't any separate boot
ROMs on the DEUNA. I bet the "small program" is just part of the T11
code.
And I'm assuming that this scheme didn't work on any UNIBUS VAXes - only
PDPs.
I'd assume so as well. On a semi-related note, wasn't it not possible
to netboot a UNIBUS VAX with a DEUNA? I seem to recall that from the
manual. It required an intermediate bootloader or something, right?
Unibus VAXen basically means VAX-11 machines. They booted either from VMB on console media, or (for the 11/750) from a boot block. No network capabilities there. They could not even boot from tape.
Johnny
