"Gary Whitten" writes:

> All
>
> I've recently dusted off my 4/233 (Model PB51C-AA) after about 3 years of
> having it just sit there and I've subsequently forgetten what got it into
> its present vegetative state.
>
> Right now, when I boot, I get the following on the screen:
>
> eb......ea.e9.e8.e7.e6.e5
> v6.2-2, built on Aug 7, 1996
>>>>

Congratulations, your system is working and you are at the famous SRM
prompt.

> At the prompt I can do some rudimentary things but nothing of substance.
> Additionally, my unix and linux are very rusty, which is why I'm doing this
> in the first place so unlike Windows/X86 systems, I don't have my usual list
> of tricks and procedures to try to sort things out.
>
> What I'd like to do is drop Alpha RedHat 7.2 on this. I have the CDs burnt
> from the website but my system doesn't seem to want to boot from a CD.
> Also, I have a Red Hat 9 boot floppy for X86 systems which won't work but I
> think that's 'normal'.
>
> Whatever state that the system comes up in doesn't have dd readily
> available, nor does it have ls, so I'm kind of stuck.
>
> I don't even remember how to get into the CMOS of the system to check the
> settings.

You are there. Run the command "sh dev" and see which device is the
CDROM. It'll be something like dka4. Then insert the CD and run
"boot dka4". If there's a hard disk in the machine you can try "boot"
without any arguments and see if something starts up.

--
Måns Rullgård
mru@kth.se

Gary:

*whew*!!!
It must have been a LONG time since you've booted this system.
Long enough to have forgotten what the SRM console is.
There is no CMOS in this system per se, you could load the AlphaBios/ARC
console which *simulates* a PeeCee style CMOS.
Try typing "ARC" from the >>> prompt (no quotes).
If it works, you'll have something which might be a little more
familiar. The caveat is, the AlphaBios/ARC console will only boot NT or
Linux, not OpenVMS or Unix.
If it doesn't change anything, then you have a 'half flash' system and
it's probably not worth the hassle of updating the firmware so that you
could use the AlphaBios/ARC console.
SRM does everything that this system needs to boot any *REAL* operating
system. ;-) (including Linux).


Here's a link to AlphaLinux.org, probably the best source of Linux for
Alpha information on the net:
http://www.alphalinux.org/

Good luck!

TomC

Gary Whitten wrote:
> All
>
> I've recently dusted off my 4/233 (Model PB51C-AA) after about 3 years of
> having it just sit there and I've subsequently forgetten what got it into
> its present vegetative state.
>
> Right now, when I boot, I get the following on the screen:
>
> eb......ea.e9.e8.e7.e6.e5
> v6.2-2, built on Aug 7, 1996
>
>
>
> At the prompt I can do some rudimentary things but nothing of substance.
> Additionally, my unix and linux are very rusty, which is why I'm doing this
> in the first place so unlike Windows/X86 systems, I don't have my usual list
> of tricks and procedures to try to sort things out.
>
> What I'd like to do is drop Alpha RedHat 7.2 on this. I have the CDs burnt
> from the website but my system doesn't seem to want to boot from a CD.
> Also, I have a Red Hat 9 boot floppy for X86 systems which won't work but I
> think that's 'normal'.
>
> Whatever state that the system comes up in doesn't have dd readily
> available, nor does it have ls, so I'm kind of stuck.
>
> I don't even remember how to get into the CMOS of the system to check the
> settings.
>
> Any help is greatly appreciated. Thanks in advance.
>
> Gary
>
>

> You are there. Run the command "sh dev" and see which device is the
> CDROM. It'll be something like dka4. Then insert the CD and run
> "boot dka4". If there's a hard disk in the machine you can try "boot"
> without any arguments and see if something starts up.

Ok, great news, much easier than I dared to hope for. It was dka400,
actually. I got an abnormal termination on the boot, however.

Started with the line:
----

Traceback (innermost last):

file "/usr/bin/anaconda" , line 477, in ?
from splashscreen import splashScreenShow
file "/usr/bin/anaconda/splashscreen.py", line 21, in ?
from gtk import *
file "/usr/lib/python1.5/site-packages/gtk.py", line 29, in ?
from _gtk.gtk_init()

RuntimeError: cannot open display

----
Does any of the above ring any bells?

I had seen something about a 'wrapper' needed for some systems here:

ftp://ftp2.compaq.com/pub/linux/RedH...pha/HPWrapper/

This may be appropriate for me but I'm not sure.

"Gary Whitten" writes:

>> You are there. Run the command "sh dev" and see which device is the
>> CDROM. It'll be something like dka4. Then insert the CD and run
>> "boot dka4". If there's a hard disk in the machine you can try "boot"
>> without any arguments and see if something starts up.
>
> Ok, great news, much easier than I dared to hope for. It was dka400,
> actually. I got an abnormal termination on the boot, however.
>
> Started with the line:
> ----
>
> Traceback (innermost last):
>
> file "/usr/bin/anaconda" , line 477, in ?
> from splashscreen import splashScreenShow
> file "/usr/bin/anaconda/splashscreen.py", line 21, in ?
> from gtk import *
> file "/usr/lib/python1.5/site-packages/gtk.py", line 29, in ?
> from _gtk.gtk_init()
>
> RuntimeError: cannot open display
>
> ----
> Does any of the above ring any bells?

Maybe the graphics card isn't supported or isn't properly detected.
There should be a way to choose a text installer instead. What
graphics card is in the machine?

--
Måns Rullgård
mru@kth.se

Gary Whitten wrote:
>> You are there. Run the command "sh dev" and see which device is the
>> CDROM. It'll be something like dka4. Then insert the CD and run
>> "boot dka4". If there's a hard disk in the machine you can try "boot"
>> without any arguments and see if something starts up.
>
> Ok, great news, much easier than I dared to hope for. It was dka400,
> actually.

It happens, most likely, to be the same thing in your case.

> I got an abnormal termination on the boot, however.
.....
>
> RuntimeError: cannot open display

You have problems with your graphics card and X. If so then
at >>> prompt type:

b dka400 -fl 3

to install in a "text mode". You will worry about X later.

If you will do 'b dka4 -fl i' instead then you end up at
"interactive" aboot prompt and you can list all predefined boot lines.


> I had seen something about a 'wrapper' needed for some systems here:
>
> ftp://ftp2.compaq.com/pub/linux/RedH...pha/HPWrapper/
>
> This may be appropriate for me but I'm not sure.

It is appropriate and has a potential to make your life much easier.
See README at that location. Check also all updates (although at the
location above they got stuck on 2003-Sep-29 while the last "official"
ones were from 2003-Dec-10; to be found at
ftp://alpha.crl.dec.com/pub/linux/re...alpha/updates/ and for
later ones you will likely need to recompile srpms from Legacy project).

Michal

Tom Crabtree wrote:
> Try typing "ARC" from the >>> prompt (no quotes).
> If it works, you'll have something which might be a little more
> familiar.
......
> SRM does everything that this system needs to boot any *REAL* operating
> system. ;-) (including Linux).

There are technical reasons to prefer SRM over ARC to boot Linux
kernels so unless you are __really forced__ in some way to use ARC
(like no SRM on your board or you have some hardware which prevents
booting with all available versions of SRM) then you have to be
truly insane to mess with ARC and milo. Besides booting with SRM
is so much simpler ...

Michal

On Wed, 28 Jan 2004 18:00:51 +0000 (UTC), Michal Jaegermann
wrote:
>There are technical reasons to prefer SRM over ARC to boot Linux
>kernels

Like what? I've always thought it was six of one, half a dozen of the other.

> so unless you are __really forced__ in some way to use ARC
>(like no SRM on your board or you have some hardware which prevents
>booting with all available versions of SRM) then you have to be
>truly insane to mess with ARC and milo.

Not really. I've been using milo for years on my PC164/500, and it has been
really simple.

>Besides booting with SRM is so much simpler ...

....that is, assuming you can deal with BSD partitions and getting aboot to
work. That's gotten better over time, but it hasn't always been
straightforward.

jmaynard@thebrain.conmicro.cx (Jay Maynard) writes:

> On Wed, 28 Jan 2004 18:00:51 +0000 (UTC), Michal Jaegermann
> wrote:
>>There are technical reasons to prefer SRM over ARC to boot Linux
>>kernels
>
> Like what? I've always thought it was six of one, half a dozen of the other.

PALcode, IIRC.

--
Måns Rullgård
mru@kth.se

On Wed, 28 Jan 2004 19:49:37 +0100, Måns Rullgård wrote:
>jmaynard@thebrain.conmicro.cx (Jay Maynard) writes:
>> On Wed, 28 Jan 2004 18:00:51 +0000 (UTC), Michal Jaegermann
>> wrote:
>>>There are technical reasons to prefer SRM over ARC to boot Linux
>>>kernels
>> Like what? I've always thought it was six of one, half a dozen of the other.
>PALcode, IIRC.

I thought the PALcode was the same for everything but VMS. In particular, I
know that Unix uses the Tru64 PALcode, and thought NT did as well.

jmaynard@thebrain.conmicro.cx (Jay Maynard) writes:

> On Wed, 28 Jan 2004 19:49:37 +0100, Måns Rullgård wrote:
>>jmaynard@thebrain.conmicro.cx (Jay Maynard) writes:
>>> On Wed, 28 Jan 2004 18:00:51 +0000 (UTC), Michal Jaegermann
>>> wrote:
>>>>There are technical reasons to prefer SRM over ARC to boot Linux
>>>>kernels
>>> Like what? I've always thought it was six of one, half a dozen of the other.
>>PALcode, IIRC.
>
> I thought the PALcode was the same for everything but VMS. In particular, I
> know that Unix uses the Tru64 PALcode, and thought NT did as well.

Well, it doesn't. NT uses its own. I don't know what MILO does.

--
Måns Rullgård
mru@kth.se

Well, I used the text mode and started with the wrapper and things seemed to
go rather well. However, after it was all said and done and I rebooted, I
went back to the SRM prompt.

I thought that LILO or something else would be kicking in at this point.

Thoughts? Suggestions?

Gary

"Gary Whitten" writes:

> Well, I used the text mode and started with the wrapper and things seemed to
> go rather well. However, after it was all said and done and I rebooted, I
> went back to the SRM prompt.

Do "set auto_action boot" at the SRM prompt. This will make SRM boot
automatically at power on. In case things get ugly and you want the
prompt back, the trick is to hit ctrl-c at the right moment just
before it starts booting.

> I thought that LILO or something else would be kicking in at this point.

It's called aboot on Alphas.

There's a slightly oldish FAQ at http://www.alphalinux.org/faq/srm.html.

--
Måns Rullgård
mru@kth.se

In article , Måns Rullgård wrote:
> "Gary Whitten" writes:
>
>> Well, I used the text mode and started with the wrapper and things seemed to
>> go rather well. However, after it was all said and done and I rebooted, I
>> went back to the SRM prompt.
>
> Do "set auto_action boot" at the SRM prompt. This will make SRM boot
> automatically at power on. In case things get ugly and you want the
> prompt back, the trick is to hit ctrl-c at the right moment just
> before it starts booting.
>
>> I thought that LILO or something else would be kicking in at this point.
>
> It's called aboot on Alphas.
>
> There's a slightly oldish FAQ at http://www.alphalinux.org/faq/srm.html.

Or, just set the relevant boot_* variables and type 'boot'
(without the quote marks) at the prompt. That way, you
don't have to do the CTRL-C dance if you need to get back to
the prompt.

Good luck.

Robert Riches
spamtrap42@verizon.net
(Yes, that is one of my email addresses.)

OK, thanks guys....all the vars are set, so I can just type 'boot' now and
it'll come up. When I'm sure everything is working, I'll set the
auto_action var. Now I just need to relearn my Linux

"Robert M. Riches Jr." wrote in message
news:slrnc1g8d6.4na.rob@one.localnet...
> In article , Måns Rullgård wrote:
> > "Gary Whitten" writes:
> >
> >> Well, I used the text mode and started with the wrapper and things
seemed to
> >> go rather well. However, after it was all said and done and I
rebooted, I
> >> went back to the SRM prompt.
> >
> > Do "set auto_action boot" at the SRM prompt. This will make SRM boot
> > automatically at power on. In case things get ugly and you want the
> > prompt back, the trick is to hit ctrl-c at the right moment just
> > before it starts booting.
> >
> >> I thought that LILO or something else would be kicking in at this
point.
> >
> > It's called aboot on Alphas.
> >
> > There's a slightly oldish FAQ at http://www.alphalinux.org/faq/srm.html.
>
> Or, just set the relevant boot_* variables and type 'boot'
> (without the quote marks) at the prompt. That way, you
> don't have to do the CTRL-C dance if you need to get back to
> the prompt.
>
> Good luck.
>
> Robert Riches
> spamtrap42@verizon.net
> (Yes, that is one of my email addresses.)

Gary Whitten wrote:
>
> All
>
> I've recently dusted off my 4/233 (Model PB51C-AA) after about 3 years of
> having it just sit there and I've subsequently forgetten what got it into
> its present vegetative state.
>
> Right now, when I boot, I get the following on the screen:
>
> eb......ea.e9.e8.e7.e6.e5
> v6.2-2, built on Aug 7, 1996
> >>>
>
> At the prompt I can do some rudimentary things but nothing of substance.

O.k. then. Try this:

>>> SHOW *

....to display the console environment variables. "bootdef_dev" will
indicate which device it will try to boot from if you just issue:

>>> b

"boot_osflags" will show the "-fl" value used by default on the b[oot]
command (yes, command CAN be abbreviated!).

>>> show devices

....will display the devices that the console can see.

There is, as mentioned in another thread, a "help" command, also.

> Additionally, my unix and linux are very rusty, which is why I'm doing this
> in the first place so unlike Windows/X86 systems, I don't have my usual list
> of tricks and procedures to try to sort things out.
>
> What I'd like to do is drop Alpha RedHat 7.2 on this. I have the CDs burnt
> from the website but my system doesn't seem to want to boot from a CD.

Use SHOW DEVICES (sh dev for short) to find your CD-ROM device. By
default as set up at the factory, it was frequently something like
DKA500 or DKA400.

> Also, I have a Red Hat 9 boot floppy for X86 systems which won't work but I
> think that's 'normal'.
>
> Whatever state that the system comes up in doesn't have dd readily
> available, nor does it have ls, so I'm kind of stuck.
>
> I don't even remember how to get into the CMOS of the system to check the
> settings.

There's a utility diskette that you may need. Also, some setup utilities
only run from the "ARC" console.

> Any help is greatly appreciated. Thanks in advance.

Hope this helps.

--
David J. Dachtera
dba DJE Systems
http://www.djesys.com/

Unofficial Affordable OpenVMS Home Page:
http://www.djesys.com/vms/soho/

Måns Rullgård wrote:
> jmaynard@thebrain.conmicro.cx (Jay Maynard) writes:
>>
>> I thought the PALcode was the same for everything but VMS. In particular, I
>> know that Unix uses the Tru64 PALcode, and thought NT did as well.
>
> Well, it doesn't. NT uses its own. I don't know what MILO does.

PAL code and HAL (Hardware Abstraction Layer) are supplied to MILO in
a form of binary object files, created originally for NT, which a very
long time ago were provided by DEC for that purpose. From what I was
told by people who should know these are old, obsolete and were not
so great in the first place. 'aboot' relies on SRM to do the
corresponding job and this directly translates into a performance.
Nothing earth shattering but still... (and if you have reasonably
recent SRM this apparently avoids some other problems too).

Also kernels which were compiled to boot "high end" Alpha (wildfire,
etc.) will not work with ARC/milo but they are fine with SRM/aboot.

Michal

Michal Jaegermann writes:

> Also kernels which were compiled to boot "high end" Alpha (wildfire,
> etc.) will not work with ARC/milo but they are fine with SRM/aboot.

That's because of the new kernel

--
Måns Rullgård
mru@kth.se

Michal Jaegermann writes:

> Also kernels which were compiled to boot "high end" Alpha (wildfire,
> etc.) will not work with ARC/milo but they are fine with SRM/aboot.

That's because of the new kernel start address, right?

--
Måns Rullgård
mru@kth.se

Måns Rullgård wrote:
> Michal Jaegermann writes:
>
>> Also kernels which were compiled to boot "high end" Alpha (wildfire,
>> etc.) will not work with ARC/milo but they are fine with SRM/aboot.
>
> That's because of the new kernel start address, right?

Right; although this is rather "moveable" address as aboot will be
not fazed by a different load address. I think that some variants
of milo may be capable of handling that too, so this does not seem
to a be a killer obstacle, but one can hardly expect much development
happening with milo.

Michal