[comp.sys.dec] VMS upgrade path question

klr@hadron.COM (Kurt L. Reisler) (06/13/91)

Disclaimer, I know zilch about VMS or DECNET.

With that out of the way, is it possible to upgrade from VMS 4.4 to VMS
5.1 or higher, without first upgrading to VMS 5.0?  The reason for this
is simple ( and stupid ).  It seems that in the process of relocating
the collection of upgrade tapes from location A to location B, the car
in which they were being carried attempted to occupy the same physical
space as a bus.  Along with the structural realignment of the car, the
box containing the 5.0 tapes was, shall we say, damaged beyond
recognition.  The other tapes are in their original condition.

So, do we need to get replacements for the 5.0 tapes, or can we just
start the upgrade with the 5.1 tapes?

Thanks in advance.

Kurt Reisler

diamond@jit533.swstokyo.dec.com (Norman Diamond) (06/13/91)

In article <962@hadron.COM> klr@hadron.COM (Kurt L. Reisler) writes:
>It seems that in the process of relocating the collection of upgrade tapes
>from location A to location B, the car in which they were being carried
>attempted to occupy the same physical space as a bus.

Should have copied the upgrades over an Ethernet connection.  That has
collision detection.
--
Norman Diamond       diamond@tkov50.enet.dec.com
If this were the company's opinion, I wouldn't be allowed to post it.
Permission is granted to feel this signature, but not to look at it.

gpwrmdh@gp.co.nz (06/13/91)

In article <962@hadron.COM>, klr@hadron.COM (Kurt L. Reisler) writes:
> Disclaimer, I know zilch about VMS or DECNET.
> 
> With that out of the way, is it possible to upgrade from VMS 4.4 to VMS
> 5.1 or higher, without first upgrading to VMS 5.0?  
[funny story about car crash deleted]
> So, do we need to get replacements for the 5.0 tapes, or can we just
> start the upgrade with the 5.1 tapes?

Afraid so. You will need V5.0, but I don't think you need the V4.6 tape - I 
may be wrong here, but I haven't got the manuals handy.

> 
> Thanks in advance.
> 
> Kurt Reisler
-- 
----------------------------------------------------------------------------
Martin D. Hunt			
GP Print Limited		USEnet address : martinh@gp.co.nz
Wellington			PSI address    : PSI%0530147000028::martinh
New Zealand			Phone	       : +64 4 4965790
-----------------------------------------------------------------------------

mcculley@alien.enet.dec.com (06/14/91)

In article <1991Jun13.065949.29415@tkou02.enet.dec.com>, diamond@jit533.swstokyo.dec.com (Norman Diamond) writes...
#>In article <962@hadron.COM> klr@hadron.COM (Kurt L. Reisler) writes:
#>>It seems that in the process of relocating the collection of upgrade tapes
#>>from location A to location B, the car in which they were being carried
#>>attempted to occupy the same physical space as a bus.

Would that make the problem a bus error?

#>Should have copied the upgrades over an Ethernet connection.  That has
#>collision detection.

I got the idea that the problem was not the collision going undetected, 
but that it couldn't be corrected by retries.

--bruce

k2@bl.physik.tu-muenchen.de (Klaus Steinberger) (06/14/91)

gpwrmdh@gp.co.nz writes:

>In article <962@hadron.COM>, klr@hadron.COM (Kurt L. Reisler) writes:
>> Disclaimer, I know zilch about VMS or DECNET.
>> 
>> With that out of the way, is it possible to upgrade from VMS 4.4 to VMS
>> 5.1 or higher, without first upgrading to VMS 5.0?  
>[funny story about car crash deleted]
>> So, do we need to get replacements for the 5.0 tapes, or can we just
>> start the upgrade with the 5.1 tapes?

>Afraid so. You will need V5.0, but I don't think you need the V4.6 tape - I 
>may be wrong here, but I haven't got the manuals handy.

Sure, but it's a long long way to do the upgrade, i believe you better do
an fresh install of the highest version of VMS, you have. I spent at least
two days to walk through the upgrade from 4.7 to 5.0-2. And every major
upgrade (5.0 -> 5.1, 5.1 -> 5.2 and so on) eats one more day of your 
live. One nice effect: A fresh install defragments your disk (INITIALIZE !).

Sincerely,
Klaus Steinberger
--
Klaus Steinberger               Beschleunigerlabor der TU und LMU Muenchen
Phone: (+49 89)3209 4287        Hochschulgelaende
FAX:   (+49 89)3209 4280        D-8046 Garching, Germany
BITNET: K2@DGABLG5P             Internet: k2@bl.physik.tu-muenchen.de

k2@bl.physik.tu-muenchen.de (Klaus Steinberger) (06/14/91)

mcculley@alien.enet.dec.com writes:


>In article <1991Jun13.065949.29415@tkou02.enet.dec.com>, diamond@jit533.swstokyo.dec.com (Norman Diamond) writes...
>#>In article <962@hadron.COM> klr@hadron.COM (Kurt L. Reisler) writes:
>#>>It seems that in the process of relocating the collection of upgrade tapes
>#>>from location A to location B, the car in which they were being carried
>#>>attempted to occupy the same physical space as a bus.

>Would that make the problem a bus error?

>#>Should have copied the upgrades over an Ethernet connection.  That has
>#>collision detection.

>I got the idea that the problem was not the collision going undetected, 
>but that it couldn't be corrected by retries.

The solution will be CSMA/CA  (Collision avoidance).

Klaus
--
Klaus Steinberger               Beschleunigerlabor der TU und LMU Muenchen
Phone: (+49 89)3209 4287        Hochschulgelaende
FAX:   (+49 89)3209 4280        D-8046 Garching, Germany
BITNET: K2@DGABLG5P             Internet: k2@bl.physik.tu-muenchen.de

martin%boot.decnet@edwards-tems.af.mil (06/15/91)

In article <k2.676880102@woodstock>, k2@bl.physik.tu-muenchen.de (Klaus Steinberger) writes:
> 
> Sure, but it's a long long way to do the upgrade, i believe you better do
> an fresh install of the highest version of VMS, you have. I spent at least
> two days to walk through the upgrade from 4.7 to 5.0-2. And every major
> upgrade (5.0 -> 5.1, 5.1 -> 5.2 and so on) eats one more day of your 
> live. One nice effect: A fresh install defragments your disk (INITIALIZE !).
> 

If I remember correctly, there is no full VMS tape between 4.6 (maybe 4.7) and
5.2.  So you can do a fresh install with only a 5.2 tape, but not with only 5.0
or 5.1.  To get to 5.1, you must install 4.6 (or maybe 4.7), and then do the
upgrades up through each rev to 5.1.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
Gary S. Martin                !  (805)277-4509  DSN 527-4509
6510th Test Wing/TSWS         !  Martin@Edwards-TEMS.af.mil
Edwards AFB, CA 93523-5000    ! 
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 

gpwrmdh@gp.co.nz (06/15/91)

In article <1991Jun14.112726.76@%boot.decnet@edwards-tems.af.mil>, martin%boot.decnet@edwards-tems.af.mil writes:
>
> If I remember correctly, there is no full VMS tape between 4.6 (maybe 4.7) and
> 5.2.  So you can do a fresh install with only a 5.2 tape, but not with only 5.0
> or 5.1.  To get to 5.1, you must install 4.6 (or maybe 4.7), and then do the
> upgrades up through each rev to 5.1.
> 

No, you don't remember correcly. V5.0 was a full VMS tape - "point nought" 
releases always will be. During V4.x, the "point even number" (i.e, 4.0, 
4.2, 4.4, 4.6) were full VMS tapes. V4.7 definitely was not. The strategy 
has changed for V5.x, with generally only the "dash releases" not being 
full releases - I think V5.1 may be an exception to this. Certainly V5.0, 
5.2, 5.3, and 5.4 have been full releases.

-- 
----------------------------------------------------------------------------
Martin D. Hunt			
GP Print Limited		USEnet address : martinh@gp.co.nz
Wellington			PSI address    : PSI%0530147000028::martinh
New Zealand			Phone	       : +64 4 4965790
-----------------------------------------------------------------------------

bruns@fallout.uucp (Alan Bruns) (06/16/91)

In article <962@hadron.COM>, klr@hadron.COM (Kurt L. Reisler) writes:
> Disclaimer, I know zilch about VMS or DECNET.
> 
> With that out of the way, is it possible to upgrade from VMS 4.4 to VMS
> 5.1 or higher, without first upgrading to VMS 5.0?

As several postings have already mentioned, your best bet is to do a virgin
install.  But...

	o With version 5 there was a change made to the layout of
	  JBCSYSQUE.DAT.  There's a conversion procedure that shipped with
	  V5.0 that does this.  You can probably pick it up from somebody.
	  Or, use a command procedure that was recently posted on DECUServe
	  called FIXQUE.  I think a pre-V5.0 version of it was posted as
	  well as the post-V5 copy I grabbed.  If you don't have much in the
	  way of forms, ques, etc., defined, this is a don't-care item.

	o There have been at least two changes to the SYSUAF file, in V5.2
	  (I think) and V5.4.  CAPTIVE/RESTRICTED flag is new since V4.x, and
	  the password dictionary/validation stuff was added in V5.4.  Once
	  again, if you only have a few UAF entries, do a UAF LIST/FU before
	  the install and key it all in afterwards.

There have been a number of postings on the same basic subject on DECUServe
over the past year or so.  These are the main sticking points (as I remember
them) in doing a jump past V5.0.  Doubtless some other folks remember other
things and will jump in.

Kurt, if you're not still on DECUServe and would like, I'd be glad to extract
as much of the discussions as I can find there on this and send them to you.

Alan Bruns
Allied Electronics, Inc.		Bruns@Fallout.UUCP
Fort Worth, TX  76118			Bruns@Eisner.DECUS.Org
(817) 595-6420

martin%boot.decnet@edwards-tems.af.mil (06/17/91)

In article <1991Jun15.133712.1542@gp.co.nz>, gpwrmdh@gp.co.nz writes:
> 
> No, you don't remember correcly. V5.0 was a full VMS tape - "point nought" 
> releases always will be. During V4.x, the "point even number" (i.e, 4.0, 
> 4.2, 4.4, 4.6) were full VMS tapes. V4.7 definitely was not. The strategy 
> has changed for V5.x, with generally only the "dash releases" not being 
> full releases - I think V5.1 may be an exception to this. Certainly V5.0, 
> 5.2, 5.3, and 5.4 have been full releases.
> 

Agreed that .0 releases are ordinarily full VMS releases but 5.0 was not a full
VMS tape because there was no working standalone backup for 5.0 or 5.1. 
Standalone backup was fixed with the release of 5.2, which was a full tape.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
Gary S. Martin                !  (805)277-4509  DSN 527-4509
6510th Test Wing/TSWS         !  Martin@Edwards-TEMS.af.mil
Edwards AFB, CA 93523-5000    ! 
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 

herman@corpane.uucp (Harry Herman) (06/21/91)

In <1991Jun17.094735.78%boot.decnet@edwards-tems.af.mil> martin%boot.decnet@edwards-tems.af.mil writes:

>In article <1991Jun15.133712.1542@gp.co.nz>, gpwrmdh@gp.co.nz writes:
>> 
>> No, you don't remember correcly. V5.0 was a full VMS tape - "point nought" 
>> releases always will be. During V4.x, the "point even number" (i.e, 4.0, 
>> 4.2, 4.4, 4.6) were full VMS tapes. V4.7 definitely was not. The strategy 
>> has changed for V5.x, with generally only the "dash releases" not being 
>> full releases - I think V5.1 may be an exception to this. Certainly V5.0, 
>> 5.2, 5.3, and 5.4 have been full releases.
>> 

>Agreed that .0 releases are ordinarily full VMS releases but 5.0 was not a full
>VMS tape because there was no working standalone backup for 5.0 or 5.1. 
>Standalone backup was fixed with the release of 5.2, which was a full tape.

>- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
>Gary S. Martin                !  (805)277-4509  DSN 527-4509
>6510th Test Wing/TSWS         !  Martin@Edwards-TEMS.af.mil
>Edwards AFB, CA 93523-5000    ! 
>- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
My VMS V.50 came complete with Standalone Backup on the TK50.  I just checked
it out to make sure.  I know that I used it to install V5.0 from scratch on
a test disk when V5.0 first came out.

I seem to recall the release notes for early VMS V5.x systems cautioning not
to generate new Standalone Backup's with STABACKIT.COM due to some problems,
but the distribution itself does have a Standalone Backup on it, making it
a complete distribution capable of being installed from scratch on a new
disk.

			Harry Herman
			herman@corpane