T.R | Title | User | Personal Name | Date | Lines |
---|
1494.1 | By George, I think it might work! | IOSG::TALLETT | Arranging bits for a living... | Thu Sep 24 1992 10:45 | 22 |
|
Hmmm, this is an interesting one! You'd have to run the old
queue manager of course and I would do the ALL-IN-1 installation
or link or whatever on a V5.4 system so that the code thinks its
running on V5.4 and doesn't try to use any V5.5 features, but my
guess is that it would probably work. I've never heard of anyone
doing this so you are on your own, but you knew that already! :-)
The things I'm thinking of, just for others to check my logic:
OA$FORMATTER - should be okay as long as you don't mix
queue managers.
UAI$/set_password - links SYS.STB which shouldn't have
changed
FileCab Server - uses its own threads package, not the V5.5
one
Deferred image activation - should be ok, works on 5.4 & 5.5
Maybe someone can think of something that wouldn't work?
Regards,
Paul
|
1494.2 | Missed the second system disk on the first pass | IOSG::TALLETT | Arranging bits for a living... | Thu Sep 24 1992 10:49 | 8 |
|
Just noticed that you would, of course, have a separate system
disk. Would you have a separate ALL-IN-1 installation and
separate UAF files? You might notice the inconsistency but
I still think it would work.
Regards,
Paul
|
1494.3 | system disk should be the same | MSDSWS::DUNCAN | Things can't get much worse..I Hope! | Thu Sep 24 1992 12:13 | 25 |
| I forgot to say this was ALL-IN-1 V2.4.
I will have to double check how they built the system disk, but a lot
of the second system disk files came from the first system disk.
No problem with the queue manager, they are at A5.5-2HW. I forgot
about the "A".
The only thing I can think of that might cause a problem is if any of
the sys$share files are not there, but I think I could take care of
this.
Note that I do not want to do a new installation. I want to use all
the existing ALL-IN-1 files which are available to both the A5.5-2HW
node and the V5.4-x nodes. The UAF, NETPROXY... are all the same as
they were copied form the 5.4 systems.
Again, I think is should work and of course will test before we turn
users loose. I just want to make sure that nothing major will break
and screw up the entire cluster.
Thanks,
Darryl
|
1494.4 | Common UAF seems to make sense | IOSG::TALLETT | Arranging bits for a living... | Thu Sep 24 1992 18:48 | 8 |
|
I would define SYSUAF etc. to point to common files rather than
taking a copy.
Let us know how you get on.
Regards,
Paul
|