[comp.sys.apollo] install++ troubles persist

rtp1@tank.uchicago.edu (raymond thomas pierrehumbert) (01/18/90)

I am still trying to install fortran 10.7.p on my DN10000 running OS10.1
Using either install++ or minst, I get the message:
    ERROR:  Product Apollo os 10.2.p not found.
or words to that effect.  Following a previous thread on this 
newsgroup, I assumed the problem was that I had deleted my os AA
and followd a suggestion that I could solve the problem by simply
restoring the release index from tape (which I did).  I restored
the 10.1.p release index, and IT DIDN't MAKE A BIT OF DIFFERENCE.
I got the same error message, and the product failed to install.
    Any ideas as to what I should do now?  I don't have the time
or the disk space to restore the whole os, and anyway, that probably
wouldn't make any difference, since install++ would think it hadn't
been "installed".  I haven't received 10.2.p yet, and anyway, don't
intend to install it until I have a working version of NFS for 10.2
in hand, so that isn't a way out.
   What in the world is going on here?  

dbfunk@ICAEN.UIOWA.EDU (David B Funk) (01/18/90)

> I am still trying to install fortran 10.7.p on my DN10000 running OS10.1
> Using either install++ or minst, I get the message:
>     ERROR:  Product Apollo os 10.2.p not found.

You may have a messed up baseline file ("/install/baseline/baseline.*").
Look at the one with the hightest version number and see if there is a
line that has the 10.2 version number in it. If you have more than one
version, you may be able to fall back to an older one.

For another work-around look at page 4-24 of "Installing Software with
Apollo's Release and Installation Tools" (#008860-A00).

Dave

rtp1@tank.uchicago.edu (raymond thomas pierrehumbert) (01/19/90)

I seem to have only one baseline file around.  I don't know what happened
to the rest of them; I don't think I deleted them.  The baseline
file thinks that I don't have any OS installed.  Any ideas how to
fix this?  Can I force the recreation of a baseline file??

 RAI V0.16 Thu Nov  9 12:08:10 1989
Release ri.apollo.os.v.10.1.p disk none
Option apollo.sys5.3.text N/A
Set apollo.sys5.3.crosdev N/A
Option apollo.sys5.3.pgmdev N/A
Option apollo.sys5.3.graphics N/A
Option apollo.sys5.3.terminfo N/A
Option apollo.sys5.3.lp_site N/A
Option apollo.sys5.3.include N/A
Option apollo.sys5.3.manuals N/A
Option apollo.sys5.3.all_links N/A
Option apollo.bsd4.3.text copy
Set apollo.bsd4.3.crosdev both
Option apollo.bsd4.3.pgmdev copy
Option apollo.bsd4.3.graphics copy
Option apollo.bsd4.3.lp_site copy
Option apollo.bsd4.3.include copy
Option apollo.bsd4.3.manuals copy
Option apollo.bsd4.3.all_links copy
Option apollo.uucp_site copy
Option apollo.unix_mail_site copy
Option apollo.new_site copy
Option apollo.games_site copy
Option apollo.aegis.manuals copy
Option com_db N/A
<Lots of stuff deleted here...>
Release ri.apollo.ftn.v.10.5.p disk version
Option apollo_ftn.a88k.a88k_install_examples copy
Option apollo_ftn.a88k.a88k_install_help copy
Option apollo_ftn.a88k.a88k_option_link copy
<More stuff deleted>

It looks like the first line is the one causing the problem.  Is
there some simple way I can change it?  Thanks.  I have read the
manuals, but I am still confused.

rtp1@tank.uchicago.edu (raymond thomas pierrehumbert) (01/19/90)

My install++ problem is now resolved.  The problem had nothing to do with
release indexes, bad baseline files, or anything of the sort.  the
problem was that when install++ complains:

None of the following products were found:
      Apollo os.10.2.p
      ftnlib (something.something.something)

what it really means is that you need to have Either installed 10.2,
OR have installed the current ftnlib.  This is a rather mysterious
message. Anyway, when I installed the ftnlib, everything went fine. This
is a problem with the release notes that came with the tape, as the
release notes imply that you do not need to install the newest ftnlib
if you are not using BYTE variables.
    Thanks to Dan Brooks of the 800-number, for a conversation that 
somehow jogged my brain into thinking of this interpretation of the
error message.