Conference iosg::all-in-1_v30

    P.S.
    
    I guess it would be easier if I said is this (2 x SYSUAF) even
    supported ?
    
    Thanks,
    
    Clive

A cause of only the last attachment being printed is that an old V2.2 or
earlier version of a print completion script (probably WPPSYSTEM.SCP) is
still being used. 

Up to V2.2 the print completion scripts only had to process one listing 
file.  From V2.3 print completion scripts have to process multiple files 
because the message and each attachment are separately formatted.  If a
V2.2 print completion script is used then only the last attachment listing
file is printed.  The reason this is not happening with background
formatting is that the WPPBGFORMAT script was new in V2.3 so this script
has only ever existed in a form that processes multiple listing files. 


I don't know have access to the STARS article about RMS-FNF errors so I may be 
repeating what it says.  

The background formatter symbiont processes run under the system uic.  The
print jobs they create include a Username item code with the username of
the user creating the formatter job (equilvalent to PRINT/USER=username at
DCL) so that the print job also belongs to the user and not to the system.

The system has 2 SYSUAFs and 2 sets of users.  Is it the case that each user 
is restricted to using two nodes of the cluster?  If so it is possible 
that:

    o	a background formatter symbiont process is running on one of the
    	nodes of the cluster

    o	it is processing a job for a user, but it is not running on one of
    	the nodes that the user is allowed to use

    o	when it creates the print job the SYSUAF record cannot be accessed,
    	just as if the user had tried to login to that node


The above could be verified from DCL by:

    o	logging into the node on which the background formatter symbiont is 
    	running into the SYSTEM account

    o	try to create a print job with a /USER=username qualifier



Richard

    Thanks Richard,
    
    > The system has 2 SYSUAFs and 2 sets of users. Is it the case that
    > each user is restricted to using two nodes of the cluster?
    
    Yes
    
    Unfortuantely it did not appear to be an old WPPSYSTEM since it has a
    date of 30-MAY-1990 like my 2.4 system. We recompiled the A1TXL with no
    change.
    Strangely, when printing to FILE, it contains codes for the body text
    and the attachment (though I haven't tried the codes yet to see how it
    prints here). Since it contains the codes I presume this confirms this
    is not an old script but I find it difficult to understand why it won't
    print the body. 2xSYSUAF couldn't cause this surely? (customer is sure
    it does). Is it supported ?
    
    Thanks,
    
    Clive
    seems to have escape codes
    for the body text and the attachment text. 

An ALL-IN-1 print operation, as performed by the WPPRINT script, consists
of two stages, 'formatting' and 'printing'.  During the formatting stage a
listing file for the message and each attachment is created.  It then calls
a 'print completion' script to perform the printing stage, which consists
of doing something with the listing files created during the previous
stage.  Different print completion scripts are called depending on the type
of print destination. 

For example, WPPFILE is called when the print destination is FILE.  This
script asks the user for the name of a VMS file and appends all the listing
files into that file.  WPPSYSTEM is called when the print destination is a
VMS system queue.  This scripts creates a print job to print the listing
files. 

Since a print to FILE is working and print to a system queue is not I still 
think that the problem is associated with the WPPSYSTEM script.  And more 
specifically, the VMS print job is being created with only one the listing 
file instead of multiple listing files.

Is there an old WPPSYSTEM.SCP in the CMTXL, because if there is it will be 
used in preference to the Digital element.  The ALL-IN-1 command 

    oa$txl_list wppsystem,do

will show which WPPSYSTEM script is being used.


I think the WPPSYSTEM and the 2 x SYSUAF problems are separate problems, 
and should be investigated as such.

Did you look any further into the question of 

    - which nodes the OA$FORMATTER symbionts are executing on

    - which users have problems with background formatter jobs

    - and which nodes those users are allowed to log into



Richard

    Apologies and thanks,
    
    Today a 2.2 script miraculously appeared (in do_british as opposed to
    do_share).
    
    Thanks,
    
    Clive