Conference galvia::decintact

435.0. "terminal manager bug/ invalid atb" by CSC32::J_HENSON (Don't get even, get ahead!) Fri Feb 21 1997 22:46

Decintact V?.?, ovms v5.5-2

A customer (Equitable Life Insurance/Canada) has reported the following
errors.  This is effectively disabling half of their production. 

This is occurring from an application that has been stable for the past
2  years, and only began occurring 2 days ago.  The only recent 
change that the user can identify is that they installed (and then
deleted) a new liberator global section.

The customer needs to know how to resolve this as soon as possible.

Thanks,

Jerry Henson

U.S. CSC

P.S.  I supported decintact about 4 years ago, but this is the first
such issue I've worked since then.  I have forgot just about everything
I ever knew about this product.

============================================================

From:	US3RMC::"[email protected]" 21-FEB-1997 15:00:38.94
To:	<csc32::j_henson>
CC:	
Subj:	C970221-3825  . . . DECintact call



   Hi Jerry:

   Here are the details that we could find.

From sys$manager:operator.log:

%%%%%%%%%%%  OPCOM  21-FEB-1997 13:56:05.20  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
%INTACT-E-TMG_BUGCHK, Terminal Manager Bug at PC=00039FFF R11=003FF750
R10=0000BDD0
         R9=003FEA00 R8=0047C200 R7=00000008 R6=00000000 on ?? :

%%%%%%%%%%%  OPCOM  21-FEB-1997 13:56:05.21  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
-INTACT-E-TMG_INVATBRES, invalid ATB in restore_ctx
         hdr = 00000000 00000004 00000000 00000000 00000988 00000A98


%%%%%%%%%%%  OPCOM  21-FEB-1997 14:21:27.32  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
%INTACT-E-TMG_BUGCHK, Terminal Manager Bug at PC=00039FFF R11=003FED40
R10=0000BDD0
         R9=003FEA00 R8=0047C200 R7=00000008 R6=00000000 on ?? :

%%%%%%%%%%%  OPCOM  21-FEB-1997 14:21:27.32  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
-INTACT-E-TMG_INVATBRES, invalid ATB in restore_ctx
         hdr = 00000000 00000004 00000000 00000000 00000988 00000A98


%%%%%%%%%%%  OPCOM  21-FEB-1997 15:45:04.34  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
%INTACT-E-TMG_BUGCHK, Terminal Manager Bug at PC=00039FFF R11=00407998
R10=0000BDD0
         R9=003FEA00 R8=0047C200 R7=00000008 R6=00000000 on ?? :

%%%%%%%%%%%  OPCOM  21-FEB-1997 15:45:04.35  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
-INTACT-E-TMG_INVATBRES, invalid ATB in restore_ctx
         hdr = 00000000 00000004 00000000 00000000 FFFF7168 00001EB9



%%%%%%%%%%%  OPCOM  21-FEB-1997 15:45:04.34  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
%INTACT-E-TMG_BUGCHK, Terminal Manager Bug at PC=00039FFF R11=00407998
R10=0000BDD0
         R9=003FEA00 R8=0047C200 R7=00000008 R6=00000000 on ?? :

%%%%%%%%%%%  OPCOM  21-FEB-1997 15:45:04.35  %%%%%%%%%%%
Message from user INTACT_PRODA on EQVAX6
-INTACT-E-TMG_INVATBRES, invalid ATB in restore_ctx
         hdr = 00000000 00000004 00000000 00000000 FFFF7168 00001EB9




   On the terminals themselves, one of two errors come up.

  Unexpected Error in Liberace,  Reason Code=44,
      System-F-Abort, abort
      Abort File Name "   ".crs  (blank is highlighted area
                                  that can't be read).

  Unexpected Error in Liberace,  Reason Code=131185098,


fyi

% ====== Internet headers and postmarks (see DECWRL::GATEWAY.DOC) ======
% Received: from mail2.digital.com by us3rmc.pa.dec.com (5.65/rmc-22feb94) id AA25945; Fri, 21 Feb 97 13:50:00 -0800
% Received: from equitable.ca by mail2.digital.com (5.65 EXP 4/12/95 for V3.2/1.0/WV) id AA20572; Fri, 21 Feb 1997 13:41:27 -0800
% Received: by gateway.equitable.ca id <39686>; Fri, 21 Feb 1997 19:13:33 -0500
% Date: Fri, 21 Feb 1997 16:36:59 -0500
% From: [email protected]
% To: <csc32::j_henson>
% X-Vmsmail-To: MX%"[email protected]"
% Subject: C970221-3825  . . . DECintact call
% Message-Id: <[email protected]>

>>       <<< Note 435.0 by CSC32::J_HENSON "Don't get even, get ahead!" >>>
>>                     -< terminal manager bug/ invalid atb >-


Anyone?  I don't mean to be pushy, but I expect to be pushed by my
customer, and need an answer for him.

Thanks,

Jerry

I am looking into this, but bear in mind that you submitted the original
note on Friday night and looked for an update at lunchtime on Monday...

If they've been running with no problems for 2 years it sounds like they
know what they're doing, but ...

Global sections remain mapped in memory even when DECintact is shut down.
To delete a global section properly you have to shut down DECintact and then
use the INTACT$INIGSD utility to delete it. The same is true for modifying
an existing one i.e. you have to delete it first and then create a new one
with the updated information. Section 15.2 of the System Management Guide
describes the procedure for configuring global sections.

Is it possible that they didn't shut down DECintact before trying to delete
the new global section? This is particularly important if the DECintact system
is configured to use system buffers (nonpaged pool) because in this case any
sections that you delete using INTACT$INIGSD will be deleted immediately. This
will make it impossible for INTACT$COPY_CONTROL.COM to stop the DECintact
components and could also cause the copy of DECintact to fail if it continues
to use these sections after they have been deleted. In this case the only
effective cure is to reboot the machine.

I would recommend that they review the global sections they currently have
defined. If there's any doubt the best thing to do would be to delete and
recreate all the global sections following the procedures in section 15.2
of the System Management Guide. This shouldn't be too difficult if they have
followed the recommendations in 15.2.4 and recorded the answers in a command
file. If it were me, I'd reboot the machine as well to make sure all traces
of the old sections are removed.

Joe.

Thanks,

I have asked the customer to reboot, as he claims to have tried the
other suggestions.  He will not be able to reboot until this weekend.
I'll let you know his results.

thanks again,

Jerry

I just received confirmation from the customer that rebooting his system
resolved the problem.

Thanks,

Jerry