Release notes for portingset: 7.1d.03/SUN_SPARC/SOLARIS2.6/PA.2348Release notes 
for portingset: 7.1d.03



MaintCorelli: # 20106 (BDUX13233): ipc_boot distorts bshell arguments

Date: Fri, 26 Sep 2003 14:58:57 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
When the BW is used to start a bshell, and arguments are specified,
ip_boot might distort the arguments.

Problem Description (Technical terms)
ipc_boot eliminates some arguments that BW passes because these are not meant for the bshell.
This is done by using memcpy() on overlapping memory areas. Note that the behavior in this situation is
undefined!
During installation on Itanium, this causes argument corruption, and thus failing install.

Test Procedure
Install on HP-Itanium.




MaintCorelli: # 20100 (BDUX13227): Core dump on NULL dereference

Date: Fri, 26 Sep 2003 09:48:53 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Core dump on many platforms in <rdbms>_maint due to a NULL pointer dereference.

Workaround
When crashes occur in <rdbms>_maint, use the 7.1d.02 version.

Test Procedure
Use ora_maint on SunOS (not on HP)




MaintCorelli: # 20089 (BDNT13215): Fixed bug in BW msi. License file is not 
correct

Date: Thu, 25 Sep 2003 12:25:23 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
License Dialog was not correct in BW MSI..

Problem Description (Technical terms)
See above. Problem caused by migration from IS for MSI pro 1.0 to IS Devloper 7

Workaround
NA

Test Procedure
Run BW setup, and check license dialog.




MaintCorelli: # 20059 (BDUX13183): Merge BDNT13122/BDUX13060

Date: Fri, 19 Sep 2003 15:01:47 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
see BDNT13122 and BDUX13060

Workaround
n/a	

Test Procedure
see BDNT13122 and BDUX13060




MaintCorelli: # 20053 (BDNT13159): Portingset update: update also zoneinfo and 
nlsinf directories

Date: Fri, 19 Sep 2003 14:27:43 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
When updating the portingset the lib\zoneinfo, lib\zonerules and lib\nlsinf directories are not updated.
These directories should be updated as well. Because sometimes this information is updated.

Problem Description (Technical terms)
Installer does not install these directories.

Test Procedure
Before starting updating the portingset, empty the directories lib\zoneinfo, lib\zonerules and lib\nlsinf.
Update the portingset by running the setup.
When setup is finished, the mentioned directories should be filled again.




MaintCorelli: # 20046 (BDUX13179): Removed X11R5 references. BX is not available 
on ERP

Date: Fri, 19 Sep 2003 09:31:59 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Do not include X11 libs. It may result in confusion.

Test Procedure
NA, X11 was not used anymore




MaintCorelli: # 20038 (BDUX13167): core dump on bshellxma solution caused 
bshellxma to take all resources from system ( SUN only )

Date: Thu, 18 Sep 2003 10:36:56 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
bshellxma was bringing the complete system down after a fix for a core dump, and enabling all trace options 

Problem Description (Technical terms)
Problem after a correct fix for a core dump problem. the bshellxma was taking the complete system down. this was only for a SUN OS system, on other systems no problem was seen

Workaround
Disable all trace options

Test Procedure
for testing a script should be created, test procedures with the correct script are mentioned in the defect. check that OS you connect to is SUN, and tracing on that system is switched on to max.


Motive source
DF:132229




MaintCorelli: # 20024 (BDUX13155): Changed Names/Projects

Date: Tue, 16 Sep 2003 12:04:18 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Changed names




MaintCorelli: # 20017 (BDNT13152): Updated bison from version 1.35-2 to 1.35-4

Date: Tue, 16 Sep 2003 11:57:57 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Bison did not work on some NT4 platforms. 

Workaround
NA

Test Procedure
Try to build blat_temp.y from al_1 library on a NT4 system.




MaintCorelli: # 20003 (BDUX13137): Fixed compilation error on AIX and AS/400

Date: Mon, 15 Sep 2003 09:55:31 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Fixed compilation errors in al_alloc.c. Caused by missing initializer.




MaintCorelli: # 19997 (BDNT13133): Fixed compilation error

Date: Fri, 12 Sep 2003 16:41:07 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Fixed compilation error, missing include file

Workaround
NA

Test Procedure
Build bshellxma




MaintCorelli: # 19994 (BDNT13130): Re-added log message, but also better test on 
setting vrcpath resources (merge from BDNT13088)

Date: Fri, 12 Sep 2003 16:33:32 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Re-added log message for max_vrc entries, together with a better test, since log message is still useful.

Workaround
NA

Test Procedure
Very difficult to test, but try next:
- start Shared memory on windows
- start shared memory manager, and press load button -> no message in event viewer should appear about max_vrc entries
- change u<user> file (add or change a vrcpath entry)
- press again on load button in SHM manager -> message should appear that either vrc entries exceeded max, or changed vrc path




MaintCorelli: # 19987 (BDUX13001): 2nd shmtimer is started

Date: Fri, 12 Sep 2003 15:26:46 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Customer terms)
After starting a Baan environment with "shmmanager6.2 -i" a shmtimer6.2 is
running.
That running shmtimer6.2 has the UID of the user that run "shmmanager6.2 -i".
When that user run "shmtimer6.2 -i" it get the message:
 "shmtimer6.2: Timer already running"
and no new shmtimer6.2 is started.

When another user run "shmtimer6.2 -i" it get the message:
 "shmtimer6.2: Cannot access shmtimer (pid 1738) - error 1 (Not owner)"
 "shmtimer6.2 started: pid = 1740, time = 1050065601 (Fri Apr 11 12:53:21 2003)"
And a new shmtimer6.2 is started.

Also another user should not be possible to start a second shmtimer6.2.


Problem Description (Technical terms)
Test on running shmtimer was wrong.

Workaround
Do not start 2nd shmtimer.

Test Procedure
See problem description

Motive source
TCS:100-131790
TCS:100-131325




MaintCorelli: # 19982 (BDUX12474): Call Graph Profiler backport

Date: Fri, 12 Sep 2003 14:36:09 +0200

Created on:  DevCorelli
Type:        bugfix

Problem Description (Customer terms)
Analysing performance problems can be very difficult. The Call Graph Profiler 
is a new tool that shows a relation between functions, queries and other db 
activities using an HTML output format
Also Profiling 3GL is almost impossible as it requires profiled objects.
This no longer needed.

Problem Description (Technical terms)
idem

Workaround
not really a workaround

Test Procedure
See the manual
Start a bshell with -set PROFILE_ALL=1 -set PROF_DIR=<directory-for-the-output>
Start any session and close it. In the <directory-for-the-output> HTML files
are found for further inspection




MaintCorelli: # 19954 (BDUX13024): AS400: Add strsrvdbg.qclsrc to clearcase

Date: Thu, 11 Sep 2003 12:23:40 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Add the source for the program starting the AS400 debugger. Unfortunately it is not complete.




MaintCorelli: # 19953 (BDNT13050): Fixed bshell compilation warnings on Windows 
platform

Date: Wed, 10 Sep 2003 16:46:40 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Fixed lot of compilation warnings in bshell on Windows, and a few on Unix.

Workaround
NA




MaintCorelli: # 19948 (BDNT13078): Added bdbreconfig to dependency list

Date: Wed, 10 Sep 2003 13:40:37 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
bdbreconfig was not build anymore on Windows, due to removal (by accident) of dependency

Workaround
NA




MaintCorelli: # 19925 (BDNT13041): Added object section in ini file for 
performance counters

Date: Mon, 8 Sep 2003 15:50:33 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Customer terms)
During installation on Windows, two messages in the in the event viewer were posted about a suggestion to improve performance for 
Windows performance counters (application log, category perflib). This projects prevents that previous messages are logged.

Problem Description (Technical terms)
Missing section in ini file. Format of file seems to be changed by Microsoft.

Workaround
NA. Warning in event viewer can safely be ignored.

Test Procedure
Run installer and check that no informational messages about performance in the perflib are displayed.




MaintCorelli: # 19924 (BDNT13047): Do not reconnect BW in case of Automation

Date: Mon, 8 Sep 2003 13:53:46 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
When Worktop is used and the connection with the bshell aborts (for instance because the Bshell crashes), 
BW will give the user a chance to reconnect. However the Worktop will also beform some retries. Therefore
BW should not reconnect in case the Worktop is used.


Problem Description (Technical terms)
The fact that BW was started by the Worktop is determined by the fact that BW was 
started through Automation. In this case BW will only give a message to tell the user that the connection 
with the Bshell was broken, however the user will not have a chance to reconnect.


Workaround
Press "NO" when the question comes to reconnect.


Test Procedure
1)
Login Baan using Worktop
Unplug the network cable of your client computer running Worktop
Wait till a message pops up "Connection with server lost"
Press OK
Plug in the network cable again.
Doubleclik on a session in Worktop. 
Now Worktop should start the session, and not hang.

2)
Login Baan using Worktop
Unplug the network cable of your client computer running Worktop
Wait till a message pops up "Connection with server lost"
Press OK
Doubleclik on a session in Worktop. 
Plug in the network cable again.
Now Worktop should start the session, and not hang.


Note that for the last scenario BW can generate errors like "Cannot setup a connection with the host".
After clicking on the OK button to remove these messages, Worktop should be still able to establesh a new connection and start the session.


Motive source
DF:139607




MaintCorelli: # 19913 (BDNT13030): 2003 Certificate from Verisign for BI

Date: Fri, 5 Sep 2003 14:20:55 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Yearly update of certificate for BI Applet required.

Problem Description (Technical terms)
The current certificate was expired. This prevents building a new BI version.
A new certificate from Verisign must be installed. Note: also the password for
signing has been changed from "baanbi" to "baan".


Workaround
Normal usage of BI can still continue.
New BI builds cannot be made.




MaintCorelli: # 19907 (BDNT12530): Shared memory leak

Date: Fri, 5 Sep 2003 13:18:15 +0200

Created on:  MaintReger
Type:        bugfix

Problem Description (Customer terms)
Shared memory keeps on growing when using the Crystal Reports integration.

Problem Description (Technical terms)
Crystal is generating temporary BSE environments, each time with a random name.
This causes the DB driver cache in shared memory to be filled.

The fix was initially made for Unix environments and ported to Windows. The
Windows Oracle driver is still using shared memory, because the shared memory
was disabled to late. The original fix for this problem was made in BDUX11861.

Test Procedure
This problem can only be reproduced on a Windows Baan environment. The shared
memory leak does not occur on Unix systems. 

Make sure that you are the only user on the Windows system. 

Connect once using Baan Windows or Worktop to the system, because the shared
memory leak did not occur after a rebooted system. 

Use Crystal reports to connect to the Baan environment. Check the amount
of memory before connect and afterwards. The shared memory usage should not
change at all  with the new driver. 

Motive source
DF:139779




MaintCorelli: # 19904 (BDUX13021): AS400: update qpc_yy_aix.c and qp_token_aix.h 
after BDUX10007

Date: Thu, 4 Sep 2003 10:16:42 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
GMAKE[4]: Entering directory `/home/port/MaintCorelli/lib/qp'
Yacc source /view/MaintCorelli.night/vobs/tt/lib/qp/qc_qpc.y has been modified
Please, update /view/MaintCorelli.night/vobs/tt/lib/qp/qpc_yy_aix.c
GMAKE[4]: [/view/MaintCorelli.night/vobs/tt/lib/qp/qpc_yy_aix.c] Error 1 (ignored)
Yacc source /view/MaintCorelli.night/vobs/tt/lib/qp/qc_qpc.y has been modified
Please, update /view/MaintCorelli.night/vobs/tt/lib/qp/qp_token_aix.h
GMAKE[4]: [/view/MaintCorelli.night/vobs/tt/lib/qp/qp_token_aix.h] Error 1 (ignored)

Workaround
Update qp_token_aix.h and qpc_yy_aix.c`

Test Procedure
Nightly build




MaintCorelli: # 19899 (BDUX13017): Remove alloc.mem soft memory limit message

Date: Wed, 3 Sep 2003 16:52:18 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
alloc.mem calls above a certain memory limit produces an error message.
The alloc.mem function does return 0 indicating that the call to alloc.mem
is succesful


Problem Description (Technical terms)
idem

Workaround
Ignore the error message. The program continues to execute.

Test Procedure
function main()
{
	string aaa(1) based

	alloc.mem(aaa, 31*1024*1024)
}

will not give an error message.

Motive source
DF:214622




MaintCorelli: # 19892 (BDUX13014): Blat: transaction duration in db drivers too 
short

Date: Wed, 3 Sep 2003 14:50:31 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
The commit/abort time as generated by BLAT in the database drivers are too short (big gap with
bshell times). The real commit/abort time in the RDBMS is missing. Only the post processing in the drivers
was measured.

Test Procedure
Run some session with BLAT trans logging enabled (in db driver!).
Compare the timings. The new version gives longer durations for the commit/abort in the database driver.




MaintCorelli: # 19872 (BDNT12821): Manual merge of BDNT12989

Date: Tue, 2 Sep 2003 09:44:46 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
In certain circumstances it can be very handy to know from which client computer a login request came.
For example when you want to know from which client computer the Baan login is made.
In case of Windows os this it is not possible to determine this. 
(On unix the name of the client computer is part of the argv of the bshell process).

Problem Description (Technical terms)
Need to add new functionality to log this information.

Workaround
Use netstat. But does not show you the time the user connected and if it concerns a Baan Login.

Test Procedure
1. To enable the logging, modify the Windows registry as follows:
	- Open the Windows registry using regedit.exe
	- Open key HKEY_LOCAL_MACHINE\SOFTWARE\Baan\Services
	- Select from the menubar: Edit: New -> DWORD Value
	- Type in the name of this value: RexecLogClient, and press Enter
	- Double-click this value, to open the properties
	- Fill in the Value data: 1, and press OK
	- Exit regedit.exe
	
2. Restart the Baan Logic Service:
	- Open the Baan NT Manager 
	- Right-click the entry "Logic server", and select "Stop" from the drop-down menu.
	The Output window should display the following text now: "Logic service stopped"
	- Right-click the entry "Logic service" again, and select "Start" from the drop-down menu
	The Output window should display the following text now: "Logic service started"

3. Log in Baan, and check Event log.
	- the following logmessage should be logged, for example:

	
Env	 Services (C:\WINNT\Baan)
Prog	 rexecd file \general\nw\blogind.c # 1922
Keyword	 BaanLoginDoLogClient()
Username	 baan	type S	language 2
Process	 0x6b0
Lasterror	 0
Errno	 0
bdberrno	 0
Message	 2003-08-29[15:18:15]: User 'kvddool' login request from '10.21.111.33'


Note: Only for SSPI protocol the username is empty.





MaintCorelli: # 19862 (BDUX12991): Merged DevCorelli changes to MaintCorelli

Date: Fri, 29 Aug 2003 09:52:30 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Only a few files were different, trying to remove DevCorelli release




MaintCorelli: # 19851 (BDUX12966): mb.display function problem

Date: Thu, 28 Aug 2003 11:22:11 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
A 3gl program has been written that reproduces the problem found in repdll. The
problem is with mb.display(), but until that is fixed we can probably work
around the problem in repdll by just truncating the string coming back from
mb.display() to the length it is supposed to be. What appears to happen is if
the string is ended with a tab character immediately AFTER the last character
that would fit in the display, it pads out the rest of the string with the tab
character. Below is the script:

function main()
{
    long fd
    string var(255)
    string long_str(255)

    var = "01234567890123456789         "
    fd = mb.display(var, long_str, 20)
}

Problem Description (Technical terms)
Debugger shows following value for long_str after mb.display().

long_str = "01234567890123456789^I^I^I^I^I^I^I^I^I^I^I^I^..."


Function mb.display() will place any escape strings or code features which maybe
present at the end of string into the target. It includes TAB characters also.
This should not happend.

Workaround
n/a

Test Procedure
Run following test program and check the value of long_str

function main()
{
    long fd
    string var(255)
    string long_str(255)

    var = "01234567890123456789         " | TAB characters at the end
    fd = mb.display(var, long_str, 20)
}

check value of long_str. It should be

long_str = "01234567890123456789"

Motive source
DF:215793




MaintCorelli: # 19846 (BDUX12983): AS400: Bug in al_sli.c when calling <srvpgm 
name>_srvpgm_init

Date: Thu, 28 Aug 2003 10:03:47 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
On AS400, using dlopen400 to load a .SRVPGM (shared library) and trying to run the initial procedure <srvpgm name>_srvpgm_init could result in the error "qsh: 001-0079 Process ended by exception 9901".

Workaround
Do not use "for(; *p != 0x00; *p = tolower(*(p++)));"

Test Procedure
Nightly build




MaintCorelli: # 19843 (BDUX10007): BaanSQL: Timestamp literal

Date: Wed, 27 Aug 2003 16:42:54 +0200

Created on:  DevCorelli
Type:        enhancement

Problem Description (Customer terms)
This project implements the following new date/time features to BaanSQL:

1. date literal
   For example: DATE '2002-12-31'

2. timestamp literal
   For example: TIMESTAMP '2002-12-31 13:32:45'
   Date and time are specified in local time, but internally
   represented as UTC.

3. current_date function

4. current_timestamp function

All these features comply with the ANSI SQL standard.


Motive source
IRD:58




MaintCorelli: # 19840 (BDUX12979): Bshell debugger history fix

Date: Wed, 27 Aug 2003 16:34:26 +0200

Created on:  MaintReger
Type:        bugfix

Problem Description (Technical terms)
If a command were used from the history stack position 0, it was saved again.

Motive source
TCS:800-133513




MaintCorelli: # 19835 (BDNT12890): Bwprint crashs on Windows 9x

Date: Wed, 27 Aug 2003 15:11:52 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
When printing to Bwprint, and when having no valid printers in your windows printer setup, Bwprint generates an Access Violation and crashes.

Problem Description (Technical terms)
The access violation occurs when function PrepareDevMode wants to log a message (using log_mesg) because there is no valid printer available.
Function log_mesg generates the access violation, because tss locale is not initialized yet. (log_mesg calls function TSStsswcs2extmbs).
Function PrepareDevMode is called from function PrepareDevmodeAndUnlock, and PrepareDevmodeAndUnlock is called from OnInitialUpdate.

Workaround
Have a printer in your Windows printer setup.

Test Procedure
On Windows98: Take care that there are no printers installed.
Print to Bwprint (turn on Preview mode).
Now, Bwprint should generate the message "There is currently no valid printer in your print setup.", but should not crash.

Motive source
DF:139572




DevCorelli: # 19795 (BDUX11556): Data validation tool (backport)

Date: Fri, 22 Aug 2003 14:14:53 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Long down time during migrations

Problem Description (Technical terms)
Domain validations during/after migrations are now done by (ab)using bdbreconfig.
A table is exported and imported to use the domain check during import.

A new tool (bdbvalidate) does the domain check by reading the database.




MaintCorelli: # 19794 (BDUX11556): Data validation tool (backport)

Date: Fri, 22 Aug 2003 14:14:05

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Long down time during migrations

Problem Description (Technical terms)
Domain validations during/after migrations are now done by (ab)using bdbreconfig.
A table is exported and imported to use the domain check during import.

A new tool (bdbvalidate) does the domain check by reading the database.




MaintCorelli: # 19791 (BDUX12926): typo in Project.all

Date: Fri, 22 Aug 2003 10:27:04 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
compilation error

Test Procedure
compile




MaintCorelli: # 19783 (BDNT12780): BW does not start URL

Date: Thu, 21 Aug 2003 17:00:48 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
When an URL is defined starting for example \\VIGO (servername), then this creates an error, 
because the \\ is converted to // somehow in the process.
Customer uses this URL to point to intranet pages with additional help information.


Problem Description (Technical terms)
In case a htm(l) file is started, the backslashes are converted from backslashes to unix slashes (forward slashes).
This conversion is added in the past (VHNT2682) to prevent problems when using Netscape.
But the Microsoft internet explorer does not accept unix slashes in some cases, unless the standard file tag is used:
an absolute path like "//cnl09659/c$/WINNT/Help/debug/sdbug.htm" will not work when starting it in this way via the internet explorer.
But file:////cnl09659/c$/WINNT/Help/debug/sdbug.htm wil work, and still works also when using Netscape.
This is because the file tag is standarized (the same like http://).
Therefore, in case of an absolute path, start the path with "file://", to make sure it will always work. 


Workaround
type "file://" in front of the absolute path.


Test Procedure
Via session tgbrg5500m000, fill in an URL like \\VIGO\...htm(l), and try to open this URL. 
Refer also to defect 139193 for more info.


Motive source
DF:139193




DevCorelli: # 19782 (BDUX12917): DF215719 - Tax types being ignored

Date: Thu, 21 Aug 2003 17:00:19 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
Although different tax types are selected for processing through tax provider,
only Sales or Override Sales types are being used.

Problem Description (Technical terms)
In deciding which type to assign, individual "case" statements are not
terminated with "breaks".  And so, the last case "Sales" (or override sales)
is always selected.

Test Procedure
Build a new api from the correct library (libtvtx.a or libtboth.a) and test
different types of tax.  All should now work correctly.

Motive source
DF:215719




MaintCorelli: # 19780 (BDUX12917): DF215719 - Tax types being ignored

Date: Thu, 21 Aug 2003 16:58:52 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
Although different tax types are selected for processing through tax provider,
only Sales or Override Sales types are being used.

Problem Description (Technical terms)
In deciding which type to assign, individual "case" statements are not
terminated with "breaks".  And so, the last case "Sales" (or override sales)
is always selected.

Test Procedure
Build a new api from the correct library (libtvtx.a or libtboth.a) and test
different types of tax.  All should now work correctly.

Motive source
DF:215719




MaintCorelli: # 19776 (BDUX12911): Solaris8 build error

Date: Thu, 21 Aug 2003 16:56:05 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
in obj*.SOL8/servers the ORACLE7_2 directory was removed in project BDUX10913 (Solaris 8 port). During nightly build we get an error because ORA7 define is on for Solaris 8.

Test Procedure
Nightly build on rise.




MaintCorelli: # 19734 (BDUX12117): Retained cursors

Date: Fri, 15 Aug 2003 15:09:18 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Technical terms)
Performance improvement with higher retained_cursors default.

Workaround
manual increase 'retained_cursors' resource.




MaintCorelli: # 19722 (BDNT12863): Fix build of bic.exe on NT

Date: Thu, 14 Aug 2003 17:22:33 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Build of bic.exe on NT didn't work because of YYBACKUP not being supported okay by bison.
Now use bic_yy_aix.c and bic_tokes_aix.h for build on NT.

Test Procedure
Copile program script on MS Windows




MaintCorelli: # 19702 (BDNT12795): Error 518 when starting more then 250 
sessions

Date: Wed, 13 Aug 2003 10:28:29 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
When starting to much Baan sessions (or zooming to much times), that is more then about 250 times, 
the following error occurs (And several other errors):

Env	 corelli (d:\corelli)
Prog	 bshell file \db\lib\dbc\C_OPEN.C # 823
Keyword	 Invalid transaction
Username	 kvddool	type S	language 2
Process	 0x107c
ProcessSet	 kvddool@CNL08696:3984
Lasterror	 997 (Overlapped I/O operation is in progress.)
Errno	 0
bdberrno	 518 (Invalid session code specified)
Message	 Invalid transaction id 0 specified.
fied).

Because of this error you are going to think something is wrong in the session code of the Baan session,
and the actual problem is not clear. To save support time in analyzing the problem, a more nice message should appear. 
E.g.: maximum number of session reached.

Problem Description (Technical terms)
To start a new db session, bdb_start_session is done. This function reserves a global transaction,
which appears to the bdb user as a session. The number of transactions is limited by MAX_PROC_SLOTS. 
When starting more the MAX_PROC_SLOTS, no new bdb session can be created.
But there is no error generated informing the enduser that this is the case. 
Also execution of the calling function is not stopped in case  bdb_start_session cannot start a new session. 

Workaround
Do not start more then 250 (zoom)sessions, or use multiple Baan logins.

Test Procedure
Log in Baan and start more then 250 sessions.


Motive source
DF:139257




DevCorelli: # 19639 (BDUX12746): QP: wrong result on retry of WITH RETRY REPEAT 
LAST ROW query

Date: Fri, 1 Aug 2003 16:27:54 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
If a query contains and ORDER BY clause with multiple columns and contains the
WITH RETRY REPEAT LAST ROW clause, then the query processor generates a wrong
"retry expression". The retry expression is used to restrict the result set of 
the query to those rows that are not yet committed.

In case the ORDER BY clause was "ORDER BY A,B,C WITH RETRY REPEAT LAST ROW" the 
query processor generated the following expression:

	(A >= :1) OR (A = :1 AND B >= :2) OR ({A,B} = {:1,:2} AND C >= :3)

This is wrong. It should have generated the following (or functionally equivalent)
expression:

	(A > :1) OR (A = :1 AND B > :2) OR ({A,B} = {:1,:2} AND C >= :3)

Note that only the last (rightmost) expression has the ">=" operator.

Test Procedure
Assume the following table dbtst120, sorted by (sex, workdept, empno):

	sex  workdept  empno  ...
	---  --------  -----  
	1    'A00   '  110 
	1    'A00   '  120 
	1    'B01   '  20 
	1    'D11   '  60 
	1    'D11   '  150 
	1    'D11   '  170 
	.                     .
	.                      .
	.                       .


Execute the following command:

	qptool6.2 -a peffffsffrff -q "
		select sex,workdept,empno 
		from dbtst120
		order by sex,workdept,empno with retry repeat last row
	"

Explanation: We parse ('p') and execute ('e') the query. We fetch the first 
four rows ('ffff') (upto 1,'D11',60) and then save these values as the retry 
values ('s') (the bshell does that automatically during the 
commit.transaction()). Then the next two values are fetched ('ff'). Next, a 
retry ('r') is done, followed by two fetches ('ff'), of which the first should 
return the row (1,'D11',60).

This should give the following result:

	sex  workdept  empno   Comment
	---  --------  -----   -------------------------------
	1    'A00   '  110 
	1    'A00   '  120 
	1    'B01   '  20 
	1    'D11   '  60      Saved retry values
	1    'D11   '  150 
	1    'D11   '  170 
	1    'D11   '  60      Values returned after retry
	1    'D11   '  150 

Just for reference, before the fix the result would be:

	sex  workdept  empno   Comment
	---  --------  -----   -------------------------------
	1    'A00   '  110 
	1    'A00   '  120 
	1    'B01   '  20 
	1    'D11   '  60      Saved retry values
	1    'D11   '  150 
	1    'D11   '  170 
	1    'A00   '  110     Values returned after retry
	1    'A00   '  120 

Note that the values returned after the retry are incorrect.

Motive source
DF:137839




MaintCorelli: # 19638 (BDUX12746): QP: wrong result on retry of WITH RETRY 
REPEAT LAST ROW query

Date: Fri, 1 Aug 2003 16:12:16 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
If a query contains and ORDER BY clause with multiple columns and contains the
WITH RETRY REPEAT LAST ROW clause, then the query processor generates a wrong
"retry expression". The retry expression is used to restrict the result set of 
the query to those rows that are not yet committed.

In case the ORDER BY clause was "ORDER BY A,B,C WITH RETRY REPEAT LAST ROW" the 
query processor generated the following expression:

	(A >= :1) OR (A = :1 AND B >= :2) OR ({A,B} = {:1,:2} AND C >= :3)

This is wrong. It should have generated the following (or functionally equivalent)
expression:

	(A > :1) OR (A = :1 AND B > :2) OR ({A,B} = {:1,:2} AND C >= :3)

Note that only the last (rightmost) expression has the ">=" operator.

Test Procedure
Assume the following table dbtst120, sorted by (sex, workdept, empno):

	sex  workdept  empno  ...
	---  --------  -----  
	1    'A00   '  110 
	1    'A00   '  120 
	1    'B01   '  20 
	1    'D11   '  60 
	1    'D11   '  150 
	1    'D11   '  170 
	.                     .
	.                      .
	.                       .


Execute the following command:

	qptool6.2 -a peffffsffrff -q "
		select sex,workdept,empno 
		from dbtst120
		order by sex,workdept,empno with retry repeat last row
	"

Explanation: We parse ('p') and execute ('e') the query. We fetch the first 
four rows ('ffff') (upto 1,'D11',60) and then save these values as the retry 
values ('s') (the bshell does that automatically during the 
commit.transaction()). Then the next two values are fetched ('ff'). Next, a 
retry ('r') is done, followed by two fetches ('ff'), of which the first should 
return the row (1,'D11',60).

This should give the following result:

	sex  workdept  empno   Comment
	---  --------  -----   -------------------------------
	1    'A00   '  110 
	1    'A00   '  120 
	1    'B01   '  20 
	1    'D11   '  60      Saved retry values
	1    'D11   '  150 
	1    'D11   '  170 
	1    'D11   '  60      Values returned after retry
	1    'D11   '  150 

Just for reference, before the fix the result would be:

	sex  workdept  empno   Comment
	---  --------  -----   -------------------------------
	1    'A00   '  110 
	1    'A00   '  120 
	1    'B01   '  20 
	1    'D11   '  60      Saved retry values
	1    'D11   '  150 
	1    'D11   '  170 
	1    'A00   '  110     Values returned after retry
	1    'A00   '  120 

Note that the values returned after the retry are incorrect.

Motive source
DF:137839




DevCorelli: # 19629 (BDNT12768): Fixed bug in deliver if destination contains 
variables, like

Date: Fri, 1 Aug 2003 13:39:45 +0200

Created on:  MaintGemini
Type:        porting specific change

Problem Description (Technical terms)
Delivery utility on Windows doesn't fill content.txt file if delivery is in a directory containing a variable, like $PORTINGNR

Workaround
NA

Test Procedure
Run deliver to a directory with a variable in deliver.cfg. 
Check that $DEST\content.txt is filled.




MaintCorelli: # 19628 (BDNT12768): Fixed bug in deliver if destination contains 
variables, like

Date: Fri, 1 Aug 2003 13:38:10 +0200

Created on:  MaintGemini
Type:        porting specific change

Problem Description (Technical terms)
Delivery utility on Windows doesn't fill content.txt file if delivery is in a directory containing a variable, like $PORTINGNR

Workaround
NA

Test Procedure
Run deliver to a directory with a variable in deliver.cfg. 
Check that $DEST\content.txt is filled.




MaintCorelli: # 19622 (BDUX12715): Incorrect returnvalue of bdbpost when using 
patterns

Date: Fri, 1 Aug 2003 12:25:19 +0200

Created on:  MaintReger
Type:        bugfix

Problem Description (Customer terms)
When importing tables using bdbpost in the following way, the return value of bdbpost is not 0, while
there were no errors. In case there are no errors, the return value of bdbpost should be 0.
Because the return value is not 0, the session reports that something went wrong.

5.2> bdbpost6.2 -pb52a_own -kfmn -D. -t"|" -c998 "ttfgld005*"
Table        M  Idx  RowCount  Messages              
tfgld005998       3         0  No rows in table
			       Adding indexes ...
5.2> echo $?
32


Problem Description (Technical terms)
The variable ret_val got the non-initialized value of ret_tmp assigned. 
This should not happen.

Workaround
Run bdbpost and ignore the returnvalue.

Test Procedure
Run bdbpost as follows:
bdbpost6.2 -pb52a_own -kfmn -D. -t"|" -c998 "ttfgld005*"

(this is just an example, but it is important to use pattern)
after running do:

echo $?

this should return 0, in case there were no errors.

Motive source
DF:134269




MaintCorelli: # 19616 (BDNT12735): access violation when deleting tables

Date: Fri, 1 Aug 2003 12:22:35 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
When running session ttaad4231m000 (Delete Tables) for a second time, the bshell generates
and access violation and the connection is lost.


Problem Description (Technical terms)
In case a tabledefinition is not present, no hash_unlink was done.
Therefore, when this tabledefinition was needed a second time, an incomplete structure of the tables was used.
This caused the access violation.


Workaround
Take care that all tabledefinition are really present.


Test Procedure
The problem is in general reproducable when using a non-existing table definition for a second time.
This problem does only occur on Windows platforms.

The testscenario as described in the Problem description can be used, as well as the following test script:

function main()
{
        error.bypass = 1
        sql.parse("select * from cfcec400")
        sql.parse("select * from cfcec400")
}

The bshell should not generate an access violation when the second sql.parse is done.


Motive source
DF:138921




MaintCorelli: # 19593 (BDUX12295): AS400 : Unnecessary debug information 
(Compiler option) changed

Date: Thu, 31 Jul 2003 11:02:08 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Debug binary has expansion of Macro's and inclusion of header files, which 
makes the debugging process painful.

Test Procedure
Build debug binary and try to debug it.




MaintCorelli: # 19580 (BDUX12697): Lock problem in logging due to jvm thread.

Date: Wed, 30 Jul 2003 15:51:03 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Technical terms)
The bshell would go exit when the option -dbgjvmi is used and after 5 times, a thread could not set a file lock on the log file. 


Workaround
The abort in the al_log.c file has been removed.

Test Procedure
Web defect 8153




DevCorelli: # 19574 (BDUX12712): Level-1 DB drivers: Second update on same row 
is not executed on RDBMS

Date: Tue, 29 Jul 2003 16:45:42 +0200

Created on:  DevCorelli
Type:        bugfix

Problem Description (Customer terms)
In certain scenarios, data corruption can occur. To be more precise,
if a row is updated twice in succession within the same transaction,
the second update was not executed on the database.

This problem is introduced in porting set 7.1d.02, and was not in
7.1d.01.

Problem Description (Technical terms)
See also Solution of MaintCorelli: # 16570 (BDUX9888): DB: Error 201 on 
delete after update:

	"If, within the same transaction, an update is followed by another 
	update or delete, then for the second update or delete we do not 
	check whether another user has modified the record. This is possible, 
	because after the first update we hold a database lock and no other 
	user can change the record."

This is still true, however the actual update should still be performed.

Test Procedure
Execute the following gcommand script. It should complete without errors.
It tests both for the "delete after update" and "update after update"
scenarios.

	*
	* Insert new row (empno=21,salary=1.4)
	*
	dbtst120 RINS 0 -i
	= dbtst120.empno = 21
	= dbtst120.salary = 1.4
	= dbtst120.workdept = A00
	*
	* Put a delayed lock on (empno=21,salary=1.4)
	*
	dbtst120 REQUAL# 0 -i
	= dbtst120.empno = 21
	*
	* Update salary from 1.4 to 1.3
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 1.3
	dbtst120 REQUAL # 0 -c
	c dbtst120.salary = 1.3
	*
	* Update salary from 1.3 to 1.0
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 1.0
	dbtst120 REQUAL # 0 -c
	c dbtst120.salary = 1.0
	*
	* Update salary from 1.0 to 1.4
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 1.4
	dbtst120 REQUAL # 0 -c
	c dbtst120.salary = 1.4
	*
	* Update the salary from 1.4 to 2.2204460492503131e-16
	* The value in the Delayed Lock Buffer (DLB) is updated to
	* this new value.
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 2.2204460492503131e-16
	*
	* Delete the row. The DLB value (2.2204460492503131e-16) was 
	* compared with a number close to it, resulting in error 201.
	*
	dbtst120 RDEL% 0
	*
	* Verify that the row is deleted (results in flush of the delete).
	*
	dbtst120 REQUAL 0 -i -E NOREC
	= dbtst120.empno = 21
	*
	* Abort transaction (removes row empno=21)
	*
	dbtst120 DABORT 0


Motive source
DF:129680




MaintCorelli: # 19573 (BDUX12712): Level-1 DB drivers: Second update on same row 
is not executed on RDBMS

Date: Tue, 29 Jul 2003 16:44:02 +0200

Created on:  DevCorelli
Type:        bugfix

Problem Description (Customer terms)
In certain scenarios, data corruption can occur. To be more precise,
if a row is updated twice in succession within the same transaction,
the second update was not executed on the database.

This problem is introduced in porting set 7.1d.02, and was not in
7.1d.01.

Problem Description (Technical terms)
See also Solution of MaintCorelli: # 16570 (BDUX9888): DB: Error 201 on 
delete after update:

	"If, within the same transaction, an update is followed by another 
	update or delete, then for the second update or delete we do not 
	check whether another user has modified the record. This is possible, 
	because after the first update we hold a database lock and no other 
	user can change the record."

This is still true, however the actual update should still be performed.

Test Procedure
Execute the following gcommand script. It should complete without errors.
It tests both for the "delete after update" and "update after update"
scenarios.

	*
	* Insert new row (empno=21,salary=1.4)
	*
	dbtst120 RINS 0 -i
	= dbtst120.empno = 21
	= dbtst120.salary = 1.4
	= dbtst120.workdept = A00
	*
	* Put a delayed lock on (empno=21,salary=1.4)
	*
	dbtst120 REQUAL# 0 -i
	= dbtst120.empno = 21
	*
	* Update salary from 1.4 to 1.3
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 1.3
	dbtst120 REQUAL # 0 -c
	c dbtst120.salary = 1.3
	*
	* Update salary from 1.3 to 1.0
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 1.0
	dbtst120 REQUAL # 0 -c
	c dbtst120.salary = 1.0
	*
	* Update salary from 1.0 to 1.4
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 1.4
	dbtst120 REQUAL # 0 -c
	c dbtst120.salary = 1.4
	*
	* Update the salary from 1.4 to 2.2204460492503131e-16
	* The value in the Delayed Lock Buffer (DLB) is updated to
	* this new value.
	*
	dbtst120 RUPD% 0 -i
	= dbtst120.salary = 2.2204460492503131e-16
	*
	* Delete the row. The DLB value (2.2204460492503131e-16) was 
	* compared with a number close to it, resulting in error 201.
	*
	dbtst120 RDEL% 0
	*
	* Verify that the row is deleted (results in flush of the delete).
	*
	dbtst120 REQUAL 0 -i -E NOREC
	= dbtst120.empno = 21
	*
	* Abort transaction (removes row empno=21)
	*
	dbtst120 DABORT 0


Motive source
DF:129680




MaintCorelli: # 19567 (BDUX12585): Fast reconfig: Database drivers dumps ccore 
in dbsShmInvalidateDbsDD

Date: Mon, 28 Jul 2003 16:36:49 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Customer terms)
The database driver dumps a core when using bdbreconfig. This is the
case for an update from Reger to Gemini for example. The problem
occurs if the shared memory built in the Reger environment is not
restarted before the update.

Problem Description (Technical terms)
The layout of the shared memory for the database driver cache changed
from portingset 7.3a.01 to 7.3a.02. All functions, except dbsShmInvalidateDbsDD
take care of this.

Workaround
Restart the shared memory before upgrading.

Test Procedure
On a 7.3a.01 enviroment, use a 7.3a.02 (or later) bdbreconfig and database 
driver to reconfigure a table. Without this fix, the database driver dumps a
core while attempting to invalidate the DD in shared memory. With this fix,
the driver should detect the incompatibility and should not attempt to 
invalidate the DD in shared memory.




MaintCorelli: # 19539 (BDUX12625): tmboaserver failed to execute SQL query calls 
in weird circumstances

Date: Thu, 24 Jul 2003 13:11:05 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Customer terms)
tmboaserver failed to execute SQL queries in case 
tmboaserver was started again.

Problem Description (Technical terms)
The QP layer closes all queries that were executed in
case a session that executed them was closed. However
the XML query layer assumed the queries were still present.

Test Procedure
Start tmboaserver on Gemini
fire a query (e.g. by importing metadata through BOI Studio)
stop tmboaserver session
start tmboaserver
fire the same query again.




MaintCorelli: # 19531 (BDUX12677): ReliantUNIX: add vers_ident to 
dummy_compdate_force_loader_stuff

Date: Thu, 24 Jul 2003 11:50:15 +0200

Created on:  MaintBaanIVc
Type:        porting specific change

Problem Description (Customer terms)
On ReliantUNIX, the compiler removes strings with "@(#)(B1 ) Prod. ver.  :" and "@(#)(B2 ) Target   :" from executables due to optimalisation. These strings are used to quickly determine the version of the execuatbles.

Problem Description (Technical terms)
Because no reference was done to the vers_ident variable, the version string was removed by optimalisation.

Workaround
Add reference to the variable to fool the compiler.

Test Procedure
Nightly build, what bshell6.2 | grep '@(#)(B' on ReliantUNIX and check for strings.




MaintCorelli: # 19523 (BDUX12673): Refint: increase number of max companies

Date: Thu, 24 Jul 2003 11:04:54 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
While running refint (refint6.1 -pP_EUR_BV -I13H22M05S -c -r -C450), refint core dumps.
Using gdb, a backtrace is retrieved from the core:
#0  0x00025764 in get_included_compnrs ()
#1  0x00025838 in build_new_cnt ()
#2  0x0002567c in do_actions ()
#3  0x000254ac in do_on_child_table ()

While running a debugable refint, refint does not core dump anymore, but the output shows the following error:
[1] Syntax error: ''selec'' not expected
[0] Found 1 errors during scanning/parsing .. Stopped.
sql_parse error 302 Query ''selec from tccom737 a''

Problem Description (Technical terms)
An internal array no_compnrs[] is to small to hold the number of companies.

Workaround
The compnr file should not contain more then 200 different companies (logical companies).

Test Procedure
The fix is tested at customer system, because we did not have the same setup on an internal system.

Code change Type
bugfix

Motive source
DF:138897




DevCorelli: # 19507 (BDUX12147): QP: bad performance (Corelli)

Date: Tue, 22 Jul 2003 17:16:24 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
The query processor does not fill the search operator correctly if it
encounters any OR or UNION expressions. This is specific only to the LEVEL1
driver.

Test Procedure
Write a small query containing nested or and save it as file "test_query".
For eg. 
	select _index1 from dbtst180
	where _index1 = {10}
	and (projno between 'MA2000' and 'ZZZZZZ')
	and (emendate > date(1984,11,01) or (emendate = date(1982,11,01) and em>
	
Execute DBSLOG=02000 TT_SQL_TRACE=04000 qptool6.2 -f test_query
Verify that the trace pertaining to the SrchOper looks like :
	...
	Field empno:
    Equal: [=] Bind (Col empno) Value: '10'
Flag: 04
    Field projno:
    Lower: [>=] Bind (Col projno) Value: 'MA2000'
Flag: 04
    Upper: [<=] Bind (Col projno) Value: 'ZZZZZZ'
Flag: 04
    Field emstdate:
    Field emendate:
	...

Before the fix, the second expression (projno between 'MA2000' and 'ZZZZZZ') was not
present in the SrchOper.
	
Motive source
DF:214029




MaintCorelli: # 19506 (BDUX12147): QP: bad performance (Corelli)

Date: Tue, 22 Jul 2003 17:13:00 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
The query processor does not fill the search operator correctly if it
encounters any OR or UNION expressions. This is specific only to the LEVEL1
driver.

Test Procedure
Write a small query containing nested or and save it as file "test_query".
For eg. 
	select _index1 from dbtst180
	where _index1 = {10}
	and (projno between 'MA2000' and 'ZZZZZZ')
	and (emendate > date(1984,11,01) or (emendate = date(1982,11,01) and em>
	
Execute DBSLOG=02000 TT_SQL_TRACE=04000 qptool6.2 -f test_query
Verify that the trace pertaining to the SrchOper looks like :
	...
	Field empno:
    Equal: [=] Bind (Col empno) Value: '10'
Flag: 04
    Field projno:
    Lower: [>=] Bind (Col projno) Value: 'MA2000'
Flag: 04
    Upper: [<=] Bind (Col projno) Value: 'ZZZZZZ'
Flag: 04
    Field emstdate:
    Field emendate:
	...

Before the fix, the second expression (projno between 'MA2000' and 'ZZZZZZ') was not
present in the SrchOper.
	
Motive source
DF:214029




MaintCorelli: # 19494 (BDUX12646): AS400 : Un-used include file sqlcli removed

Date: Tue, 22 Jul 2003 14:25:48 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Compilation failed in V5R2. SQL_UPDATE & SQL_DELETE typedef's are included as 
system define in SQLCLI system header file. CLI interface is not used in AS400
driver and hence removed unneccessary inclusion of sqlcli.h file.

Test Procedure
Build AS400 driver




MaintCorelli: # 19476 (BDNT12632): Manually ported BDNT12631 to 4c/Corelli

Date: Mon, 21 Jul 2003 16:14:38 +0200

Created on:  MaintBaanIVc
Type:        porting specific change

Problem Description (Technical terms)
Updated script to make generating reports easier (for Peter V).

Workaround
NA




MaintCorelli: # 19469 (BDUX12608): SIGKILL, in system log, after running job.

Date: Mon, 21 Jul 2003 14:26:03 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
A sigkill messages appears in the operating systems logfiles after running a
job. 

Problem Description (Technical terms)
The sigkill messages gives the idea that something has gone wrong while running
the job. 

Workaround
Use IPC_NOSPAWN=1. This causes a fork() i.s.o. a spawn() between ba and bshell).

Test Procedure
1) Start ba6.2 ttaad4100
2) Exit the session
3) Wait for the command line and check the operating system log if a sigkill
message appears. 

The logfile of the os/390 can be checked by connecting to the os/390 system,
using a 3270 emulator. Select s <enter> in the main menu, followed by log
<enter> in the next menu. 

Motive source
DF:137621




MaintCorelli: # 19414 (BDUX12556): AS400: Defintion of RmRecurseCmd is incorrect 
when RmCmd is defined as del

Date: Wed, 16 Jul 2003 09:06:31 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
On AS400 del can be used instead of rm. If the target is a link to a Library, del will also delete the object thelink is pointing to. The del command does not support the -r (recursion) flag so for recursive delete you must use rm -r.

Test Procedure
Nightly build on MaintGemini/DevNewton which use recursive rm.




MaintCorelli: # 19393 (BDUX12529): Delay undefined function message for bshell 
functions until function execution attempt

Date: Mon, 14 Jul 2003 16:05:12 +0200

Created on:  MaintReger
Type:        bugfix

Problem Description (Customer terms)
Bshell programs abort when encountering an undefined bshell function during 
object loading.
It is preferred to delay the process abortion until an attempt to execute
a function. 

Problem Description (Technical terms)
If a function does not exists in function.c the loading of the object aborts
and the process terminates.


Test Procedure
Compile and run the following source. Old bshell's will not even attempt to run
the object. Now a message is displayed when actual one of these functions 
executed.

# 1 "bic_test"
function global long undefined.long.function()
function global string undefined.string.function()
# 5 "ptest"

function long fff()
{
        long bbb

        bbb = undefined.long.function()
        return (bbb)
}

function string ggg()
{
        string bbb

        bbb = undefined.string.function()
        return (bbb)
}


function main()
{

        long a
        string s

        s = ggg()
        a = fff()
}






MaintCorelli: # 19361 (BDUX12506): AS400, V5R2: cd: 001-0009 Error found getting 
information for symbolic link . No such path or directory.

Date: Fri, 11 Jul 2003 15:52:31 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
On OS400 V5R2 the bjvmi.jar was not created correctly.

Test Procedure
Nightly build




MaintCorelli: # 19337 (BDNT12431): Fixed problems with baan MMC manager

Date: Thu, 10 Jul 2003 16:35:15 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Customer terms)
Shared memory could not be configured if the service was not running.

Problem Description (Technical terms)
SHM tried to do a link_shm, which failed, and as a result of this all previous values were restored.
Due to this behavior, it looked like that SHM was not configurable if SHM was not running.
Also solved some other minor problems:
- animation was executed infinitely if an error occurred during starting a service, now limited to 10 seconds
- BSE name was not shown in overall property overview, but the (default) word "BSE"
- a start/stop BSE menu entry was enabled for client-only installs (while totally unuseful)

Workaround
NA

Motive source
TCS:800-324




MaintCorelli: # 19312 (BDUX12448): AS400: Fix #elif IBM_AS400 directives

Date: Thu, 10 Jul 2003 09:48:05 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
On AS400, the pre-processor directive #elif IBM_AS400 does not compute on OS version V5R2 if IBM_AS400 was defined but not set to a 
value (#define IBM_AS400 1).

Workaround
#elif defined(IBM_AS400)

Test Procedure
Nightly build




MaintCorelli: # 19307 (BDUX12461): Several enhancements in reports created for 
build

Date: Wed, 9 Jul 2003 16:42:38 +0200

Created on:  MaintGemini
Type:        porting specific change

Problem Description (Technical terms)
Replaced i386 string with Intel string when sending reports




DevCorelli: # 19297 (BDUX12367): Bic macro arguments limit change

Date: Wed, 9 Jul 2003 11:57:28 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Technical terms)
Bic doesn't made errors when the maximum number of arguments in macro definitions are reached.


Workaround
Use less arguments.

Test Procedure
Create a macro with more than 128 arguments.

Motive source
TCS:106-138597




MaintCorelli: # 19294 (BDUX12367): Bic macro arguments limit change

Date: Wed, 9 Jul 2003 11:08:27 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Technical terms)
Bic doesn't made errors when the maximum number of arguments in macro definitions are reached.


Workaround
Use less arguments.

Test Procedure
Create a macro with more than 128 arguments.

Motive source
TCS:106-138597




MaintCorelli: # 19275 (BDUX12430): DF213570

Date: Tue, 8 Jul 2003 15:31:22 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
When trying to use a separation character other than the default(","), DDC does
work correctly.

Problem Description (Technical terms)
Alternate separation characters have been disabled by a previous change.

Workaround
Do not use an alternate separation character.

Test Procedure
Attempt to use a character other than the default, and verify that the
specified character is now being used in messages.

Motive source
DF:213570




MaintCorelli: # 19244 (BDNT12421): Converted MMC file to MMC1.2 format

Date: Mon, 7 Jul 2003 11:01:14 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Converted baanman.msc to MMC1.2 format.

Workaround
NA (The old format still worked correctly, but MMC asked for conversion when opened)

Test Procedure
Check that a conversion request message do not appear anymore




MaintCorelli: # 19243 (BDUX12415): Merged project BDUX12243 manually

Date: Mon, 7 Jul 2003 10:59:44 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
see project BDUX12243.

Workaround
NA

Test Procedure
see BDUX12243




MaintCorelli: # 19235 (BDUX12395): dirtest failed on OS390

Date: Fri, 4 Jul 2003 16:29:43 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
dirtest failed during NightlyBuild

Test Procedure
NightlyBuild




MaintCorelli: # 19231 (BDUX12406): AS400: al_ebcdic.c errors on V5R2

Date: Fri, 4 Jul 2003 16:26:38 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Module name al_ebcdic has been changed to al_ebcdi53
Pointer not set for location referenced.
Application error.  MCH3601 unmonitored by QCZPXLC at statement 0000000003, instruction X'0000'.

Workaround
Use #include to include extra include files

Test Procedure
Nightly build




MaintCorelli: # 19219 (BDUX12384): DF212605 - DDC background handling

Date: Fri, 4 Jul 2003 08:20:38 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
DDC server stops processing transactions after a while when running in
background mode (i.e. started from ddc driver via ddc_start_bshell.


Problem Description (Technical terms)
While the bshell is running in background mode, data is still being written
to ds_outstream and/or ds_instream.  Since a data is not being picked up
off of these, the limit is reached on that stream and no more processing is
done.

Workaround
Run DDC in foreground mode.

Test Procedure
Ran ddc in background mode processing many more transactions than were
previously processed.

Motive source
DF:212605




MaintCorelli: # 19213 (BDNT12386): BwPrint: avoid MFC messagebox in silent mode

Date: Thu, 3 Jul 2003 16:41:18 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Technical terms)
In silent mode, it must be avoided that BwPrint hangs in an MFC MessageBox




MaintCorelli: # 19203 (BDNT12378): BwPrint: merge from BDNT12353 (BwPrint silent 
mode)

Date: Thu, 3 Jul 2003 13:34:20 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Technical terms)
add silent mode, which replaces messageboxes by logged events




MaintCorelli: # 19189 (BDNT12342): BGAS Win98 problems

Date: Wed, 2 Jul 2003 12:24:52 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Baan Automation Server problem on Windows 98 platform.

Problem Description (Technical terms)
Last File Size not properly set, could be overwritten with next request.




MaintCorelli: # 19158 (BDNT12186): BGAS w98 problem

Date: Tue, 1 Jul 2003 09:03:54 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Baan Automation Server gives problem when run on Windows 98 platform.
Session start hanging at random times.

Problem Description (Technical terms)
Requests send to the BAS client software were not recognized. Win98 has
different behaviour regarding file times.

Test Procedure
Use ODM to see item attachments on a Win98 client system.

Motive source
DF:137785




MaintCorelli: # 19147 (BDUX12320): AS400: Corelli V5R2

Date: Mon, 30 Jun 2003 15:14:49 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
OS400 version V5R2 does not like the icc compiler front-end option "-qOUTPUT=" without "*NONE" or "*PRINT".

Workaround
Use our own front-end program acc to fill in -qOUTPUT=*NONE when the environment variable GENSPLF is not set.

Test Procedure
nightly build




MaintCorelli: # 19118 (BDUX12310): Bshell core dumps when doing BDB_DEBUG=100 
tracing

Date: Fri, 27 Jun 2003 16:32:56 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
For a normal user database roles are defined.
When logging into Baan having the following command set bshell core dumps:
-- -set BDB_DEBUG=00100 -logfile /tmp/bdb_debug -keeplog

Problem Description (Technical terms)
Because of the tracing a fprintf is done containing a null pointer.
This is because there doesn't need to be always a condition.

Workaround
Do not use BDB_DEBUG=0100 tracing

Test Procedure
Pre-requisite: Have a Baan user (normal user), and have roles to it assigned.
In that case directory $BSE/lib/roles/db will have role files.
Do not assign conditions to the roles.
Now log in Baan, but have the following filled in the Command field of BW:
-- -set BDB_DEBUG=00100 -logfile /tmp/bdb_debug -keeplog
The bshell should not core dump now when logging in Baan.


Motive source
DF:137511




MaintCorelli: # 19110 (BDUX12299): Restore imake rule NormalSharedLibraryTarget 
to original except for AS400

Date: Fri, 27 Jun 2003 14:46:05 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Tru64 and Linix did not like the new NormalSharedLibraryTarget rule so the original was restored exceot for AS400

Workaround
#ifdef BUILD_AS400

Test Procedure
Nightly build




DevCorelli: # 19104 (BDUX12297): AS400: Update /vobs/logic/bic/bic_yy_aix.c and 
/vobs/logic/bic/bic_tokens_aix.h

Date: Fri, 27 Jun 2003 11:35:42 +0200

Created on:  DevCorelli
Type:        porting specific change

Problem Description (Technical terms)
Please, update /view/DevCorelli.night/vobs/tt/bic/bic_yy_aix.c
Please, update /view/DevCorelli.night/vobs/tt/bic/bic_tokens_aix.h

Test Procedure
Nightly build




MaintCorelli: # 19073 (BDUX12248): AS400: Another fix due to 5799PTL

Date: Thu, 26 Jun 2003 09:01:57 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Another round of fixes due to the new 5799PTL packages and a stricter compiler.

Test Procedure
Nightly build




MaintCorelli: # 19044 (BDNT11826): Bwprint: Make Show Barcode Text optional

Date: Wed, 25 Jun 2003 08:30:56 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
The barcode image, created by Bwprint, contains by default the barcode text (caption).
We would like to have a barcode without the barcode text.

Problem Description (Technical terms)
Retrieve value ShowBarcodeText from registry. This value has to be entered manually into the registry.
key:	HKEY_LOCAL_MACHINE\SOFTWARE\Baan\BwPrint\BarcodeText
value:	ShowBarcodeText (DWORD value): 
		0:	Barcode Text is not displayed
		1:	Barcode Text is displayed (this is the default)

Future enhancement: make this option controlable using bshell function bc$.
	

Workaround
No workaround

Test Procedure
1. Create the registry key: HKEY_LOCAL_MACHINE\SOFTWARE\Baan\BwPrint\BarcodeText
2. For this key create the value ShowBarcodeText (DWORD value), and assign the value "0" to it.
3. Start Bwprint.exe 
4. Open the barcode test
Now the barcode images should not contain barcode text.





MaintCorelli: # 19040 (BDNT12220): BW freezes when selecting a menu from menubar

Date: Wed, 25 Jun 2003 08:27:59 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Baan system is locking after opening "specific menu" and clicking "Window" menu

1. Go to the session "Document revision"
2. Mark a revision (in this case GHS-0006)
3. Open the specific menu very shortly after marking the revision. To reproduce easily use "alt+S"
4. Open with the mouse the menu "Window" in the menu bar.
Now BW freezes and consumes a lot of cpu time.
 

Problem Description (Technical terms)
When a menu is updated while the user selects another menu this problem appears.
Updates should be delayed till windows exit the menuloop.

This problem was already solved by the following projects:
7.1c.05: BDNT803
7.2b: BDNT6352
but this fix does not solve the problem in case you use the keyboard (alt-s) to select the menu.

To solve this the windows messages WM_ENTERMENULOOP and WM_EXITMENULOOP should be handled seperately.

Workaround
Do not use the keyboard to select the menu, but the mouse.

Test Procedure
1. Go to the session "Document revision"
2. Mark a revision
3. Open the specific menu very shortly after marking the revision. To reproduce easily use "alt+S"
4. Secondly open with the mouse the menu "Window" in the menu bar.
Now BW should not freeze and continue to work correctly.


Motive source
DF:136628




MaintCorelli: # 19027 (BDNT11967): EVTLISTBOXSELECT not allways raised by bw.exe

Date: Tue, 24 Jun 2003 12:48:15 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
When using multiple pull-down menus in 5.0b or 5.0c the value selected is
reverted back to the original value when <TAB> is pressed quickly after the 
selection is made.

Problem Description (Technical terms)
Windows will not create a CBN_SELCHANGE message in the described situation.
Because of this BW will not end a EVTSELCHANGE to the bshell. 

==> bshell is not aware about the content of the field. This caused
    the strange behaviour.


Workaround
n/a

Motive source
DF:214450




DevCorelli: # 19021 (BDUX12221): MSQL: Error 1169: ORDER BY cannot contain 
duplicates

Date: Tue, 24 Jun 2003 09:46:42 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
"Database error 169 happens deleting one or more records in session 'Signals by
Item/Planner' (cprao1125m000)".

Problem Description (Technical terms)
If a query's ORDER BY clause contains duplicate references to the same column 
of the SELECT clause, the MSQL level-2 driver gives error 1169. In the event 
viewer the following message can be found:

	dbs_errno = 1169 Error 1169 occurred:
	Error sqlcode 169 (sqlstate: 42000, Return code: -1, diagRegNum: 1) during SQLExecute.
	[Microsoft][ODBC SQL Server Driver][SQLServer] A column has been specified more than 
	once in the order by list.  Columns in the order by list must be unique.

The same problem occurs if a db.command (e.g. db.first()) is done on a non-
unique index (that is an index that allows duplicate values). In that case a
query is generated that contains duplicate references to the same column.

Workaround
The second duplicate reference in the ORDER BY clause, has no function.  If 
possible, remove the second duplicate reference from the ORDER BY clause.

Test Procedure
Execute the following command:

	qptool6.2 -q "select * from dbtst120 order by empno, firstnme, empno"

Observe that the result is correct. Re-run the test with the environment
variable DBSLOG set to 400. Search the dbs.log file for the corresponding
query and observe that the ORDER BY clause contains just 1 reference to
the column empno.


Motive source
DF:137191




MaintCorelli: # 19020 (BDUX12221): MSQL: Error 1169: ORDER BY cannot contain 
duplicates

Date: Tue, 24 Jun 2003 09:46:04 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
"Database error 169 happens deleting one or more records in session 'Signals by
Item/Planner' (cprao1125m000)".

Problem Description (Technical terms)
If a query's ORDER BY clause contains duplicate references to the same column 
of the SELECT clause, the MSQL level-2 driver gives error 1169. In the event 
viewer the following message can be found:

	dbs_errno = 1169 Error 1169 occurred:
	Error sqlcode 169 (sqlstate: 42000, Return code: -1, diagRegNum: 1) during SQLExecute.
	[Microsoft][ODBC SQL Server Driver][SQLServer] A column has been specified more than 
	once in the order by list.  Columns in the order by list must be unique.

The same problem occurs if a db.command (e.g. db.first()) is done on a non-
unique index (that is an index that allows duplicate values). In that case a
query is generated that contains duplicate references to the same column.

Workaround
The second duplicate reference in the ORDER BY clause, has no function.  If 
possible, remove the second duplicate reference from the ORDER BY clause.

Test Procedure
Execute the following command:

	qptool6.2 -q "select * from dbtst120 order by empno, firstnme, empno"

Observe that the result is correct. Re-run the test with the environment
variable DBSLOG set to 400. Search the dbs.log file for the corresponding
query and observe that the ORDER BY clause contains just 1 reference to
the column empno.


Motive source
DF:137191




MaintCorelli: # 19004 (BDUX12153): AS400: Repair name mangling when using icc 
(QAPTL/QCCMAPOUT.SRVPGM)

Date: Mon, 23 Jun 2003 12:18:08 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
icc -o hello6.2 hello.c produced the following files:

	/qsys.lib/port.lib/hello6.2.module
	/qsys.lib/port.lib/hello6.pgm


The last one was /qsys.lib/port.lib/hello6.2.pgm when using our old version of the 5799PTL package. This name mangling was done by GNU/QCCMAPOUT.SRVPGM which is now moved to QAPTL/QCCMAPOUT.SRVPGM. Also the name mangling program is now a C++ object instead of a C object. We need our own name mangling because we want to preserve program names like bshell6.2.pgm (RSTFRMSTMF commands when installing portingsets).

Workaround
Manually move /qsys.lib/QAPTL.LIB/QCCMAPOUT.SRVPGM to /qsys.lib/QAPTL.LIB/QCCMAPOU2.SRVPGM and put our own QCCMAPOUT.SRVPGM in its place. This solves the name mangling problem.

Test Procedure
Nightly build




MaintCorelli: # 18997 (BDUX12207): bugfix #endif __WIN32__ in al_env.c

Date: Mon, 23 Jun 2003 09:30:24 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
cpp: "/vobs/tt/lib/al_1/al_env.c", line 263: warning 2010: Extra characters on #endif.




MaintCorelli: # 18996 (HYUX11500): AS400 : Handling of Table without indexes 
(Level1)

Date: Mon, 23 Jun 2003 08:42:41 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
BDBPOST with "-f" option doesn't work on Level1 AS400 driver. 

Problem Description (Technical terms)
In Level1 AS400 driver, tables are accessed using the logical file (index).
While using "-f" option, index won't be created and hence while accessing the
table, it results with E_BDB_NO_TABLE error.

Workaround
Without "-f" option it works fine.

Test Procedure
Take a sequential dump of a table using bdbpre and then import the dump
using bdbpost along with "-f" option.

Motive source
DF:133932




DevCorelli: # 18987 (BDNT12199): TSS: merge from projects BDNT12185 and 
BDNT12190

Date: Fri, 20 Jun 2003 15:52:05 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
merge needed




MaintCorelli: # 18986 (BDNT12199): TSS: merge from projects BDNT12185 and 
BDNT12190

Date: Fri, 20 Jun 2003 15:51:39 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
merge needed




DevCorelli: # 18948 (BDUX12165): Bison on DevCorelli

Date: Wed, 18 Jun 2003 16:23:11 +0200

Created on:  DevCorelli
Type:        bugfix

Problem Description (Technical terms)
Bison on DevCorelli

Test Procedure
Build a DevCorelli porting set and ensure that the manual copying of yacc output is not needed on NT.




MaintCorelli: # 18936 (BDUX12163): Debugable Cpp failed to run on AIX (red) 
without any real reason

Date: Wed, 18 Jun 2003 13:59:07 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
Debugable cpp was not able to load libc.a. Very strange.

Workaround
Export LIBPATH=/usr/lib.

Test Procedure
run Util/Boot in debugable clearcase object tree





MaintCorelli: # 18934 (BDNT12009): Making Bison available on Corelli

Date: Wed, 18 Jun 2003 13:56:33 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Making Bison available for corelli on NT

Test Procedure
Build the corelli porting set and verify that the manual process of copying the yacc generated files from unix to NT should not be needed anymore.




MaintCorelli: # 18929 (BDNT12161): Fixed compilation error

Date: Wed, 18 Jun 2003 12:43:21 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Fixed compilation error in ogclient

Test Procedure
Compile




MaintCorelli: # 18924 (BDUX12159): Fixed compilation errors

Date: Wed, 18 Jun 2003 11:57:30 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Fixed compilation errors, introduced by BDNT10939




MaintCorelli: # 18920 (BDUX12142): Rename back to 7.1d.03

Date: Tue, 17 Jun 2003 16:31:42 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Updated BW (7.1d.03.155) number and changed portingset number back to 7.1d.03

Workaround
NA

Test Procedure
should already be known




MaintCorelli: # 18913 (BDNT12138): Manual merge of BDNT12131 to Corelli

Date: Tue, 17 Jun 2003 16:21:17 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Customer terms)
Installer problems when installing an update CD on Windows NT without tools update. See also BDNT12131.

Workaround
NA




MaintCorelli: # 18911 (BDUX12137): Temp rename of PS back to 7.1d.02, due to 
required rebuild for NT and AS/400

Date: Tue, 17 Jun 2003 15:21:15 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
BW version rename, due to rebuild of 7.1d.02 for NT (and AS/400)




MaintCorelli: # 18910 (BDNT10939): Misc. changes. Use library function instead 
om similar local one.

Date: Tue, 17 Jun 2003 15:14:26 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Some miscelaneous changes. Initially the project was used for making an overleaped bshell/MSQL driver.

Workaround
NA




MaintCorelli: # 18894 (BDUX12134): Manual merge of project BDUX12120. Test on 
empty variables failed for some platforms

Date: Tue, 17 Jun 2003 14:32:18 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Code gave errors on some platforms, e.g. Linux. Empty variable do not allow [ $(SharedLibrary) -ge 1 ]. Added -z test

Workaround
NA




MaintCorelli: # 18890 (BDUX12119): Fix al_sli null pointer on Solaris

Date: Tue, 17 Jun 2003 14:29:54 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
Trying to load an non-existent shared library on Solaris could result in a core dump due to a dlerror call that can return a NULL pointer when the dynamic load error is already cleared.

Workaround
Check the return value of dlerror before printing it.

Test Procedure
Load an nonexistent shared library (for example BCLM 2.0 libBclmEx.so).




MaintCorelli: # 18861 (BDUX12046): AS400: Do not mix options and arguments when 
using icc front-end

Date: Mon, 16 Jun 2003 11:32:20 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Compiling on AS400 did not work after installing an official version of the 5799PTL iSeries Tools for Developers package. This was caused by the fact that compilations like "icc -c hello.c -I." did no longer work but have to be rewritten to "icc -c -I. hello.c". This is now handled by our front-end to icc, the acc program.

Workaround
Use the acc front-end to sort argument and options of icc call, change some Imakefiles so that the $(LD) command is also called withthe correct sequence of options and arguments.

Test Procedure
Nightly build




DevCorelli: # 18847 (BDNT12094): Do not delete destination path of COMPRESS is 
true

Date: Fri, 13 Jun 2003 16:41:24 +0200

Created on:  MaintBaanIVc
Type:        porting specific change

Problem Description (Technical terms)
Do not delete complete directory tree if COMPRESS is on.

Workaround
NA

Test Procedure
Run deliver with COMPRESS on and chech that underlying files/directories under the DEST path are not removed




MaintCorelli: # 18846 (BDNT12094): Do not delete destination path of COMPRESS is 
true

Date: Fri, 13 Jun 2003 16:41:04 +0200

Created on:  MaintBaanIVc
Type:        porting specific change

Problem Description (Technical terms)
Do not delete complete directory tree if COMPRESS is on.

Workaround
NA

Test Procedure
Run deliver with COMPRESS on and chech that underlying files/directories under the DEST path are not removed




MaintCorelli: # 18828 (BDUX12080): Printerchange on Eddie

Date: Thu, 12 Jun 2003 16:23:57 +0200

Created on:  MaintBaanIVc
Type:        porting specific change

Problem Description (Technical terms)
PR2015 does not exist anymore. So new printer is chosen as default on Eddie

Workaround
Change every makefile by hand




DevCorelli: # 18823 (BDNT12072): Deliver should not log in destination dir, 
before it is created

Date: Thu, 12 Jun 2003 16:21:32 +0200

Created on:  MaintBaanIVc
Type:        porting specific change

Problem Description (Technical terms)
Delivery failed if destination directory was not existent and COMPRESS was true and BUILDCAMEOXML was true.
Also some other minor issues.

Workaround
Manual create destination dir before running deliver.

Test Procedure
Run deliver for a new release with specified variables set to TRUE.




MaintCorelli: # 18822 (BDNT12072): Deliver should not log in destination dir, 
before it is created

Date: Thu, 12 Jun 2003 16:19:32 +0200

Created on:  MaintBaanIVc
Type:        porting specific change

Problem Description (Technical terms)
Delivery failed if destination directory was not existent and COMPRESS was true and BUILDCAMEOXML was true.
Also some other minor issues.

Workaround
Manual create destination dir before running deliver.

Test Procedure
Run deliver for a new release with specified variables set to TRUE.




DevCorelli: # 18811 (BDNT12062): Deliver did not cleanup after error

Date: Thu, 12 Jun 2003 10:56:12 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
Deliver did not cleanup directories if an error occurred.

Workaround
NA

Test Procedure
Run deliver and check if an error occurs that c:\temp\iswi and c:\temp\port.xxx have been removed




MaintCorelli: # 18810 (BDNT12062): Deliver did not cleanup after error

Date: Thu, 12 Jun 2003 10:55:48 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
Deliver did not cleanup directories if an error occurred.

Workaround
NA

Test Procedure
Run deliver and check if an error occurs that c:\temp\iswi and c:\temp\port.xxx have been removed




MaintCorelli: # 18799 (BDUX12037): Fix multiple-macro-and-string bug

Date: Thu, 12 Jun 2003 08:10:47 +0200

Created on:  DevNewton
Type:        bugfix

Problem Description (Customer terms)
After two #define-s for the same macro name (e.g., somename(...) and somename()), and a string literal (CONSTANT)
with the same contents (e.g., "somename") the second macro definition cannot be found anymore, and when
the second macro is called, the content of the first definition is substituted.

Example:

    #define somename(...) "unreachable"
    #define somename() | this definition doesn't matter, as long as it compiles
    
    function extern void f(const string str)
    {
        somename()
        f("somename")
        | From this point on the second definition of somename is not seen
        somename()      |<----- this uses the first definition
                        |       (but it should use the second one)
    }


Problem Description (Technical terms)
After defining both macros and using the string, there are three symbols named "somename", two
of type MACRO and one of type CONSTANT.  All symbols after the CONSTANT are ignored when
expanding a macro call to 'somename'.

Probably the problem can also occur in other combinations of defines and constants with the same
name/content.

Workaround
Don't use constant strings that have the same content as macro names.

Test Procedure
Compile the above script.




MaintCorelli: # 18788 (BDUX12057): Oracle driver on AIX now shared library based

Date: Wed, 11 Jun 2003 12:55:12 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
The Oracle executables are now using the shared Oracle libraries to solve
sone AIX specific technical reasons (32-bit pw-syscall dependency in client).

Test Procedure
nightly build and regression test




DevCorelli: # 18769 (BDNT12034): Fix in deliver. Also cleanup iswi stuff

Date: Wed, 11 Jun 2003 08:42:46 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Minor bug in deliver. Also cleanup ISWI object tree, otherwise MSI based BW may reuse/misuse other sources

Workaround
NA

Test Procedure
Run a deliver and check that c:\temp\iswi is cleaned up (this directory is configured in deliver.cfg, you may use another one, see OBJBWMSI)




MaintCorelli: # 18767 (BDNT12034): Fix in deliver. Also cleanup iswi stuff

Date: Wed, 11 Jun 2003 08:41:52 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Minor bug in deliver. Also cleanup ISWI object tree, otherwise MSI based BW may reuse/misuse other sources

Workaround
NA

Test Procedure
Run a deliver and check that c:\temp\iswi is cleaned up (this directory is configured in deliver.cfg, you may use another one, see OBJBWMSI)




MaintCorelli: # 18763 (BDUX11918): AS400: New version of 5799PTL package 
installed

Date: Wed, 11 Jun 2003 08:39:04 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
The new 5799PTL package includes the icc front-end we use for comiling on AS/400. This version is more critical about placing the options and arguments in the icc command:

	icc -I. -c hello.c -o hello.o

Should now be:

	icc -I. -c -o hello.o hello.c

Workaround
Reshuffle Makefile

Test Procedure
Nightly build




DevCorelli: # 18752 (BDNT11988): Migrated BW to IS7

Date: Tue, 10 Jun 2003 13:19:28 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
InstallShield developer 7 cannot be used together with InstallShield Professional for Windows Installer edition.
Hence migrated BW MSI project to this newer IS (Developer 7) version. Installshield renamed the product to 'Developer', but is 
a sucessor of IS Prof for Windows Installer Edition.

Workaround
NA

Test Procedure
Make installer image and check whether you can install bw with the provided image (setup.exe)




MaintCorelli: # 18751 (BDNT11988): Migrated BW to IS7

Date: Tue, 10 Jun 2003 13:19:08 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
InstallShield developer 7 cannot be used together with InstallShield Professional for Windows Installer edition.
Hence migrated BW MSI project to this newer IS (Developer 7) version. Installshield renamed the product to 'Developer', but is 
a sucessor of IS Prof for Windows Installer Edition.

Workaround
NA

Test Procedure
Make installer image and check whether you can install bw with the provided image (setup.exe)




DevCorelli: # 18746 (BDNT12025): Manual merge of project BDNT11686 to Corelli

Date: Tue, 10 Jun 2003 12:16:53 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
default.bwc is overwritten after portingset update of BSE environment on Windows

Problem Description (Technical terms)
see above

Workaround
Manual edit default.bwc afterwards

Test Procedure
Do a portingset update and check that default.bwc is not changed anymore




MaintCorelli: # 18745 (BDNT12025): Manual merge of project BDNT11686 to Corelli

Date: Tue, 10 Jun 2003 12:16:09 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
default.bwc is overwritten after portingset update of BSE environment on Windows

Problem Description (Technical terms)
see above

Workaround
Manual edit default.bwc afterwards

Test Procedure
Do a portingset update and check that default.bwc is not changed anymore




MaintCorelli: # 18725 (BDNT12004): When marking records, Records doesnt remained 
marked using BI

Date: Thu, 5 Jun 2003 16:45:51 +0200

Created on:  DevCorelli
Type:        bugfix

Problem Description (Customer terms)
If you make a multi select in a multiocc session then the rows will not keep marked when using BI.
Start for example session ttaad2500m000 (maintain user) and select one row with the mouse. 
Then press shift and mark an additional row so that a multi select will be done. 
The rows are marked only for some seconds.


Problem Description (Technical terms)
Unmark was not implemented correctly in BI.

Workaround
Use single selection.

Test Procedure
See problem description.

Motive source
DF:135490




MaintCorelli: # 18701 (BDNT11981): TSS: bug in TSSutf8_n2tssmbs_n

Date: Thu, 5 Jun 2003 10:27:00 +0200

Created on:  MaintReger
Type:        bugfix

Problem Description (Technical terms)
UTF8 in the range U+0080-U+00FF was not correctly converted to TSS.

Test Procedure
See scopus defect

Motive source
DF:137389




MaintCorelli: # 18640 (BDUX11921): unresolved symbol htons

Date: Mon, 2 Jun 2003 16:54:51 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Load error on OS390:
IEW2456E 9207 SYMBOL HTONS UNRESOLVED.  MEMBER COULD NOT BE INCLUDED FROM THE DESIGNATED CALL LIBRARY.

Test Procedure
build bshell on OS390




MaintCorelli: # 18611 (BDNT11908): Added new build system to scrypt/patch list

Date: Mon, 2 Jun 2003 10:41:58 +0200

Created on:  MaintGemini
Type:        porting specific change

Problem Description (Technical terms)
Added new build systems to scrypt and patch

Workaround
NA

Test Procedure
NA (only for technicans)




MaintCorelli: # 18603 (BDNT11917): BwPrint: hha.dll for HtmlHelp

Date: Mon, 2 Jun 2003 09:59:46 +0200

Created on:  DevNewton
Type:        bugfix

Problem Description (Technical terms)
hhc.exe needs hha.dll

Test Procedure
build bwprint




DevCorelli: # 18599 (BDNT11914): BwPrint: HtmlHelp (11882 to Corelli and Reger)

Date: Fri, 30 May 2003 15:50:35 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
deliver.lst and bwprint.dsp must be merged separately to some branches




MaintCorelli: # 18598 (BDNT11914): BwPrint: HtmlHelp (11882 to Corelli and 
Reger)

Date: Fri, 30 May 2003 15:49:31 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
deliver.lst and bwprint.dsp must be merged separately to some branches




MaintCorelli: # 18591 (BDNT11910): BwPrint: HtmlHelp (merge 11882)

Date: Fri, 30 May 2003 14:41:24 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
Bwprint should use Html Help instead of WinHelp

Test Procedure
Menu item Help->Help Topics




MaintCorelli: # 18570 (BDUX11900): Wrong file permissions after installation on 
Oracle

Date: Wed, 28 May 2003 11:13:57 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
After a new BaanIVc install, some oracle driver related files have wrong
ownership.

Problem Description (Technical terms)
In the past these files were part of the distribution, so got the right
ownership.
Now they are generated when needed AFTER the setting of the uid/gid, so
they keep the ownership of the install user (root).

Workaround
cd $BSE/lib/ora
chown bsp:bsp ora_users ora_groups ora_storage
chown -R bsp:bsp oracle_home
cd $BSE/lib/defaults
chown bsp:bsp db_resource

Test Procedure
Full unstall

Motive source
TCS:100-131710




MaintCorelli: # 18569 (BDUX11903): Fixed problem in install6.2

Date: Wed, 28 May 2003 11:12:43 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Fixed problem with installing Unix CD media for Baan 5 (Corelli)
Missing components on CD, take them from BSE

Workaround
NA

Test Procedure
run install6.2




MaintCorelli: # 18566 (BDUX11796): OS390 compilation error

Date: Wed, 28 May 2003 10:18:23 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Compilation error java_integr.c due to missing JDK1.3 include files in thirdparty vob.

Test Procedure
compile




MaintCorelli: # 18530 (BDUX11861): Crystal integration causes full shared memory

Date: Tue, 27 May 2003 09:59:04 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Shared memory keeps on growing when using the Crystal Reports integration

Problem Description (Technical terms)
Crystal is generating tempory BSE environments, each time with a random name.
This causes the DB driver cache in shared memory to be filled.

Test Procedure
Set DBSLOG=1000 and verify that DD's are not retrieved from the shared memory
for the Crystal integration.

Motive source
DF:133783




MaintCorelli: # 18484 (BDUX11618): Buffersize to small during reading domain 
info

Date: Thu, 22 May 2003 14:53:28 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Cannot log in Baan.
When logging into Baan, or when running programs like qptool6.2, bdbpre6.2,
etc, the following errors are logged, e.g.:
Log_mesg: search domain 'ttadv.keyr' failed
No definition in definition file for sopen(F_BRDD:dttaad999, dttaad999)

Problem Description (Technical terms)
During reading the contents of the domain file, dtt.pd, function count_items()
is executed. This function does a sgets to read the domain descriptions from
the file.
This is done using a buffer size of 511.
But, probably because languages are added, there are domain descriptions which
are bigger then 511 bytes.
For example domain ttadv.eshc. When such a domain description is read,
the buffer is not large enough, and domain information gets corrupted.

The problem does not occur in the Gemini environment because this environment
is using a bshell containing the modifications of project BDUX9125.
(Because this was an enhancement project, this project was not merged to Corelli). Beside other modifications, the buffersize is also increased to 8Kb.


Workaround
Remove languages.

Test Procedure
Now there are no problems / errors anymore when logging into Baan or executing programs like qptool6.2, bdbpre6.2, etc.

Motive source
DF:123123




MaintCorelli: # 18478 (BDNT11728): Bwprint: some esthetical modifications

Date: Thu, 22 May 2003 13:53:48 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Barcode Name is not displayed in the Barcode Test
Barcode Name is not in error message

Problem Description (Technical terms)
Retreive Barcode Name using barcode functions.

Workaround
Not applicable.

Test Procedure
Bwprint: Run barcode test and check if Barcode names are also there.

Motive source
DF:136366




MaintCorelli: # 18429 (BDUX11790): Startup Sessions does not start when logging 
in via Worktop or Webtop

Date: Tue, 20 May 2003 15:18:36 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Within the Baan Tools you can define Startup Sessions.
These sessions are automatically started when logging into Baan.
But when logging in Baan using Worktop (or Webtop) user interface,
the Startup Sessions are not started.

Problem Description (Technical terms)
Function ds_execute_session_files() is not called in case arg_count > 0.
When logging in via Worktop, the argument is ostpoledaemon, because Worktop
communicates via ostpoledaemon with the backend.

Workaround
No workaround. (Startup manually)

Test Procedure
1. Define a Session Group (ttaad2107m000)
2. Assign sessions to this Session Group (ttaad2106m000)
3. Assign this Session Group to your user (ttaad2105m000), and Convert to Runtime

Now the file $BSE/lib/user/u<Baan Username> should contain the following entry:
bshell.sessions:<name of Session Group>.sessions

and there should be a file <Session Group> in the same directory. This file contains a list of
the Startup Sessions.

Note: When having another Bshell name (then default "bshell"), the entry in the u<Baan Username> file
should not be bshell.sessions:..., but <alternative bshell name>.sessions:...

4. Log in using BW, and verify that the Startup Sessions are really started *)
5. Log in using Worktop, and verify again that the Startup Sessions are really started.

*) This step is only possible in case of BaanERP5.0c or lower.

Motive source
DF:214069




MaintCorelli: # 18403 (BDUX11781): shdbapi build failed

Date: Mon, 19 May 2003 13:11:51 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Build of shared library failed after merge of BDUX11729.
The shared lib build on Corelli differs from Reger+.
On Corelli each single obj file is specified. This list contained a removed
object file.

Test Procedure
nightly build




MaintCorelli: # 18379 (BDUX11729): blat_templ.l converted to blat_templ.c

Date: Fri, 16 May 2003 08:56:59 +0200

Created on:  DevNewton
Type:        bugfix

Problem Description (Customer terms)
1) BLATSEP=_ is not picked up properly
2) On AS400 there are ASCII/EBCDIC related problems since the introduction of
	the _aix_ derived files

Problem Description (Technical terms)
3) Lex files are a build-nightmare because of the _aix_ derived files.
4) On OS390 lex needs a %p setting of 10 times the default (why ?????)

Workaround
ad 1) specify another separator like '='

Test Procedure
test 1) .../blat/Test/*

test 2) manual
% cd /vobs/obj.*/lib/blat/Tools
% clearmake test
% test_templ6.2
<fill the commands on the command line and look at the results>

Motive source
DF-128529
Internal (build) remarks




MaintCorelli: # 18351 (BDUX11736): do not run timetest

Date: Thu, 15 May 2003 11:18:29 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
During build a testprogram timetest is runned. This causes often errors with unknown cause. We better skip this test.

Workaround
Run timetest from commandline and check the results.




MaintCorelli: # 18347 (BDUX11734): Fixed problems for OS/390 build

Date: Thu, 15 May 2003 10:05:54 +0200

Created on:  MaintGemini
Type:        porting specific change

Problem Description (Technical terms)
OS/390 does not like /**/ construction in _Dep() rule (but all other Unix systems accept)

Workaround
NA

Test Procedure
Generate makefile on OS/390




MaintCorelli: # 18338 (BDUX11702): Incorrect named logfiles created in log 
directory

Date: Wed, 14 May 2003 13:47:32 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
We get a lot strange named log-Files in $BSE/log, example:

-rw-rw-rw-   1 scha       bsp   8858 Apr 15 18:13 log.SOCKET<Chutima
-rw-rw-rw-   1 ckoenig    bsp      0 Apr 11 10:56 log.SOCKET<Claudia
-rw-rw-rw-   1 felder     bsp    667 Apr 15 11:08 log.SOCKET<Felder
-rw-rw-rw-   1 ireinhol   bsp   3430 Apr 11 15:07 log.SOCKET<Ines
-rw-rw-rw-   1 jmellein   bsp  18078 Apr 16 09:11 log.SOCKET<Johannes
-rw-rw-rw-   1 isen       bsp   1384 Apr 14 16:23 log.SOCKET<Siegfried
-rw-rw-rw-   1 isen       bsp      0 Apr 15 08:06 log.SOCKET<Stefan
-rw-rw-rw-   1 cheinze    bsp      0 Apr 14 08:53 log.SOCKET<Thomas
-rw-rw-rw-   1 tkozlows   bsp      0 Apr 15 09:27 log.SOCKET<Tomasz
-rw-rw-rw-   1 scoerper   bsp    679 Apr 15 16:53 log.SOCKET<sven

This happens when connecting to the server with a client user (not Baan user) 
that has a space in the name. For example, in case the windows user is "Piet Paulesma".


Problem Description (Technical terms)
Function log_set_progname is called, eg. by the bshell, as follows:
log_set_progname(bshell_kvd (kvddool@cnl09659:1360/SOCKET<piet paulesma@CNL09659:1360>), TO_LOG)
>From function log_set_progname, function BaseInit (al_init.c) is called:
Within this function the program name is determined, by doing a right-search for a space.
The first space found is the one between "piet" and "paulesma".
So, prog becomes "bshell_kvd (kvddool@cnl09659:1360/SOCKET<piet", in stead of "bshell_kvd".
Later on, function al_basename, converts this in "SOCKET<piet", so the name of the log file will
become "log.SOCKET<piet".


Workaround
Do no use spaces in the name of the windows user.

Test Procedure
1. On your windows workstation, create a windows user having a space in the name. For example: Piet Paulesma
2. Log on your windows workstation using this user.
3. Now start BW and log in Baan (don't matter which Baan user is used, it has to be a Unix server)
4. Check the ${BSE}/log directory: no files like "log.SOCKET<piet" or "log.PIPE<piet" should be created.

Motive source
DF:136271




MaintCorelli: # 18329 (BDUX11716): removed obsolete file in Linux distribution

Date: Wed, 14 May 2003 10:25:46 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Linux distribution contained an obsolete file needed for a statically linked oracle
driver. As Linux now uses a shared driver, this file was not generated anymore.

Test Procedure
Nightly build




MaintCorelli: # 18325 (BDUX10690): coredump in ServiceConfiguratorExImpl

Date: Tue, 13 May 2003 17:16:19 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
bshellxma dumps a core

Problem Description (Technical terms)
attempt to write data on location which was not yet allocated.

Test Procedure
use baanscs on solaris

Motive source
TCS:132229




MaintCorelli: # 18299 (BDUX11666): bshell: 3GL concat function fix

Date: Tue, 13 May 2003 09:00:12 +0200

Created on:  MaintReger
Type:        bugfix

Problem Description (Technical terms)
3GL concat$() function calculated the result size wrong and because of this no more string operations were available.
(It gived one value more then it should be.)

function main()
{
	string r(18)
	r = concat$(".", "ttppp909", "xxxx") & "notattached" & str$(123)
}

The result for the r variable were : ttppp909.xxxx123
str$() function fixed the length, but too late because the "notattached" part lost.

Test Procedure
See problem description.

Motive source
TCS:106-136642




MaintCorelli: # 18292 (HYUX11664): Informix:Enabling the prepare statement cache

Date: Tue, 13 May 2003 08:55:32 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Enabling the Prepare Statement Cache




MaintCorelli: # 18291 (BDUX11674): build issue on non-clearcase builds

Date: Tue, 13 May 2003 08:53:13 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Syntax error for builds outside clearcase.
Make is more strict on spaces/tabs than clearmake.

Workaround
Use clearmake instead of make

Test Procedure
Just build




MaintCorelli: # 18266 (HYUX11643): Informix:BaanIV installation 
failure:TCS-100/130375

Date: Fri, 9 May 2003 16:45:19 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
Core dump in the informix driver with the prepare statment cache
turned on.

Workaround
Turn off the Prepare Statement Cache
(inf_use_preapre_cache:0)

Test Procedure
Run bdbpost with number of tables > the size of the prepare statement cache.

Motive source
TCS:100-130375




MaintCorelli: # 18263 (BDUX11630): Oracle driver: some build and config issues

Date: Fri, 9 May 2003 14:57:45 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
No runtime consequences for the outside world
Several build / code setup issues that were pending a long time

With these changes, only one parameter in the Project.all file
controls if a platform uses shared (NT/Linux) or static (other Unixes) Oracle libraries.
No additional setups are needed anymore, and one explicit #define
The shipment of client files is not needed anymore when a platform uses shared libs.

Test Procedure
Nightly build and regression test




MaintCorelli: # 18240 (BDUX11633): Make object vob ready for linux

Date: Thu, 8 May 2003 12:54:00 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Added support for Linux builds on ClearCase platform.
Renamed SOL9 vob for Linux for this purpose.




MaintCorelli: # 18234 (BDUX11610): AS400: Fix FUN_ARG in parse_arguments

Date: Thu, 8 May 2003 08:46:10 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Customer terms)
"bshell6.2 -v" did no longer work on AS400.

Problem Description (Technical terms)
The function parse_arguments from al_arg.c did no longer handle the type FUN_ARG well on AS400. Did was due to the fact that the typedef ARG_LIST used a (char *) type to store a (void *) casted function pointer which was again casted to *(int (*)(). This resulted in the error:

	qsh: 001-0079 Process ended by exception 9901.


which creates the following spooled file (WRKSPLF):

                                      From module . . . . . . . . :   AL_ARGDC
                                      From procedure  . . . . . . :   process_arg
                                      Statement . . . . . . . . . :   48
                                      To module . . . . . . . . . :   AL_ARGDC
                                      To procedure  . . . . . . . :   process_arg
                                      Statement . . . . . . . . . :   48
                                      Message . . . . :   Pointer not set for location referenced.
                                      Cause . . . . . :   A pointer was used, either directly or as a basing
                                        pointer, that has not been set to an address.


Workaround
Instead of (char *) use (void *) in typedef ARG_LIST

Test Procedure
Build bshell6.2 and execute "bshell6.2 -v".

Motive source
TCS:100-132030




MaintCorelli: # 18202 (BDUX11605): Putenv on Unix should not use automatic 
storage

Date: Tue, 6 May 2003 11:49:16 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
In Unix a putenv call should not use automatic storage ie buf[4096] as buffer.

Workaround
When not on __WIN32__ or IBM_AS400, malloc the 4096 bytes buffer and do not free it afterwards.

Test Procedure
bshell6.x -dbgjvmi -set BAANHOME="/usr/baan/shared




MaintCorelli: # 18194 (BDUX11564): AS400:Enabled DB_MANUAL Option

Date: Mon, 5 May 2003 14:43:07 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
To make debugging process more easier, DB_MANUAL option is enabled in AS400
like any OS.

Test Procedure
Use DB_MANUAL option and try debugging.




MaintCorelli: # 18190 (BDUX11571): AS400:Included DB4CONV command file in 
Porting set

Date: Mon, 5 May 2003 14:40:51 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Customer terms)
DB4CONV command file not found in porting set

Problem Description (Technical terms)
DB4CONV command file and its corresponding panelgroup files are not included
in porting set.

Workaround
Download the solution from support web.

Test Procedure
Run make deliver and check for these two files.




MaintCorelli: # 18178 (BDUX11563): AS400: Move ADDEXITPGM QUSRSYS/BAANRXCEXT to 
later stage in install process

Date: Fri, 2 May 2003 09:53:40 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Customer terms)
Corelli portingset does not install on OS400 version V5R2 due to IBM security patch in ADDEXITPGM command

Problem Description (Technical terms)
While installing the portingset, the CL command ADDEXITPGM PGM(QUSRSYS/BAANRXCEXT) was called before the pgm/baanrxcext stream file was unpacked. This was not a problem with OS400 versions V5R1 and lower but was identified by IBM as a security risk later and so was fixed which broke our install.

Workaround
Do the ADDEXITPGM later in the install process. Old portingsets can be installed by restoring BAANRXCEXT prior to installing.

Test Procedure
Install portingset




MaintCorelli: # 18169 (BDNT11491): bshell waits for DS input

Date: Fri, 2 May 2003 09:43:46 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
The bshell does not respond anymore after a large number of OpenWorld calls.

Problem Description (Technical terms)
Bshell is waiting for DS input, while UI does not intend to send any data.

Workaround
Run UI-less bshell or generate an event with UI, for example click on a button.

Test Procedure
OW 2.4 performance test.

Motive source
DF:1031692




MaintCorelli: # 18141 (BDUX11537): Do not start tgbrg4260m000 in case of Webtop 
login using DS

Date: Fri, 25 Apr 2003 16:24:56 +0200

Created on:  MaintGemini
Type:        bugfix

Problem Description (Customer terms)
When the DEM Browser is set as Startup program in the User Data, it is not possible to
login using Webtop.

Problem Description (Technical terms)
The bshell detects whether the startup program equals tgbrg4260m000.
If this is the case, it directly starts tgbrg4260m000. 
This way of working is introduced in the past with project VHUX2585, to solve a security leak.

Later on, a extra check is built in to check if the user logs on using Worktop. Refer to project BDNT8097.
Because Worktop has its own integrated DEM Browser, tgbrg4260m000 should not be started when logging in
via Worktop.

Now, it shows out also an exception has to be made when logging in via Webtop.
Webtop does not support tgbrg4260m000, because it has also its own integrated DEM Browser.
So the check is extended now specially for Webtop. (By checking for presence of (o)ttstpcondaemon.

Workaround
Do not choose DEM Browser as startup program in your User Data.
Webtop does "ignore" the setting of the startup program.
Set "Worktop Tree" and "Baan Process Browser" to get the DEM process in Webtop.

Test Procedure
Set the startup program in your User Data to "DEM Browser", and convert to Runtime.
Now try to log in using Webtop. 
You should have no problems now. 


Motive source
DF:133693




MaintCorelli: # 18127 (BDUX11496): Errormessages containing multibyte TSS not 
correctly logged

Date: Fri, 25 Apr 2003 13:39:20 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Japanese errormessages are logged as an empty string in the log.bshell.
For example, the following is logged in the log.bshell in case of an overflow error.

English translated errormessage:
Log_mesg: Overflow: format='ZZ9V.9999' value='1111111'

Japanese translated errormessage:
Log_mesg:

Problem Description (Technical terms)
Function vsprintf() is used to format the errormessages.
When the systemlocale is set to Japanese, this function generates error 47.
(Illegal byte sequence). This is because the Japanese errormessage is formatted in TSS.
(And also modified by TSSprep..)

Workaround
No workaround.
(Make a backup of ${BSE}/lib/textinf/Mrtext0j and copy ${BSE}/lib/textinf/Mrtext02 to ${BSE}/lib/textinf/Mrtext0j.) 

Test Procedure
On a japanese system containing Baan having japanese language:
Run does tccom0500m000 (Employees). 
An arbitrary record is selected, and run does tccom0101s000 (Employees).
The Employment tab is clicked. 
111111 is input to the field 'Labor Rate', and data is saved. 
Because it violates the format, log.bshell is made.
(refer to defect itself for more information) 

Motive source
DF:135564




MaintCorelli: # 18106 (BDNT11519): Merged project BDNT11514 to corelli

Date: Thu, 24 Apr 2003 11:59:23 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Allow NT bshell (optimized) to accept the -delay option, to make it possible to delay startup of a bshell (needed for e.g. ASM).

Workaround
NA

Test Procedure
run bshell with -delay 30 (or another value) on Windows NT and check that the bshell starts delayed




MaintCorelli: # 18070 (BDUX11181): Error 880 on DB2 OS390

Date: Fri, 18 Apr 2003 13:53:46 +0200

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Strange error code returned when trying to update a record which is locked by
another session.

Problem Description (Technical terms)
When errors like locktimeout happen the transaction is implicity aborted.
So the savepoint which is put at the start of the transaction is removed.
When an abort comes after a locktimeout error, we try to rollback to the 
savepoint which has been removed and hence the error.

Test Procedure
The problem can be reproduced with the below steps.

1) Lock a record by General Table Maintenance(ttaad4100)
	Example) table tcmcs045
		 Try changing one record in the table so that it is locked.
2) Start another BW and try to change the above record and save it.
	Example) Run tcmcs0145m000 and try changing the record locked 
		 by the GTM session and try saving it.
You see the following error
"Fatal error : tcmcs0145m000: Automatic rollback for retry.point failed (error 1880 database error 880)"
"Fatal error : Can not continue in tcmcs0145m000 in DLL: ottstpstandard ()[retry.point]

Note: Be sure to start a second BW in step 2 because the above problem is not
	reproducible with a single BW.

With the driver with the fix the error you get should be
"Fatal error : Error 850 (E_BDB_FATAL_START) on tcmcs045804 (Max retries (10) exceeded)"
"Fatal error : Can not continue in tcmcs045804 in DLL: ottstpstandard ()[db.update]"

Motive source
Scopus:136013




MaintCorelli: # 18041 (BDNT11312): Check if you have permission to use printer 
and other improvements

Date: Wed, 16 Apr 2003 15:27:29 +0200

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
1)
The -p option in the header of the bpf file specifies a network printer.
In case someone opens the bpf file, and has no permission to use this network printer, 
it takes a long time before bwprint opens the bpf file.

2)
In case the printer specified is not locally installed, but available in the network (and you do have permission):
When printing from Baan having Preview mode turned on, bwprint pops up in the foreground.
Now, when selecting File -> Print, the Print Dialog is started. 
This Dialog does not present you, your default printer, but always the last printer in your list of defined printers. This is not correct, and should be the default printer by default.


Problem Description (Technical terms)
1) At this moment bwprint is doing just an OpenPrinter, no access mask is specified, nor checks are done on permissions.
Need to implement functionality to do this.

2) Function FindPrinterByName: This function calls function FindLocalPrintByHandle.
If function FindLocalPrintByHandle cannot find the local printer it did not return in case ShowError = 1.
But also in case of ShowError = 0, this function should return and not continue.


Workaround
1) Ignore it and just wait. Or do not specify such a printer.
2) Select manually desired printer

Test Procedure
Have a Windows Printer device defined in Baan:
1. The Device Queue specifies a network printer "X": this printer is not locally installed in your Windows Printer setup, and you do have permission for this printer. When printing to this device (without Preview checked), the following message pops up:
"Printer 'X' not locally installed, install this printer locally, and use this printer?"
When answering OK, the printer is added to your local printer setup, and this printer is used.
When answering Cancel, the next message pops up: "Print to default printer?"
When answering OK, the default printer is used, when answering Cancel, nothing is printed.

2. The Device Queue specifies a network printer "Y": this printer is not locally installed in your Windows Printer setup, and you do NOT have permission to use this printer. When printing to this device (without Preview checked), the following message pops up:
"No permission to use printer "Y", print to default printer?
When answering OK, the default printer is used. When answering Cancel, nothing is printed.

3. The Device Queue specifies a network printer "Z": this printer does not exist in the network (and also not locally).
When printing to this device (without Preview checked), the following message pops up:
Printer "Z" not found, print to default printer?
When answering OK, the default printer is used, when answering Cancel, nothing is printed.

4. The Device Queue specifies no printer at all (so, is left empty).
When printing to this device (without Preview checked), no message pops up, and the default printer is automatically used.

5. Repeat step 1, but then select Preview mode.
The following message pops up:
"Printer 'X' not locally installed, install this printer locally, and use this printer?"
When answering OK, the printer is added to your local printer setup, and the Printer dialog presents you printer X.
When answering Cancel, bwprint pops up, and the Printer dialog presents you the Default printer.
When bwprint is there, select File->Print...
Now the print dialog should present you the default printer of your Windows printer setup

6. Repeat step 2, but then select Preview mode.
When bwprint is there, select File->Print...
Now the print dialog should present you the default printer of your Windows printer setup

7. Repeat step 3, but then select Preview mode.
When bwprint is there, select File->Print...
Now the print dialog should present you the default printer of your Windows printer setup

8. Repeat step 4, but then select Preview mode.
When bwprint is there, select File->Print...
Now the print dialog should present you the default printer of your Windows printer setup

9. Have a bpf file, which has -pX, were X is a printer which does exists, and you do have permission to use this printer.
Open this bpf file, by double-clicking on this file.
Then following message pops up:
"Printer 'X' not locally installed, install this printer locally, and use this printer?"
When answering OK, the printer is added to your local printer setup, and the Printer dialog presents you printer X.
When answering Cancel, bwprint pops up, and the Printer dialog presents you the Default printer.


10. Have a bpf file, which has -pX, were X is a printer which does exists, but you do not have permission to use this printer.
Open this bpf file, by double-clicking on this file.
No messages/questions, the print dialog presents you the default printer of your Windows printer setup

Motive source
DF:135715




MaintCorelli: # 18037 (BDNT11444): OpenWorld: text within quotes was converted 
to lowercase

Date: Wed, 16 Apr 2003 15:20:20 +0200

Created on:  MaintReger
Type:        bugfix

Problem Description (Customer terms)
If you perform the following query through OW on all specific aligned fields
(like items) you will get no results back. e.g. (the blanks are 9 blanks)
 
select * from tcibd001 where tcibd001.item = "         TESTAJS"
 
The same applies if I put a filter on an item in the items sessions. 

Problem Description (Technical terms)
The complete query text is converted to lowercase. Text within quotes however
should NOT be altered.

Problem introduced by project BDNT8265

Test Procedure
A small 3GL script to reproduce the problem:

|******************************************************************************
|* tmtstxmlsql  0  VRC 2.3O a  test
|* Test for using xmlExecuteSql
|* baan
|* 03-04-12 [20:51]
|******************************************************************************
|* Script Type: 0
|******************************************************************************
 
 long companyNo,ret,fp,QueryNode,DatasetNode
 string errmsg(200),infile(100),outfile(100)
 
function main()
{
 companyNo = 500
 infile = "d:\b52a\tmp\xmlqueries\query.xml"
 outfile = "d:\b52a\tmp\xmlqueries\query_ret.xml"
 
 fp = seq.open(infile,"r")
 QueryNode = xmlread(fp,errmsg)
 seq.close(fp)
 if QueryNode = 0 then
  message("error reading file: " & errmsg)
 else
  datasetNode = xmlNewNode("dataset")
  ret = xmlInsert(queryNode, datasetNode)
  ret = xmlAppendToChilds(datasetNode, queryNode)
  ret = xmlExecuteSql(QueryNode, companyNo)
  if ret < 0 then
   message ("error executing query: " & str$(ret))
  endif
  fp = seq.open(outfile,"a")
  ret = xmlWritePretty(fp,DatasetNode)
  if ret < 0 then
   message ("error writing to file: " & str$(ret))
  endif
  seq.close(fp)
 endif
}


Where the input document is:
<constructor id="OledbSqlQuery" language="BaanSQL">
    <query>select * from tcibd001 where tcibd001.item = "         BAS"</query>
    <cursor numRows="250"/>
</constructor>
 

and the returned output is:
<?xml version="1.0"?>
<dataset>
 <constructor
  id="OledbSqlQuery"
  language="BaanSQL">
  <query>select * from tcibd001 where tcibd001.item = "         BAS"</query>
  <cursor
   numRows="250"
  />
 </constructor>
</dataset>


Motive source
DF:1031118




MaintCorelli: # 18005 (BDUX11438): New version number for 7.1d.03

Date: Tue, 15 Apr 2003 12:41:18 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Updated BW version number, due to code freeze of 7.1d.02. New version number is 7.1d.03 (BW=7.1d.03.154)




DevCorelli: # 17797 (BDUX11275): Minor bug in cd_read, occurring on Linux

Date: Wed, 2 Apr 2003 15:42:59 +0200

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
cd_read on Linux used less program to show content of files. Less caused confusion, because 'q' need to be pressed, when shown file reached EOF

Workaround
type 'q' when less has shown the installation info (readme.mas)

Test Procedure
Run cd_read6.x and verify that no 'q' key press is needed anymore




DevCorelli: # 17537 (BDNT10963): Be able to handle install.info file in Unix 
style text format and removed full update/new components only dialog

Date: Tue, 11 Mar 2003 12:35:35 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Installer comes with error popup if install.info file is in Unix style text format

Problem Description (Technical terms)
see above

Workaround
- Open install.info with wordpad (write.exe)
- Save file in DOS style text format
This workaround will only work if the medium is writable.

Test Procedure
Run installer, install update cd (be sure that the file update\install.info is in Unix style text format) and check that the installer do
not popup with a message that the locale could not be retrieved from CD.




DevCorelli: # 17534 (BDUX10540): Made sources Bison and Flex compatible

Date: Tue, 11 Mar 2003 12:23:21 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Technical terms)
With this fix the yacc and lex files of BIC and libraries YY_1 and QP are
made compatible with bison (version 1.28) and flex (version 2.5.4)

Test Procedure
Modify the Makefile's of BIC, YY_1 and QP by adding the entry:

YACC = bison -y

Compile BIC, YY_1 and QP.




DevCorelli: # 17381 (BDUX10839): minor install bug

Date: Tue, 25 Feb 2003 12:14:58 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
double check on exitance of bw.z
typo in error message: intentet iso intended




DevCorelli: # 17361 (BDUX10812): Delivery temporary dir fix

Date: Mon, 24 Feb 2003 09:49:41 +0100

Created on:  MaintReger
Type:        bugfix

Problem Description (Technical terms)
NT delivery leaves empty temporary directories.

Workaround
Delete them.

Test Procedure
Do a delivery on NT systems.




DevCorelli: # 17356 (BDUX10811): AS400: build DevCorelli failed

Date: Mon, 24 Feb 2003 09:47:08 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Technical terms)
Build on AS400 failed because in project bugf_BDUX10734 the file qc_qpc.y was updated,
but the derived files qpc_yy_aix.c and qp_token_aix.h were not updated.

Test Procedure
Compile on AS400.




DevCorelli: # 17337 (BDUX10734): QP: comparison with 0.0 inconsistent between 
level 1 and 2

Date: Fri, 21 Feb 2003 09:23:02 +0100

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Customer terms)
In the WHERE clause of queries, the literals "0" and "0.0" behave differently.

Problem Description (Technical terms)
The double literal 0.0 is interpreted as a double with a precision of 1 digit
after the decimal point. As a consequence, the query processor will evaluate
the expression "0.0 = 0.04" to True.

In constrast, the expression "0 = 0.04" yields False.

Also, on level-2 comparing 0.0 to 0.04 yields False.

The behavior is inconsistent between level-1 and level-2. Comparing 0.0 to 0.04 
should yield False.


Workaround
Either change "0.0" to "0" or add more 0's after the decimal point.

Test Procedure
Execute the following command:

	qptool6.1 -q "select * from dbtst120 where 0.0 = 0.04"

The query should return "No rows selected", because 0.0 does not equal 0.04.
It should _not_ return all rows of table dbtst120.


Motive source
DF:131631




DevCorelli: # 17159 (HYUX10636): Informix:No out of range error with date 
-1:TCS-100/130521

Date: Wed, 12 Feb 2003 13:45:55 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
The informix driver inserts the date 2 January 0001 when an attempt to
insert the date -1 is made. All other drivers return error 205.

Motive source
TCS:100-130521




DevCorelli: # 17138 (BDUX10600): Informix level-2: Wrong bind of short values

Date: Wed, 12 Feb 2003 09:34:41 +0100

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
The Informix level-2 driver does not bind correct values to parameters of
type short if the bind values are outside the range of short values.

As a consequence, the Informix level-2 driver may return an incorrect result
for a query, if the query contains an expression in which a column of type
integer is compared to a constant or parameter which value is less than the
minimal integer value (-32768) or greater than the maximum integer value
(32767)

For example:

select empno
from   dbtst120
where  edlevel inrange -100000 and 100000

Column edlevel is of type integer. The Informix level-2 driver returns no
rows for this query. It should return all rows of table dbtst120, because
-100000 is less than the minimal integer value and 100000 is greater than
the maximum integer value.

Workaround
In the query, use the minimum and maximum integer values.

Test Procedure
Create a file 'query' that contains the following query:

select empno
from   dbtst120
where  edlevel inrange -100000 and 100000

Use qptool to execute this query:

qptool6.2 -c 0 -f query

Execute this query on a level-2 Informix driver. Verify that it returns all
rows of table dbtst120.

Motive source
DF:133420




DevCorelli: # 17082 (BDUX10545): Add timing to art trace logging

Date: Thu, 6 Feb 2003 16:09:10 +0100

Created on:  DevCorelli
Type:        enhancement

Problem Description (Customer terms)
No timing information available when testing with
art functions with use of art trace logging. 

Problem Description (Technical terms)
idem

Workaround
Do use tracing and ARM products like HP Openview
and Tivoli

Test Procedure
Enable ART in $BSE/lib/defaults/all with
	art_enable:1
	art_trace:1
	art_user:baan
Start some 4GL session and examine
	$BSE/tmp/bshell.<pid>

Motive source
DF:212459




DevCorelli: # 17072 (BDUX10390): Removed dead code (SQLserver 6.5 import) from 
installer and portset

Date: Thu, 6 Feb 2003 09:44:23 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
After the media replacement the alternative way to import SQLServer dumps
is not used anymore.

Test Procedure
Full install from the new media




DevCorelli: # 17012 (HYUX10475): Informix:Error -1210 with 
pdaemon6.2:TCS-100/128821

Date: Mon, 3 Feb 2003 15:45:54 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Technical terms)
Error 2210 while starting pdaemon6.2.

Motive source
TCS:100-128821




DevCorelli: # 17008 (BDUX9880): DF211808 - shutdown/cleanup if parent goes away

Date: Mon, 3 Feb 2003 10:42:55 +0100

Created on:  DevCorelli
Type:        enhancement

Problem Description (Customer terms)
When bshell goes down (abnormally or) with SESSION_TIMEOUT reached then all
subprocesses (of the bshell) except apix.x goes away.  Additionally, apix.x
uses message queues which are still active when it's parent processes are gone.

Problem Description (Technical terms)
apix.x is started from the application via run.baan.prog function.
When processes are started this way, the bshell opens up a sh_server to manage
the external application.  When the bshell goes away, it takes care of it's
child processes, but not it's grandchildren.  Likewise, sh_server does not
currently have a mechanism to remove it's child processes.  Additionally, apix.x
will still have active message queues even if it is killed.  So these resources
are not freed up.

A later fix may be realized in the run.[baan.]prog function to dictate the
behavior of child (of sh_server) processes in case of abnormal exits.

Workaround
1)  (if possible) do not kill bshell, but exit normally
2)  do not use SESSION_TIMEOUT to kill processes which use apix.x
3)  manually kill apix.x process; and remove message queues

Test Procedure
Start an application session that will start apix.x.
Leave session open so that the apix.x remains active.
Tested by "killing" the bshell.
Also tested by allowing bshell to timeout vie SESSION_TIMEOUT set.
Verify process apix.x goes away.  Also verify that message queues are gone.

Motive source
DF:211808




DevCorelli: # 16997 (BDUX10471): Modification in comments

Date: Fri, 31 Jan 2003 14:38:58 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Fixed compilation problem caused by a typo
Test Procedure
Recompiled and verified the fix




DevCorelli: # 16994 (BDUX10339): Missing bind values in trace log

Date: Fri, 31 Jan 2003 13:17:34 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
If TT_SQL_TRACE option is specified with value 02000, generally the trace log 
contains the value(s) of the bind parameter(s) specified in the query, along with 
other information.
However, when the value of the bind variable remains unchanged for more than one 
bind, the trace log contains values only for the first bind. 
For other bind(s), the value(s) of the bind variable(s) is missing from the trace log.

Test Procedure
QPTool was tested with the following options:
TT_SQL_TRACE=02000 qptool6.2 -c 0 -apefbefbc -f "query file".
The query file contained the query: "select user, name from ttaad200 where user = :user".
This means that QPTool was run with actions intending to have two fetches for the value
specified in the bind variable "user".:

The following are snippets of the trace log, after the fix.

Tracing on: TT_SQL_TRACE = 02000
[SqlParse: before parse] SqlId : 40066008 - QID : 1 - Status : 0
Trans 1 (qptool) Mode 0 Query:
select user, name
from   ttaad200
where  user = :user

[SqlParse: after parse (e = 0)] SqlId : 40066008 - QID : 1 - Status : 2
[SqlExec before exec: (CompNr 000)] SqlId : 40066008 - QID : 1 - Status : 2
Automatic Bind for user
   Value(str): 'gerardv'
...

[SqlExec before exec: (CompNr 000)] SqlId : 40066008 - QID : 1 - Status : 7
Automatic Bind for user
   Value(str): 'gerardv'  	[This line was missing for second or more binds]	
...




DevCorelli: # 16907 (HYNT10200): L1 to L2 Migration Tool fails when a table has 
> 11 hash columns

Date: Tue, 28 Jan 2003 11:14:29 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
L1 to L2 migration tool fails when a table has more than 11 hash columns

Workaround
Drop the hash columns manually for the failed tables and reexecute the migration tool.

Test Procedure
Execute the tool on a table with more than 11 hash columns and it should succeed.

Motive source
TCS:100-129723




DevCorelli: # 16839 (BDUX10291): Updated yacc aix file to fix AS/400 compilation 
errors

Date: Wed, 22 Jan 2003 14:26:42 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Technical terms)
Updated the files qpc_yy_aix.c and qp_token_aix.h to fix compilations
errors on the AS/400




DevCorelli: # 16805 (BDNT10146): cd.read6.x scripts uses LANG variable

Date: Mon, 20 Jan 2003 15:06:37 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
cd.read6.x script uses LANG variable, SunOS platform gives error message on using this LANG variable as an internal variable. Changed the variable in LANGUAGE.

Test Procedure
Use cd.read6.x on a SunOS system, if you do not see the error message:
	"couldn't set locale correctly"
the scripts works ok




DevCorelli: # 16792 (BDUX10007): BaanSQL: Timestamp literal

Date: Mon, 20 Jan 2003 11:53:20 +0100

Created on:  DevCorelli
Type:        enhancement

Problem Description (Customer terms)
This project implements the following new date/time features to BaanSQL:

1. date literal
   For example: DATE '2002-12-31'

2. timestamp literal
   For example: TIMESTAMP '2002-12-31 13:32:45'
   Date and time are specified in local time, but internally
   represented as UTC.

3. current_date function

4. current_timestamp function

All these features comply with the ANSI SQL standard.


Motive source
IRD:58




DevCorelli: # 16780 (BDNT10213): Changed text in dialog

Date: Mon, 20 Jan 2003 09:08:01 +0100

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Customer terms)
Adjusted text in installer dialog, to prevent confusion

Problem Description (Technical terms)
Adjusted text in installer dialog, to prevent confusion




DevCorelli: # 16698 (BDNT10068): cd.read contains a construction, which is not 
valid for sun

Date: Tue, 14 Jan 2003 16:03:37 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
cd.read contains a construction, which is not valid for sun




DevCorelli: # 16615 (BDNT9952): Merged changed for deliver.exe (BDNT9903) to 
Corelli

Date: Thu, 9 Jan 2003 11:06:52 +0100

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
Deliver did not:
- build the cameo/ddc examples in the right way
- some settings from Reger were mistakely added in deliver
- the content.txt (bill of materials) file was not filled in any case (bug in windows?)

Workaround
- use an older deliver version, e.g. 2.27
- disable automatic build of ddc/cameo examples in deliver.cfg

Test Procedure
Run deliver with building cameo/ddc on. Check the content.txt file, and check that e.g. cameo1.exe was 
created in the destination path (under $DEST\bshell\api\bin)




DevCorelli: # 16567 (BDNT9977): Adjusted text in installer dialog for locales

Date: Tue, 7 Jan 2003 10:44:03 +0100

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Customer terms)
Adjusted text in locale dialog in installer. Made a reference to the documentation.
Without reference, it might be unclear for customers to choose the right locale.

Problem Description (Technical terms)
see above.

Test Procedure
run installer and check text in dialog for choosing the locale.




DevCorelli: # 16501 (BDUX9931): adjust copyright string

Date: Thu, 2 Jan 2003 11:39:34 +0100

Created on:  MaintBaanIVc
Type:        bugfix

Problem Description (Technical terms)
Copyrightstring is changed from 2002 to 2003




DevCorelli: # 16468 (BDNT9883): cd.read6.x: stopped prematurely while creating 
install.info, selecting of tools package not handled correctly

Date: Mon, 23 Dec 2002 13:44:02 +0100

Created on:  MaintCorelli
Type:        porting specific change

Problem Description (Technical terms)
cd.read6.x:
	stopped prematurely while creating install.info when only installing applications
	selecting of tools package not handled correctly when the tools package had the same vrc another package

Workaround
Allways start installing with a portingset cd, always select everything

Test Procedure
Do start installing a applications cd from that cd see if the script ends with: cd.read6.1: Ended
If so, it's ok




DevCorelli: # 16401 (BDNT9805): modify identing with quotes

Date: Tue, 17 Dec 2002 16:57:48 +0100

Created on:  DevEscher
Type:        bugfix

Problem Description (Technical terms)
The text after #ident needs quotes around it.

Workaround
where identstring is put, put quotes too (in deliver.exe)

Test Procedure
running modified deliver.exe with modified deliver.lst




DevCorelli: # 16354 (BDNT9776): Fix BOM file in NT deliver

Date: Fri, 13 Dec 2002 08:56:26 +0100

Created on:  DevEscher
Type:        bugfix

Problem Description (Customer terms)
BOM file were empty when COMPRESS-ed arhcive were created in the NT delivery.

Problem Description (Technical terms)
The destination path weren't found in the destination path of the copied file, because it were copied into a temporary place before compression.

Workaround
When compression flag is set, check for the temporary dir path also.

Test Procedure
Do a deliver, and check the BOM file's content.




DevCorelli: # 16348 (BDNT9771): Make all variants of deliver on all releases 
identical

Date: Fri, 13 Dec 2002 08:52:55 +0100

Created on:  MaintReger
Type:        bugfix

Problem Description (Customer terms)
Deliver were not identical on every releases

Problem Description (Technical terms)
Because of an enhancement project, a change weren't merge to all releases.

Workaround
Do it manually.

Test Procedure
Build a deliver executable.




DevCorelli: # 16325 (BDNT9743): Fix RUNPROGRAM command in the NT deliver.

Date: Thu, 12 Dec 2002 12:29:36 +0100

Created on:  DevEscher
Type:        bugfix

Problem Description (Customer terms)
Using the RUNPROGRAM command in the deliver.lst for an executable, which is reachable only in the path, hangs with error.

Problem Description (Technical terms)
The code tries to access the given file, but doesn't handle the path.

Workaround
Let this access check when a path is given in the RUNPROGRAM argument.

Test Procedure
Do a delivery on Gemini.




DevCorelli: # 16291 (BDUX9680): AS/400: enhance RSTFRMSTMF command

Date: Wed, 11 Dec 2002 10:09:47 +0100

Created on:  DevCorelli
Type:        enhancement

Problem Description (Technical terms)
When testing the multibyte bshell6.2 problems occuured when using the RSTFRMSTMF command. The command itself is a front-end for calling WRKO4OBJS.SRVPGM and the back-end for the RSTFRMSTMF.cmd menu. It is now possible to run WORKO4OBJS.pgm from the command line and the WRKO4OBJS.MODULE is compiled within so it does not need the shared library anymore:

  /qsys.lib/bdux9680.lib/WORKO4OBJS.pgm -h
   Usage:
           BDUX9680/WORKO4OBJS RST <Library> <Streamfile>
           BDUX9680/WORKO4OBJS SAV "<ListOfObjects>" <Library> "<ListOfObjTypes>" <Streamfile>
           BDUX9680/WORKO4OBJS CPYTO  <Library/SaveFile> <StreamFile>
           BDUX9680/WORKO4OBJS CPYFRM <Library/SaveFile> <StreamFile>


Test Procedure
Run WORKO4OBJS to convert objects to stream and vice versa.




DevCorelli: # 16284 (BDNT9697): Fix some problems on NT delivery

Date: Tue, 10 Dec 2002 13:58:45 +0100

Created on:  DevEscher
Type:        bugfix

Problem Description (Customer terms)
Deliver.exe doesn't give error messeage, when the example compilations failed or a variable were not defined.

Problem Description (Technical terms)
msdev.exe doesn't give an error code back when it fails to open a project file.

Workaround
Put the error messages to standart error and check for compilation success.

Test Procedure
Do a nightly build with DDC and CAMEO examples enabled.




DevCorelli: # 16278 (BDUX9719): Merge of delivery binary changes

Date: Tue, 10 Dec 2002 13:40:00 +0100

Created on:  DevCorelli
Type:        enhancement

Problem Description (Technical terms)
Delivery.exe needs to be the same in all versions

Workaround
Merge From Project BDNT9273




DevCorelli: # 16192 (BDNT9649): Repair messages in Event viewer after fatal exit 
of installation of an update

Date: Tue, 3 Dec 2002 13:19:08 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Customer terms)
Event viewer is showing 'unregistered source baan' after running an update (CD).

Problem Description (Technical terms)
Event Viewer do not have baan_msg.dll registered after an update was ran, causing unknown source/categories in event log.

Test Procedure
Run update from installer and check that the 'unknown source' disappeared in the Event Viewer.




DevCorelli: # 16184 (BDUX9639): UNIX installer fix

Date: Mon, 2 Dec 2002 16:37:36 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
datab.mas and displ.mas were not deleted after installation.




DevCorelli: # 16054 (HYNT9535): MSQL: Remove L1 to L2 migration tool from the 
porting set delivery list

Date: Wed, 20 Nov 2002 18:21:39 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
L1 to L2 migration tool should be removed from the porting set delivery list.

Test Procedure
After installing the new porting set and the directory %BSE%\bin\migrate should not be seen.




DevCorelli: # 16014 (BDNT9448): cd.read6.x sometimes shows tools package twice

Date: Tue, 19 Nov 2002 08:55:23 +0100

Created on:  MaintCorelli
Type:        bugfix

Problem Description (Technical terms)
Sometimes the cd.read script shows the tools package twice

Test Procedure
Run this script for a cd where the tools vrc is the same as the apps vrc




DevCorelli: # 15965 (HYNT9420): MSQL:Fix for binary columns in L1toL2Migration 
tool

Date: Thu, 14 Nov 2002 10:32:51 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Customer terms)
We seem to have a problem when converting binary data. 
It seems as though all binary data is getting converted to TinyInt.  
In general, this is usually the proper conversion, but in ttttxt010 
the original binary data is actually text.  
When it is converted to TinyInt it converts it to an improper data type 
and the text is lost. 

Problem Description (Technical terms)
The Fast Migration tool did not consider the existence of msql_binarycolumns file
which overrides the Data dictionary to convert the database data type to binary.
This leads to binary columns of size greater than 1 to exist in database.

We also have binary columns of size greater than 1 when an array column is compressed.
But, compression is a very rare scenario in MSQL driver as the Max number of columns 
in a table is 1024, and Max string length is 8000.

Workaround
use bdbpre / bdbpost to migrate from Level 1 database to Level 2 database.

Test Procedure
Apply the L1 to L2 migration tool on a company number having tables which 
have an entry in msql_binarycolumns. The binary columns of length greater than 
1 is left untouched.




DevCorelli: # 15919 (BDNT9402): Deliver IE-Wrapper with Corelli BW for E-boks 
help

Date: Tue, 12 Nov 2002 09:09:00 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Customer terms)
For Corelli backend (BaanERP 50c) Eboks should be supported to show Baan Help information.
Therefore the BW installation must be extended with an IE-wrapper application (iewrapperdde.exe).

Problem Description (Technical terms)
Make sure the iewrapperdde.exe file is also delivered with the Corelli version of BW.

Workaround
Install this file manually.

Test Procedure
Start Baan Help from a Corelli Backend which has been updated for eboks help.




DevCorelli: # 15873 (BDNT9368): cd.read stops if the df command gives error

Date: Wed, 6 Nov 2002 16:49:15 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Technical terms)
cd.read stops if the df command gives error

Test Procedure
can only be tested if df gives error on the system




DevCorelli: # 15830 (HYNT9203): MSQL:Multiple instances support to the installer

Date: Fri, 1 Nov 2002 17:17:13 +0100

Created on:  DevCorelli
Type:        bugfix

Problem Description (Customer terms)
Installer does not support multiple SQL Server instances. 

Problem Description (Technical terms)
This means that BaanERP can not be installed straight on a named instance of SQL Server.
Currently msql7adm.dll doesn't use extended SQL-DMO objects to utilize the multiple 
instances of SQL Server 2000.  Changes are needed in the installer scripts as well as
the msql7adm.dll.

Workaround
1. Install Baan on the default instance of SQL Server 2000.
2. Now copy the baan database to the named instance of user's choice.
3. For each database entry in the Database definitions session (ttaad4110m000 session ) 
   that has to point to the named instance of SQL Server, in the parameter field,  
   modify the MSQL_SERVERHOST entry as below:
 
                MSQL_SERVERHOST=<computername>\\<instancename>
 
   for ex: MSQL_SERVERHOST= NEDERLANDS\\BARNEVELD  where NEDERLANDS is the name of the 
   machine and BARNEVELD is the name of a SQL Server instance.
4. Perform a 'convert to runtime'
5. Logoff and logon to the Baan environment


Test Procedure
Try the following test scenarios:
1. Number of of instances: 2 (SQL 2000default instance + SQL 2000 named instance)- level2 
	- Install Baan on the named instance
2. Number of of instances: 2 (SQL 2000 default instance + SQL 2000named instance) - level1 
	-Install Baan on the default instance
3. Number of instances: 1 (SQL 2000 named instance only) - level1 
4. Number of instances: 1 (SQL 2000 default instance only) - level2



Motive source
Scopus defect 119024




DevCorelli: # 15795 (BDUX9299): New version number for 7.1e

Date: Tue, 29 Oct 2002 14:17:14 +0100

Created on:  DevCorelli
Type:        porting specific change

Problem Description (Technical terms)
New version number for portingset (7.1e) and BW (7.1e.152)



Total number of changIDs displayed: 203