List of changes for portingset: 9.0b ------------------------------------------------------------------------ Elgar: # 65619 (BDNT59420): repgen: number of fields allocated is off by one Date: Wed, 06 May 2015 09:27:23 +0200 Created on: Elgar Type: bugfix _Problem Description (Customer terms)_ mlog and minidump found after running repgen _Test Procedure_ reports should compile without minidumps _Affected Executables_ repgen6.2 ------------------------------------------------------------------------ Elgar: # 65595 (BDNT59391): Repgen generate wrong 3GL code Date: Fri, 01 May 2015 14:16:34 +0200 Created on: Elgar Type: bugfix _Problem Description (Customer terms)_ Reports complain: Error in reading sort file record: 1 fields scanned, xxyyzz expected _Affected Executables_ repgen6.2 _Motive source_ TCS:800-203891 ------------------------------------------------------------------------ Elgar: # 65534 (BDUX59267): Adding CDFs to input fields and remove report input field limit Date: Fri, 24 Apr 2015 15:39:52 +0200 Created on: Elgar Type: enhancement _Problem Description (Customer terms)_ CDF fields cannot be added to a report without recompiling the report. _Affected Executables_ bshell6.2 repgen6.2 _Motive source_ JIRA:LND2-3936 JIDA:LND2-1156 ------------------------------------------------------------------------ Elgar: # 65139 (BDNT58754): bdbreconfig crash. Date: Mon, 23 Mar 2015 16:47:34 +0100 Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Range set to 1 is no range but the value true. _Workaround_ N/A _Test Procedure_ Create a .pd with a RANG:1 _Affected Executables_ bdbreconfig _Motive source_ InforXtreme:488923 ------------------------------------------------------------------------ Elgar: # 65131 (BDNT58775): rexecd generates minidumps Date: Mon, 23 Mar 2015 14:43:58 +0100 Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Rexecd crashes on windows _Test Procedure_ tested at customer site _Affected Executables_ rexecd.exe ------------------------------------------------------------------------ Elgar: # 65123 (BDUX58807): tss handling different between old and new portingset Date: Fri, 20 Mar 2015 16:13:24 +0100 Created on: Elgar Type: bugfix _Problem Description (Customer terms)_ When Installing new SP on system with locale GB2312 on language 2 labels. Retrieving error 205 during the process. _Workaround_ work with some different locale _Test Procedure_ Try to insert a high ascii character that is not available in the used multi byte locale. _Affected Executables_ All database drivers in BaanIV mode ------------------------------------------------------------------------ Elgar: # 65035 (BDUX58470): LND2-2992 No export/import required if column changes from SB to MB with lower/UPPER Date: Fri, 13 Mar 2015 14:35:12 +0100 Created on: Elgar Type: enhancement _Problem Description (Customer terms)_ Changing an upper case column from SB to MB triggered an export/import cycle. When the data is pure 7-bit ASCII, this is not necessary. Also, the MSQL driver considered a SB column a different datatype than MB, but the underlying actual database type was the same. This was considerd a column-type change and ALSO triggered an export-import. This project fixes both problems. _Affected Executables_ bdbreconfig _Motive source_ JIRA:LND2-2992 ------------------------------------------------------------------------ Elgar: # 64972 (BDUX58516): Send Install Locale to BW Date: Mon, 09 Mar 2015 13:44:22 +0100 Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ BW shows unexpected characters with high_ascii_tolerance_strict_Mode depending on the User Locale. _Workaround_ keep user locale and installation locale the same. _Test Procedure_ Use different locale for installation and user settings, and check how the characters are shown _Affected Executables_ bshell _Motive source_ InforXtreme:487134 ------------------------------------------------------------------------ Elgar: # 64488 (BDUX57857): e error High Ascii character not allowed for local 88591 Date: Mon, 26 Jan 2015 16:13:43 +0100 Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Despite setting being high_ascii_tolerance:0, this error is occuring when trying to run the standard session Calculate Cost Price by project: " High ASCII characters not allowed for this field. Change the Search Key." Clearing it to blank to z's still does not allow the session to run. _Workaround_ Use locale ISO_BIN _Test Procedure_ Start the given session ( or others with the same kind of field ) where the max is set. _Affected Executables_ bshell _Motive source_ InforXtreme:484756 ------------------------------------------------------------------------ Elgar: # 64193 (BDUX57720): Oracle driver: disable FETCH FIRST ROWS because of yet another bug Date: Tue, 16 Dec 2014 11:27:57 +0100 (CET) Created on: Danzi Type: porting specific change _Problem Description (Customer terms)_ Oracle 12 has implemented the FETCH FIRST ROWS ONLY clause. This clause can be used to limit the result set for queries with the "as set with rows" clause. Without the FETCH FIRST clause the Oracle optimizer might choose a less optimal execution plan, as the result set is not limited. We implemented this clause in our oracle driver, but the Oracle implementation is far too limited to be usable. Oracle internally rewrites the query, causing ambiguous columns. The limitations have been reported to Oracle, but they do not acknowledge the limitations as bug, but they do not want to document the limitations either. Anyway, the limitations make it impossible for us to use this feature that has proven to boost performance on other databases. _Affected Executables_ ora_srv ------------------------------------------------------------------------ Elgar: # 64049 (BDUX57545): Installer: enable row versioning for fresh installations Date: Fri, 5 Dec 2014 11:41:39 +0100 (CET) Created on: Elgar Type: enhancement _Problem Description (Customer terms)_ Enable row versioning for fresh installs. After an install the db_resource file should contain the following setting: bdb_use_row_version:1 All database tables should be extended with the rcd_vers column. _Test Procedure_ Do fresh installs using the 9.0b porting set. Verify that the bdb_use_row_version:! setting is in the $BSE/lib/defaults/db_resource file. Verify that the tables have the rcd_vers column. _Affected Executables_ db2_adm msql_adm ora_adm _Motive source_ JIRA:LNTECH-4084 ------------------------------------------------------------------------ Elgar: # 64033 (BDUX57562): Wrong desktop size used in ASM 1.9 and rexecd 9.0 Date: Thu, 4 Dec 2014 17:27:33 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ ASM 1.9 and rexecd of portingset 9.0 on Windows create a Windows (non-interactive) desktop of just 128Kb, where 8Mb should be the default. Symptoms may be: failure to start bshells if the number of users increases a certain level. ------------------------------------------------------------------------ Elgar: # 63952 (BDUX57487): Oracle driver: partially enable FETCH FIRST n ROWS for Oracle 12 Date: Thu, 27 Nov 2014 16:36:00 +0100 (CET) Created on: Elgar Type: porting specific change _Problem Description (Customer terms)_ Oracle 12 has only partially implemented the FETCH FIRST n ROWS ONLY functionality _Test Procedure_ regression tests _Affected Executables_ ora_srv ------------------------------------------------------------------------ Elgar: # 63946 (BDUX57440): Apps Studio: Return dalhookerror returns multiple levels up Date: Thu, 27 Nov 2014 14:50:06 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Consider the following code example: function long f.1(long a) { return(a) } function long f.2(long a) { return(f.1(a)) } function long f.3(long a) { return(f.2(a)) } function main() { long a a = f.3(3) } When using the debugger at function f.1() at the return statement pressing 'n' jumps 'directly' back to the call point at f.3() The return statements of f.2() and f.3() are executed but there is no way to inspect local variables after the most inner return of f.1() has been executed /BAAN3GL _Affected Executables_ bic6.2 _Motive source_ TCS:800-200922 ------------------------------------------------------------------------ Elgar: # 63944 (BDNT57371): The debuggers cursor is not in sync with the script when running a script in debug mode. Date: Thu, 27 Nov 2014 14:37:54 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Repgen replaced an \r with \n while looking for comment concatenate sign "^" _Workaround_ Convert script to unix format. _Test Procedure_ Used script: declaration: long value1, value2, value3 string word1(5), word2(5), word3(5) before.program: value1 = 1 value2 = 2 value3 = 3 |Commentje word1 = "test1" word2 = "test2" word3 = "test3" _Motive source_ InforXtreme:475437 ------------------------------------------------------------------------ Elgar: # 63931 (BDNT57243): StagingWizard: Staging Wizard forces reboot Windows Server Date: Thu, 27 Nov 2014 07:44:45 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ The Staging Wizard sometimes forces a reboot, without choice between 'reboot now' and 'reboot later'. And after reboot, the Staging Area may be corrupt. _Affected Executables_ StagingWizard.exe _Motive source_ JIRA:LNTECH-4135 ------------------------------------------------------------------------ Elgar: # 63917 (BDNT57425): change several vsprintf to vsnprintf calls Date: Tue, 25 Nov 2014 17:18:06 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ During a defect an extreme large query caused the bshell to crash. The query was malformed. _Test Procedure_ none _Affected Executables_ all _Motive source_ InforXtreme:475259 ------------------------------------------------------------------------ Elgar: # 63906 (BDUX57422): mb.coerce not handling part of MB char Date: Tue, 25 Nov 2014 11:26:31 +0100 (CET) Created on: Couperin Type: bugfix _Problem Description (Customer terms)_ If a multibyte character is entered in the Weight Unit field, an BW message comes up. This issue is only reproducibed in the multibyte environment. _Workaround_ Only handle complete 4 bytes multibyte characters _Test Procedure_ function main() { long ret string rjr1(4) string rjr2(3) rjr1=chr$(155)&chr$(189)&chr$(182)&chr$(215) rjr2=rjr1 ret = mb.coerce.to.sb(rjr2) } return value should be -1 _Affected Executables_ bshell _Motive source_ InforXtreme:479161 ------------------------------------------------------------------------ Elgar: # 63896 (BDNT57421): Recurring characters when printing Hebrew Report to display Date: Tue, 25 Nov 2014 10:23:49 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ When printing Hebrew Report on the screen (device D) there are characters recurring. This happens when scrolling through the pages. _Affected Executables_ bw.exe _Motive source_ Infor Xtreme: 473663 ------------------------------------------------------------------------ Elgar: # 63856 (BDUX57360): DF:478632, TEST_RETRY should stay enabled after abort transaction(). Date: Thu, 20 Nov 2014 15:49:07 +0100 (CET) Created on: Couperin Type: bugfix _Problem Description (Customer terms)_ db.retry.point fails when we set TEST_RETRY and there is a abort.transaction in the code _Workaround_ do a commit.transaction after the abort _Test Procedure_ set retry point. do update action. do abort action do update action do commit and check if we still go to retry point. _Affected Executables_ bshell _Motive source_ InforXtreme:478632 ------------------------------------------------------------------------ Elgar: # 63813 (BDUX57351): trap on alloc.mem calls when a static array is passed Date: Tue, 18 Nov 2014 16:35:20 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ passing a static array to a function which does an alloc.mem the bshell can crash. Although this is a programming error on the 3GL side, the bshell does not detect this case _Motive source_ JIRA:LNTECH-4089 ------------------------------------------------------------------------ Elgar: # 63718 (BDUX57230): oracle: parallel execution of CREATE INDEX and ANALYZE TABLE Date: Tue, 11 Nov 2014 11:14:35 +0100 (CET) Created on: Danzi Type: porting specific change _Problem Description (Customer terms)_ This project introduses a new resource ora_parallel_degree (default value 4). If this resource has a value >= 1, a PARALLEL clause will be added to the CREATE INDEX command. The analyze table function gets a degree=> option. _Affected Executables_ ora_srv ------------------------------------------------------------------------ Elgar: # 63641 (BDUX57038): Improved stopping SHM if data segment has gone Date: Tue, 4 Nov 2014 16:21:53 +0100 (CET) Created on: Danzi Type: porting specific change _Problem Description (Customer terms)_ Crash when calling shmmanger6.2 -k when a bshell is still running. _Workaround_ 1. Stop all running bshells before stopping SHM. 2. After core dump: manually cleanup SHM ids, by using system tools (ipcs -m and ipcrm -m). The used SHM IDs are also logged in $BSE/log/log.shmmanger6.2. _Test Procedure_ 1. Start SHM 2. Connect with bshell to the BSE 3. Stop SHM. 4. Observe that stopping SHM does not destroy SHM (neither the data nor the base SHM segment) as long as a client process is using the BSE's SHM. Note that each BSE has two SHM segment: a. the base segment (small size) b. the data segment (default 96Mb). _Affected Executables_ shmmanager6.2 ------------------------------------------------------------------------ Elgar: # 63507 (BDNT57005): add version.lib for all tastes Date: Mon, 27 Oct 2014 11:43:28 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Add version.lib to release mode ------------------------------------------------------------------------ Elgar: # 63497 (BDUX57001): Core dump when attach failed Date: Mon, 27 Oct 2014 09:35:28 +0100 (CET) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ SHM manager crashes on Unix if attach failed. _Workaround_ NA. _Test Procedure_ Happens in the occasional situation if SHM attach fails. _Affected Executables_ shmman6.2 _Motive source_ Other (reported by email by ASG during system verification of 10.4 (9.0a portingset) ------------------------------------------------------------------------ Elgar: # 63482 (BDUX56942): T800-200056: compress.exe: exit status 11 (access violation) on PMC solution Date: Fri, 24 Oct 2014 11:47:28 +0200 (CEST) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ During the installation of Apps solutions with PMC some compress actions are initiated by the process. In the monthly dump of Apps (May 2014) at least one of the compress action fails. _Test Procedure_ Run compress -f on the file. _Affected Executables_ compress _Motive source_ T800-200056 ------------------------------------------------------------------------ Elgar: # 63478 (BDNT56971): startfirst needs version in dialogs Date: Fri, 24 Oct 2014 10:18:21 +0200 (CEST) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ No version in dialog header _Motive source_ TCS:800-199970 ------------------------------------------------------------------------ Elgar: # 63290 (BDNT56802): creat.tmp.file\$: Revert to using forward slash as last separator Date: Fri, 10 Oct 2014 10:56:23 +0200 (CEST) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ In porting set 8.9a.04 the function creat.tmp.file$() was changed to return a backward slash as last separator on Windows. This change is now reverted because of the possible impact it may have on existing installations. The creat.tmp.file$() function will return a forward slash as last separator. _Test Procedure_ Both on Unix/Linux and Windows use the creat.tmp.file$() and observe that the last separator is a forward slash. A regression test has been made to verify this behavior. _Affected Executables_ bshell ------------------------------------------------------------------------ Elgar: # 63201 (BDUX56565): D471209: Use of zero-width Thai character results in error 205 Date: Thu, 2 Oct 2014 16:37:52 +0200 (CEST) Created on: Couperin Type: bugfix _Problem Description (Customer terms)_ If zero-width characters are used, then the derived search key may contain too many characters. When the search key is inserted into the database, this results in error 205. _Test Procedure_ On a Unicode environment, start session tccom4500m000. Open a business partner, set the Description field with a string that begins with a Thai character with an accent (2 characters) and fill the string with ASCII characters. Saving the record must succeed, and error 205 should not be returned. A regression test has been created to verify the behavior. _Affected Executables_ bshell _Motive source_ InforXtreme:471209 ------------------------------------------------------------------------ Elgar: # 63185 (BDUX56687): Block using incorrect error constants by 3GL Date: Thu, 2 Oct 2014 10:35:20 +0200 (CEST) Created on: Elgar Type: enhancement _Problem Description (Customer terms)_ All (system) error constants for error numbers from 33 - 93 are not allowed anymore. These constants are system specific. The current macro names are misleading/incorrect. _Workaround_ Adjust code and interpret all errors != 0 as error. _Test Procedure_ Compile all code and observe that no error constants are used that are now forbidden. _Affected Executables_ none ------------------------------------------------------------------------ Elgar: # 63176 (BDNT56657): BW crashes with dialog.add.listbox Date: Wed, 1 Oct 2014 15:01:34 +0200 (CEST) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ BW crashes with dialog.add.listbox using options DLG_FIELD_TYPE, DLG_TYPE_COMBOBOX. _Motive source_ InforXtreme:469650 ------------------------------------------------------------------------ Elgar: # 63166 (BDNT56663): Mapped SQL error to BDB error Date: Wed, 1 Oct 2014 10:08:45 +0200 (CEST) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Full database not correctly reported to end-user. _Workaround_ NA. _Test Procedure_ Find a full database and try to insert data. _Motive source_ TCS:800-199238 ------------------------------------------------------------------------ Elgar: # 63031 (BDNT56410): ASCII-interface (ttaad4100) cursor position, data entry issues. Date: Tue, 23 Sep 2014 15:41:58 +0200 (CEST) Created on: Couperin Type: bugfix _Problem Description (Customer terms)_ Characters not echoed on screen in some ascii sessions in baanIV. _Workaround_ n/a _Test Procedure_ Start ttaad4100 on baanIV and check if typed characters are echoed on screen. _ _ bugfix _Motive source_ InforXtreme:463373 ------------------------------------------------------------------------ Elgar: # 62981 (BDUX56466): D470523: Query with hint give error with new portingset Date: Thu, 18 Sep 2014 16:04:13 +0200 (CEST) Created on: Couperin Type: bugfix _Problem Description (Customer terms)_ query with hint will give error with new portingset: [4] Syntax error: 'hint' not expected [0] Found 1 errors during scanning/parsing .. Stopped. sql_parse failed _Workaround_ Use older portingset or no hint functionality _Test Procedure_ Execute the next query in B50b environment: Code-chang* from cprpd100 where cprpd100._index1={1} hint use index 1 on cprpd100 _Type _ bugfix _Affected Executables_ bshell _Motive source_ InforXtreme:470523 ------------------------------------------------------------------------ Elgar: # 62977 (BDUX56475): T800-918667: Error 506 in log.reconfig Date: Thu, 18 Sep 2014 14:59:27 +0200 (CEST) Created on: Couperin Type: bugfix _Problem Description (Customer terms)_ During a Fresh installation of Infor LN 10.3 the following errors occur in the log.reconfig file: 5228: 2014-09-11[13:37:27(UTC-02:00)]:bsp: Clear reference counters error 506 on table tppdm600000 5230: 2014-09-11[13:37:27(UTC-02:00)]:bsp: Clear reference counters error 506 on table tppin010000 _Test Procedure_ Do a Fresh installation. The log.reconfig file should not contain the above error messages. _Affected Executables_ refint _Motive source_ TCS:800-198667 ------------------------------------------------------------------------ Elgar: # 62947 (BDNT56441): startcommand.exe doesnot work fine Date: Wed, 17 Sep 2014 08:38:35 +0200 (CEST) Created on: Danzi Type: bugfix _Problem Description (Customer terms)_ Batch files in start directory don't work. ------------------------------------------------------------------------ Elgar: # 62935 (BDUX56421): oracle 12 not mentioned in version string Date: Tue, 16 Sep 2014 14:26:05 +0200 (CEST) Created on: Couperin Type: porting specific change _Problem Description (Customer terms)_ ora_srv -v does not mention Oracle 12.1 as supported version. _Test Procedure_ ora8_srv -v should now list 12.1 as well. _Affected Executables_ ora8_srv ------------------------------------------------------------------------ /*Total number of changIDs displayed: 37*///// //