CONTACTS 

SUBJECTS
Prior Date Record Crashing in Schedule Gary                    04
Johnson, Gary Prior Date Diary Not Archiving Pointers Correctly May
Prior Date Diary Not Archiving Pointers Correctly May be Problem of

1505 -
1505 -    ..
1506 - Summary/Objective
1507 -
150701 - Follow up ref SDS 2 0000, ref SDS 1 I29O.
150702 -
150703 - Gary had a problem launching the SDS program, ref SDS 0 GX6L, that
150704 - turned out to be related to a recent system improvement. ref SDS 0
150705 - PZ5L  Initially did not grasp what Gary was explaining; so, called and
150706 - he clarified issues to be resolved. ref SDS 0 PQ8Y  First step was to
150707 - eliminate problems to the Schedule caused by defective code.
150708 - ref SDS 0 FV4N  Corrected the code, ref SDS 0 YI49, but did not send
150709 - to Gary, because a major upgrde will be submitted in a few days with
150710 - improvements Gary has suggested. Sent a letter explaining how to
150711 - recover and that I will call tomorrow to verify Gary is up and
150712 - running.
150713 -
150714 -     [On 040419 called Gary and explained how to solve problem.
150715 -     ref SDS 6 LS9G
150717 -      ..
150718 -     [On 040422 submitted corrected code to Gary. ref SDS 7 RT4L
150719 -
150720 -
150721 -
150722 -
150723 -
150724 -
150726 -  ..
1508 -
1509 -
1510 - Problem
1511 -
151101 - Received, ref DRT 1 0001 dated 040416, a letter from Gary with a zip
151102 - file containing his SDS records.
151104 -  ..
151105 - Gary says...
151106 -
151107 -    1.  On the web are:
151108 -
151109 -             •  my current records
151111 -  ..
151112 - Downloaded Gary's records from the web.  The Schedule is for 040415,
151113 - rather than today 040416??
151115 -  ..
151116 - Gary continues...
151117 -
151118 -             •  Backup files for the last 2 days
151120 -                 ..
151121 -             •  Prior_Date_Error.txt which explains the problem and
151122 -                what I did about it, which still hasn't fixed it.
151124 -         ..
151125 -    2.  We will be running errands, and I know that you have other
151126 -        things to do, so possibly some time over the weekend we can
151127 -        tackle this.
151128 -
151130 -         ..
151131 -        SDS Crashed Creating Prior Date Records
151132 -
151133 -    3.  This is a simple test file, since SDS is down.
151135 -  ..
151136 - Looking at Gary's records for 040415 there is no evident problem; need
151137 - to work with Gary to resolve.
151139 -  ..
151140 - Gary continues...
151141 -
151142 -    4.  The FSS Documents task, along with delays in getting SDS back
151143 -        resulted in many record that needed to be captured, and a
151144 -        series of scheduling meeting records that simply aren't worth
151145 -        the trouble.
151147 -         ..
151148 -    5.  I created records to document each class, and started through
151149 -        my written notes populating the records with meeting data.
151151 -         ..
151152 -    6.  I then began creating sets of linked Prior Date records for the
151153 -        missing meetings, including all those standing meetings that
151154 -        had been canceled.
151156 -         ..
151157 -    7.  This exercise was enough to confirm my view that the process
151158 -        for creating Prior Date records needs to be made much easier
151159 -        than it is.
151160 -
151162 -         ..
151163 -        SDS Archive Failed
151164 -
151165 -    8.  After a number of these records, the attempt to archive failed.
151166 -
151167 -            [On 040419 called Gary and explained how to solve problem.
151168 -            ref SDS 6 LS9G
151170 -         ..
151171 -    9.  Symptoms and Actions:
151172 -
151173 -            1.  The launch of SDS failed with an error -- Too many parameters - .
151174 -
151175 -            2.  The Group Manager file had my name blanked out.  I
151176 -                replaced the file with a copy of mine.
151178 -                 ..
151179 -            3.  The Schedule file, D:\sd\08\gljdy\00\00 was empty.  SDS
151180 -                prompted to create a new schedule. I escaped and
151181 -                replaced the 00 file from the most recent backup.
151183 -                 ..
151184 -            4.  The schedule showed the pending record creation, but
151185 -                the date was missing. I replaced the date.
151187 -                 ..
151188 -            5.  I looked at the new record, file 1158. The heading line
151189 -                was messed up. The fields up to the user name had been
151190 -                overwritten by a time stamp plus some garbage. I copied
151191 -                a heading from another record and fixed the line in the
151192 -                1158 file.
151194 -         ..
151195 -   10.  I created a records transfer to take home
151197 -         ..
151198 -   11.  I copied the last 2 days of backup directories from the E:
151199 -        drive to take home.
151201 -         ..
151202 -   12.  I put everything on the memory stick.
151204 -         ..
151205 -   13.  I created this file and put it on the memory stick.
151206 -
151207 -
151209 -  ..
1513 -
1514 -
1515 - 1621
1516 -
151601 - Called Gary.
151602 -
151603 - Explained the records received from him are through 040415, per above.
151604 - ref SDS 0 PPWT  There is no indication about any activity he has done
151605 - on creating a prior date record on 040416.
151607 -  ..
151608 - Gary advised there was a problem assembling a transfer.
151610 -  ..
151611 - He assembled another transfer after updating his home compter with
151612 - SDS records and files from the office as of 040416.  He then uploaded
151613 - to the web.
151615 -  ..
151616 - There were a number of problems in the Schedule file....
151617 -
151618 -    1.  Multiple SAA records listed.
151619 -
151620 -        The first SAA listing is missing the * needed for a prior date
151621 -        op.
151623 -         ..
151624 -        The 2nd SAA listing does not have a dot to identify a diary
151625 -        record.
151626 -
151627 -            [On 040419 called Gary and explained how to solve problem.
151628 -            ref SDS 6 LS9G
151630 -         ..
151631 -        There is a 3rd SAA listing at the bottom of the list of about
151632 -        20 pending tasks.
151634 -         ..
151635 -    2.  The last date in the file was missing the month string,
151636 -        somehow it got this short string got removed on the last line
151637 -        of text in the file.
151639 -  ..
151640 - Fixed these problems.
151642 -  ..
151643 - Since Gary's Schedule 00 file is dated 040416, we need to get it to
151644 - update to the date today, 040418, then create the records for a prior
151645 - date.
151647 -  ..
151648 - Made corrections, and then ran SDS and opend Gary's records.  This
151649 - returned an error message:
151650 -
151651 -                        Too many parameters...
151653 -  ..
151654 - Opened the Schedule without archiving and found some of the same
151655 - errors reported above. ref SDS 0 FV6S
151657 -  ..
151658 - Told Gary I would try to get this fixed, and call him this evening OA
151659 - 1900.  Gary said to go ahead and call, and if his situation permits
151660 - working on this matter we can do so, otherwise will do it first thing
151661 - in the morning.
151662 -
151663 -            [On 040419 called Gary and explained how to solve problem.
151664 -            ref SDS 6 LS9G
151665 -
151666 -
151667 -
151669 -  ..
1517 -
1518 -
1519 - 1956
1520 -
152001 - Researching Gary's 00 file, there is a slight difference between his
152002 - Schedule and mine, in that my 00 file has a dot in col 6 for SAA and
152003 - Gary's does not.
152005 -  ..
152006 - Will run Gary's 00 file in debug mode with the backup and global
152007 - pointer ops unplugged so that only SDS code runs.  The thing that is
152008 - suspect is SDS code may be trying to automatically extend Gary's
152009 - Schedule, and it may have a defect, that was not previously noticed.
152011 -  ..
152012 - This is a Schedule initialization issue.
152013 -
152014 -
152015 -
152016 -
1521 -

SUBJECTS
Dot Col 6 and Col 15 Changed Method for Identifying in Schedule Curr
Extend Schedule Automatically
Schedule Maintenance 008501 Develop List of Files on Disk in 00 Dire

1905 -
190601 -  ..
190602 - 04702  Automatically Extend Schedule 00 File
190603 -
190604 - Follow up ref SDS 5 HJ5H.
190605 -
190606 - This seems like a possible cause of the problem Gary reports, because
190607 - it is a recent improvment and is a little complex, which means it
190608 - could be a use case that was not tested.
190610 -  ..
190611 - Actually, it turns out that the Schedule may not be automatically
190612 - extended.  There is ability to extend the Schedule when the current
190613 - date is not in the record, but this is a rare occurrence.
190614 -
190615 -        Review of the record on 031217 shows code was created to
190616 -        automatically extend the Schedule by 6 months when less than 3
190617 -        months are in the Schedule, and this is checked each time SDS
190618 -        is started for the first time every day. ref SDS 4 EW4K
190620 -  ..
190621 - The real issue is that within the past few months on 031203 changed
190622 - the design for archiving current diary records in the Schedule.
190623 - Previously this was based on a dot in col 15, which is entered when
190624 - the user presses F8. ref SDS 3 VP4I  This was changed to archive a
190625 - task description and the associated Schedule file based on when there
190626 - is also a dot in column 6, which is only entered when a current diary
190627 - record is saved with F2 so that the pointers are written to the disk,
190628 - because this enables a diary record to be listed in the Diary Summary,
190629 - shown in macro file 0702053. ref OF 3 XW41
190631 -  ..
190632 - Line 1260, ref OF 2 U16L, -label bRno_curr in 04702
190633 -
190635 -  ..
190636 - Line 1200, ref OF 2 YK5H, -label 96rk
190637 -
190638 -    -label 96rk
190639 -    loc_cur 0 6
190640 -    ic 46
190641 -
190642 -        Added loc_cur 0 6, ic 46 to force a dot on SAA record that
190643 -        identifies it has been archived.
190644 -
190645 -
190646 -
190648 -  ..
190649 - Line 1300, ref OF 2 ZI4O, -label no_curR in 04702
190650 -
190651 -    -label no_curR
190652 -    @c:\sd\03\008
190653 -
190654 -        Extend Schedule another month.
190655 -
190657 -  ..
190658 - Testing with debug shows that when there is not a dot in col 6 of the
190659 - SAA file, 0001, then it is not archived, and we have a mess.  This
190660 - explains how Gary's SAA record shows up in multiple places.  The code
190661 - that updates the Schedule moves SAA forward, and then other code
190662 - creates a new one, and that causes a loop, which results in the error
190663 - message of "Too many parameters," per above. ref SDS 0 PY3L
190665 -  ..
190666 - This issue arises from recent improvements to archive performed tasks
190667 - based on a dot on col 6, rather than col 15, in order to ensure that
190668 - performed tasks have been saved to archive the pointers, which is
190669 - necessary for the diary sumary to function correctly. ref OF 2 T58F
190670 -
190671 -            [On 040419 called Gary and explained how to solve problem.
190672 -            ref SDS 6 LS9G
190674 -  ..
190675 - Why has this never happened before?
190677 -  ..
190678 - If the only way that SAA gets a dot is to open the file and save
190679 - it, why is there a dot in the SAA for user 00101 and not for gljdy?
190680 - How does one user get it and the other not.
190682 -  ..
190683 - Perhaps it got entered from past work, and has remainded for 00101,
190684 - and for some reason not evident at this time, the dot got removed for
190685 - gljdy, and this got replicated, which then collided with the new code
190686 - that moves none archived records to the next date.
190687 -
190688 -
190689 -
190690 -
190691 -
190692 -
190693 -
190694 -
190695 -
190696 -
190697 -
1907 -