|
Hi Al, I would order the latest Group PTF for SQL for your release of OS/400 (SF99501 for V5R1) to get up to date on that as a first step. To really debug this you are going to need a joblog, no ifs ands or buts. If it is just entering the program that causes this, maybe you can reproduce it yourself and collect what you need to show IBM that way. Otherwise, you'll have to get strict with the users and make them save this (tell them SIGNOFF *LIST for example if this happens and call you immediately). If the program has not changed, it could be an issue in the SQL runtime on the iSeries. The SQLCA is the SQL 'communications area' which is used at runtime to store SQL variables that your program is sending to the SQL runtime layer and is where the database layer can send errors back to your program. You can get SQL dumps for various reasons, but so far I have only seen these kinds of errors caused by runtime bugs in SQL itself - normally other types of SQLCA issues are caught at compile time when the SQL pre-compiler runs or when the SQL runtime gets an error back from database in the SQLCA, a much nicer SQLnnnn type of message is then sent to your program to document normal, expected errors (like 'Result of Select returned more than 1 row' etc.). Base BPCS CD has no SQL, so I imagine this is one of your modified parts. You can read more about the SQLCA here if you are really interested: http://publib.boulder.ibm.com/infocenter/db2luw/v8/index.jsp?topic=/com. ibm.db2.udb.doc/admin/r0002212.htm Thanks, Genyphyr Novak Senior System Software Engineer SSA Global R&D message: 3 date: Fri, 24 Feb 2006 04:12:14 -0600 from: Al Mac <macwheel99@xxxxxxxxxxx> subject: [BPCS-L] SQL Trace Dump in Shop Order Launch What I seek: Clue me * what the heck this is; * what it supposed to be telling me; * where to look in a program, what could trigger it. AS/400: model 170 OS/400: V5R1 BPCS: 405 CD Our AS/400 resources seem quite healthy. What tends to go down a lot is ma bell and the electric company. My users tell me this is now happening when they launch shop orders. They supply me with some screen prints that are unintelligible to me, and they apparently delete any other useful clues, other than the reports they need. ... I am lucky to get what I have got. There's several ways to launch shop orders. I launch a few to seek a sample of what's going on, but it works fine for me. This is a program that needs to be run several times a day (different people managing the work for different customers) ... I spy on joblogs of the folks generating the orders, but I not yet catching this in the act. They are supposed to run it off our default JOBQ (only one person at a time updating some of the stuff). I do not know how long this has been going on how consistent it is what if anything else is wrong ... like perhaps they get the output desired, and this dump along the way A few months ago we altered how we launch shop orders, and how we report labor against them, to save on paper and clerical time. I not know if this stuff started at the same time, or is unrelated. They have supplied me with a "User Trace Dump" from QTEMP/QAP0ZDMP consisting of data, supposedly from QSQLSRV that means nothing to me from SQLCA section: 00000056: then a 6 digit decimal # from 236872 to 717680 then D39DEDD291 then a 6 character hexadecimal # from 018A09 to 018A90 then a bunch more hexadecimal strings, most of them all zeros The dump is 130 print positions wide,of which my users supplied first 100 The program is heavily modified, approx 20% of source code came with system & 80% what we (me) done over the years. Last modified Sep 2005, tested then, working fine. When this has blown up in the past, either it has been a typo during more mods, or some of our data has gone beyond prior extremes or ceilings. This is an RPG/400 program with embedded SQL/400 which is used both through cursors and direct, to get * customer name, that the production is for * data from various engineering and shop order related files such as revision #, calculate rates, facility, type & class of item, work center, dept It also calls several sub-programs to get BOM active where-used array and identify key components. Compilation approx 150 pages. Base program size (execution object) = 700 k It also has a DDS, data structure, and the connected programs Nothing involved here modified for many months. Most recent modifications have been to unrelated software. At year end, we reset next shop order at # 1, and since then have reached 9,000 Order # 200,000 and up is still open from last year Inventory history, smallest shop order # after 10,000 is 60,000 Labor history smallest shop order # after 10,000 is 50,000 so we are 4 months away from a collision there - Al Macintyre http://www.ryze.com/go/Al9Mac Find BPCS Documentation Suppliers http://radio.weblogs.com/0107846/stories/2002/11/08/bpcsDocSources.html BPCS/400 Computer Janitor at http://www.globalwiretechnologies.com/ Replacement company web site (same company, new domain) http://www.globalwti.com/
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2024 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact [javascript protected email address].
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.