IDS Forum

RE: 4GL run-time error number -4339

Posted By: mosserp@wellsfargo.com
Date: Thursday, 11 March 2010, at 6:43 p.m.

In Response To: RE: 4GL run-time error number -4339 (LARRY SORENSEN)

> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> LARRY SORENSEN
> Sent: Wednesday, March 10, 2010 2:54 PM
> To: ids@iiug.org
> Subject: RE: 4GL run-time error number -4339 [19306]
>
> The load is handled by a separate process; each record is entered by a
> program
> one at a time and committed each time. Once the claim is in, the
> Traffic Cop
> moves the record along a series of Java programs; it is the final step
> in the
> services that spawns out the thread to a 4ge program.
>
> The 4ge program reads data from the Informix database and creates a 4GL
> report
> (that is later printed and sent out. It updates one record in the one
> table
> which indicates that the report has been completed and saves a
> compressed copy
> of the report to the another table.
>
> Each set of programs handles one record at a time and begins and
> commits work
> as appropriate. I believe that the problem arises when the process that
> calls
> the 4ge program continues on to the next record while the 4ge is still
> running. As a result we get a situation where multiple 4ge threads are
> running
> for different record reports and a wall might be hit; I don't know if
> the wall
> relates to just the Shared Memory limits or if it has to do with the
> number of
> open files being written to or some other limiting factor but once that
> wall
> is hit no more reports are created for a while. Then "magically" the
> reports
> start up again until the wall is again hit. So when you look at the
> first
> record table you see a number of entries where the record_id is set
> (indicating the report was created and saved to the correct table).
> Then there
> will be a series of records that don't have a record_id, then some with
> record_ids, and so one. So it appears that the original process
> continues to
> spawn out the threads but they are killed during or just before the 4ge
> program is run.
>

Larry,

This is a bit of a long shot, but... Do you see any long transaction rollbacks in the message log when things come to a halt?? What is your "set lock mode to wait ..." set to?

HTH,
Paul M.

Messages In This Thread

• RE: 4GL run-time error number -4339
  LARRY SORENSEN -- Tuesday, 9 March 2010, at 10:26 a.m.
  • Re: 4GL run-time error number -4339
    Fernando Nunes -- Tuesday, 9 March 2010, at 10:37 a.m.
    • RE: 4GL run-time error number -4339
      LARRY SORENSEN -- Tuesday, 9 March 2010, at 12:16 p.m.
      • Re: 4GL run-time error number -4339
        Fernando Nunes -- Tuesday, 9 March 2010, at 12:33 p.m.
  • RE: 4GL run-time error number -4339
    Andrew Lennard -- Tuesday, 9 March 2010, at 5:35 p.m.
  • Re: 4GL run-time error number -4339
    Art Kagel -- Wednesday, 10 March 2010, at 8:42 a.m.
    • RE: 4GL run-time error number -4339
      LARRY SORENSEN -- Wednesday, 10 March 2010, at 11:54 a.m.
      • RE: 4GL run-time error number -4339
        LARRY SORENSEN -- Wednesday, 10 March 2010, at 12:49 p.m.
        • Re: 4GL run-time error number -4339
          Jonathan Leffler -- Wednesday, 10 March 2010, at 4:20 p.m.
          • RE: 4GL run-time error number -4339
            LARRY SORENSEN -- Wednesday, 10 March 2010, at 4:53 p.m.
            • Re: 4GL run-time error number -4339
              Art Kagel -- Wednesday, 10 March 2010, at 7:48 p.m.
              • RE: 4GL run-time error number -4339
                LARRY SORENSEN -- Thursday, 11 March 2010, at 12:26 p.m.
                • Re: 4GL run-time error number -4339
                  Art Kagel -- Thursday, 11 March 2010, at 4:12 p.m.
            • RE: 4GL run-time error number -4339
              mosserp@wellsfargo.com -- Thursday, 11 March 2010, at 6:43 p.m.
      • Re: 4GL run-time error number -4339
        Art Kagel -- Wednesday, 10 March 2010, at 1:54 p.m.
        • RE: 4GL run-time error number -4339
          LARRY SORENSEN -- Wednesday, 10 March 2010, at 3:59 p.m.
          • Re: 4GL run-time error number -4339
            Art Kagel -- Wednesday, 10 March 2010, at 5:47 p.m.

IDS Forum is maintained by Administrator with WebBBS 5.12.