Re: PRO*C Bus Error (core dumped), but why??!

From: Jack A. Rabah <rabahya_at_my-dejanews.com>
Date: 1999/04/15
Message-ID: <7f49f2$1os$1_at_nnrp1.dejanews.com>#1/1


HI ALL, It turns out that pro*c dies if the EXEC SQL statement is too long! When I split the INSERT statement below using the line continuation "\", the file precomplied without any error. Wouldn't it be nice if proc generated an error "file.pc: Error -- Embeded SQL Statement too long on line: xx" rather than generating a Bus Error? :)

later,
jack

In article <7f34dc$2rr$1_at_nnrp1.dejanews.com>,   Jack A. Rabah <jack_at_idsintl.com> wrote:
> Ron,
>
> Thanks for the reply. I just noticed that proc is creating .cod and .cud
> files when it gives the bus error.. I guess it doesn't have the chance to
> clean up its work.. the .cud files are empty and the .cod files contain what
> seems to be the result of the precompiling. The last statement it was trying
> to precompile seems to have been the INSERT statement in the following
> function:
>
> InsertRowProg(ptr)
> EXEC SQL BEGIN DECLARE SECTION;
> DbProgStruct *ptr;
> EXEC SQL END DECLARE SECTION;
> {
>
> if (GetSerial(&ptr->pg_key) != DB_SQL_SUCCESS)
> return DB_SQL_NOT_FOUND;
>
> EXEC SQL INSERT INTO program (pg_key, ppid, prog_desc, revision, rele
> ase_date, bins, tests, indexs, conditions, sampling_method, sampling_rate,
> statu s_key, eqt_key, accept_data, res_aging_days, stats_aging_days,
> prog_type, pgc_ke y, wmap_key, rework_action, wafer_index, diex_index,
> diey_index, bin_index) VALU ES (:ptr->pg_key, :ptr->ppid, :ptr->prog_desc,
> :ptr->revision, :ptr->release_dat e, :ptr->bins, :ptr->tests, :ptr->indexs,
> :ptr->conditions, :ptr->sampling_metho d, :ptr->sampling_rate,
> :ptr->status_key, :ptr->eqt_key, :ptr->accept_data, :ptr ->res_aging_days,
> :ptr->stats_aging_days, :ptr->prog_type, :ptr->pgc_key, :ptr-> wmap_key,
> :ptr->rework_action, :ptr->wafer_index, :ptr->diex_index, :ptr->diey_i ndex,
> :ptr->bin_index);
>
> Do you see anything wrong there? Is there any limitation on the length of
> embeded SQL statements?
>
> later,
> jack
>
> Ron Reidy <rereidy_at_uswest.net> wrote:
> > Jack,
> >
> > Pro*C sometimes core dumps when there is a call to an invalid (or
 non-existent)
> > PL/SQL package, procedure, or function. Check your code for any of this.
> >
> > If this does not work, you will have to resort to calling Oracle WW Support.
> >
> > rr
> >
> > "Jack A. Rabah" wrote:
> >
> > > Hi,
> > >
> > > How can I find out why pro*c generated a bus error and core dumps without
 any
> > > error message?
> > >
> > > [phoenix] generic/> proc sqlcheck=full auto_connect=yes code=ansi_c
> > > define=__ORACLE__ def_sqlcode=yes mode=ansi user=testdb/testdb db_prog.ec
> > >
> > > Pro*C/C++: Release 8.0.4.0.0 - Production on Wed Apr 14 8:41:29 1999
> > >
> > > (c) Copyright 1997 Oracle Corporation. All rights reserved.
> > >
> > > System default option values taken from:
> > > /u01/app/oracle/product/8.0.5/precomp/admin/pcscfg.cfg
> > >
> > > Bus Error (core dumped)
> > > [phoenix] generic/>
> > >
>
> ---
> Jack A. Rabah
> IDS Software Systems
> jack_at_idsintl.com
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
>

--
---
Jack A. Rabah
IDS Software Systems
jack_at_idsintl.com

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    
Received on Thu Apr 15 1999 - 00:00:00 CEST

Original text of this message