Re: Embedded C SQL Error -600 - Mailing list pgsql-interfaces

From Christoph Haller
Subject Re: Embedded C SQL Error -600
Date
Msg-id 3E6F50F2.46B61E20@rodos.fzk.de
Whole thread Raw
In response to Embedded C SQL Error -600  (Mike Alford <mikea@syscon-intl.com>)
List pgsql-interfaces
>
> Using  PostgreSQL 7.3.1 on a customized Linux slackware host 8.0.0
> I am receiving an error -600 from this bit of code:
>
>     EXEC SQL FETCH FROM test_cursor
>         INTO :mach_seq, :ss_mseq, :child_idx, :job, :net_pieces,
:start_time;
>     printf ("b4 while test_cursor_select SQLCODE %d\n", SQLCODE);
>     while (SQLCODE != ECPG_NOT_FOUND)
> //  while (!SQLCODE)
>         {
>         EXEC SQL FETCH FROM test_cursor
>         INTO :mach_seq, :ss_mseq, :child_idx, :job, :net_pieces,
:start_time;
>         EXEC SQL FETCH FROM test_cursor
>         INTO :mach_seq, :ss_mseq, :child_idx, :job, :net_pieces,
:start_time;
>         }
>
> in ecpgerrno.h this entry:
> /* backend WARNINGs, starting at 600 */
> #define ECPG_WARNING_UNRECOGNIZED          -600
>  /* WARNING:  (transaction aborted): queries ignored until END */
> implies that a transaction with this code would be aborted, as well as
not
> being too informative about the nature of the problem.
>
> The commented line '//  while (!SQLCODE)'  fits the style of existing
code
> better.
>
> I get the output I would expect if there were no problem:
> b4 while test_cursor_select SQLCODE -600
> i 0 machseq 73374, ssmseq 3404562 child 0 job 410F 3              ,
net
> pieces 0, start_time 2003-02-10 11:00:00
> i 0 machseq 73374, ssmseq 3404562 child 1 job 410F 3              ,
net
> pieces 254, start_time 2003-02-10 11:00:00
> i 0 machseq 73374, ssmseq 3404562 child 2 job 410F 3              ,
net
> pieces 508, start_time 2003-02-10 11:00:00
> i 0 machseq 73374, ssmseq 3404562 child 3 job 410F 3              ,
net
> pieces 762, start_time 2003-02-10 11:00:00
>
> Is this an error that can be ignored? If not, what should I be looking
for?
>
No, do not ignore errors. The bit of code does not show if
it's enclosed within a transaction block (BEGIN - END). Is it?

Regards, Christoph




pgsql-interfaces by date:

Previous
From: "Hiroshi Inoue"
Date:
Subject: Re: [HACKERS] Roadmap for FE/BE protocol redesign
Next
From: Peter Eisentraut
Date:
Subject: Re: Automatic detection of PostgreSQL version