Re: Strange Windows problem, lock_timeout test request

2013-01-30 17:45 keltezéssel, Zoltán Böszörményi írta:
> 2013-01-30 16:06 keltezéssel, Hari Babu írta:
>> On Wednesday, January 30, 2013 7:59 PM Zoltán Böszörményi wrote:
>>>> 2013-01-28 15:20 keltezéssel, Hari Babu írta:
>>>> 2. regress check failed because the expected ".out" file is not
>>>> updated properly.
>>> Which regress check failed? The .out file was updated in the patch for
>> prepared_xacts.sql where >the regression tests for lock_timeout were
>> added.
>> Or do you mean the one for the sql file you >sent?
>>
>> During regress test, "prepared_xacts_1.out" expected file used for
>> comparing
>> with the result file. Which is not updated by the patch. Because of this
>> reason the regress check is failing.
>
> I see, so this is a Windows-only change that needs a different.
> Can you send the resulting prepared_xacts_1.out file so I can integrate
> its changes into my patch? That way it would be complete.
> Thanks in advance.

I have found a little time to look into this problem and
found a way to make pg_regress use prepared_xacts_1.out.
I had to change line 2193 in pg_regress.c from

     fputs("max_prepared_transactions = 2\n", pg_conf);

to

     fputs("max_prepared_transactions = 0\n", pg_conf);

The patch now passed "make check" in both cases.

Best regards,
Zoltán Böszörményi

--
----------------------------------
Zoltán Böszörményi
Cybertec Schönig & Schönig GmbH
Gröhrmühlgasse 26
A-2700 Wiener Neustadt, Austria
Web: http://www.postgresql-support.de
      http://www.postgresql.at/