thanx a lot tom, stuart.
I think using CURRENT_TIMESTAMP will be better, as it will not have the
overhead of a function call.
I have one more problem. I am using postgreSQL 7.1.3 from a C application.
I am using asyncronous query processing. If the postgreSQL server is shut
or crashed while the application is running, is there any way i can know
that something went bad at the backend.
On Thu, 28 Nov 2002, Tom Lane wrote:
> HK <harikrishnan(at)midascomm(dot)com> writes:
> > I am able to execute stored procedures. But sometimes i get this message.
> > ERROR: fmgr_info: cache lookup for language 1952725 failed
> Sounds like you dropped the plpgsql language while there were still
> functions that used it. Not a good thing to do.
> (7.3 will prevent you from making this class of mistake, btw)
> > Inside a stored procedure i am using 'now' for a datetime value.
> > Wat happens is when the stored procedure is loaded, the 'now' is
> > initialised and afterwards whenever i use this procedure, the same value
> > is being inserted.
> Use now() or CURRENT_TIMESTAMP instead of 'now'. The latter gets
> reduced "on sight" to a timestamp constant.
> regards, tom lane
In response to
pgsql-novice by date
|Next:||From: David C.Oshel||Date: 2002-11-29 20:17:06|
|Subject: 7.3 and Mac OS X 10.2.2|
|Previous:||From: Oliver Elphick||Date: 2002-11-29 05:02:33|
|Subject: Re: joins|