Re: Reporting query duration - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: Reporting query duration
Date
Msg-id 200208301614.g7UGEGU15466@candle.pha.pa.us
Whole thread Raw
In response to Re: Reporting query duration  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Tom Lane wrote:
> Bruce Momjian <pgman@candle.pha.pa.us> writes:
> >> I think the log entries should be separate: print the query text when
> >> you start, print the duration when you're done.  A little bit of
> >> postprocessing can reassemble the two entries.
> 
> > How would someone reassemble them?  You would have to have the pid, I
> > assume.
> 
> Sure.
> 
> > Do we auto-enable pid output when we enable duration?  Yuck.
> 
> No, you expect the user to select the options he needs.  Let's not
> over-engineer a perfectly simple feature.

OK, so I will rename debug_print_query to log_print_query, and Gavin's
new print query on error will also be a log_*.

I will add a new GUC variable to print the query duration, and recommend
in the docs that log_pid be enabled at the same time so you can match
the duration with the query.

Originally, I wanted to make the time just print whenever you enabled
print_query, but with them on separate lines, it should be a separate
GUC variable.

--  Bruce Momjian                        |  http://candle.pha.pa.us pgman@candle.pha.pa.us               |  (610)
359-1001+  If your life is a hard drive,     |  13 Roberts Road +  Christ can be your backup.        |  Newtown Square,
Pennsylvania19073
 


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: contrib features during beta period
Next
From: Bruce Momjian
Date:
Subject: Re: make failure on cvs tip