Re: [PATCHES] log session end - again

Also, Andrew, on your other patch for log_session_info line, Magnus had
the idea of giving each session an id based on the first transaction of
the session. That seems like a useful addition to that variable, though
I am not sure what value to give a session before it executes its first
query.

---------------------------------------------------------------------------

Andrew Dunstan wrote:
>
>
> Bruce Momjian wrote:
>
> >Andrew Dunstan wrote:
> >
> >
> >>[redirecting to hackers]
> >>
> >>If you like. I originally just made it happen if log_connections was
> >>set, but Neil wanted a separate setting for it. What is the consensus
> >>about a name?
> >>
> >>
> >
> >log_disconnect?
> >
> >
>
> Not bad. Maybe for symmetry "log_disconnections" would be better.
>
> Here's what the log lines look like:
>
> for a local connection:
> [21340] LOG: session end: duration: 0:00:23.41 user=andrew
> database=template1 host=[local]
>
> for a tcp connection:
> [21351] LOG: session end: duration: 0:00:20.22 user=andrew
> database=template1 host=127.0.0.1 port=33419
>
> for a tcp connection when log_hostname = true:
> [21373] LOG: session end: duration: 0:00:02.25 user=andrew
> database=template1 host=alphonso port=33421
>
> Do we want to change those too? Maybe "session end: duration:" should be
> "disconnection: session time:" or some such.
>
> cheers
>
> andrew
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 3: if posting/reading through Usenet, please send an appropriate
> subscribe-nomail command to majordomo(at)postgresql(dot)org so that your
> message can get through to the mailing list cleanly
>

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073