Graham Leggett writes:
> Does postgresql have an access log where the lifetime of database
> connections are tracked? Specifically tracking when the client
> connected, and how long the client stayed connected for.
See log_connections and log_disconnections. You probably want to
add PID to l
On 04 Jul 2010, at 12:49 AM, Tom Lane wrote:
I'd bet that you are using a client-side driver that feels a need to
fetch metadata about the tables it's working with. JDBC or ODBC
maybe?
A reasonable driver would cache such information, so if there are
enough
of these to present a performa
Graham Leggett writes:
> I am currently trying to analyse some performance problems within my
> postgresql v8.4.4 database, and after successfully turning sql
> statement logging on, I discover thousands and thousands of queries
> being made what looks like an internal postgresql set of tabl
Hi all,
I am currently trying to analyse some performance problems within my
postgresql v8.4.4 database, and after successfully turning sql
statement logging on, I discover thousands and thousands of queries
being made what looks like an internal postgresql set of tables in a
database cal
Richard Broersma wrote:
> I'd like to propose the following items be added to the todo list:
> (Any thoughts?)
>
> 1) Modify composite types to allow ALTER DOMAIN(s) to ADD CONSTRAINT.
> 2) Allow a since ALTER DOMAIN issue multiple ADD and DROP commands in
> a single statement.
This is not curre
Wang, Mary Y wrote:
> Hi,
>
> I'm using a psql -f command to reload the data from a dump file. I
> noticed that some tables are not populated with any rows (I mean 0
> rows), yet, if I manually insert a row (actually just copy an INSERT
> statement from that input file) in the interactive termina
On Jul 2, 12:39 am, pie...@hogranch.com (John R Pierce) wrote:
> On 07/01/10 2:27 PM, wrote:
>
> > Sorry I forgot to say I am running an Italian php website with a
> > postgresql DB. Users can write messages eachothers.
> > This is my code:
> > $connessione_db = pg_connect("dbname=x us
On Sat, Jul 03, 2010 at 09:35:56AM +0800, Craig Ringer wrote:
> On 02/07/10 21:26, saitenhe...@web.de wrote:
>
> > #ifdef PG_MODULE_MAGIC
> > PG_MODULE_MAGIC;
> > #endif
>
> Why the conditional compilation of PG_MODULE_MAGIC?
That's the recommendation, so the module compiles on all versions of
P
On 2 July 2010 23:27, A.M. wrote:
> Hello,
>
> I have encountered an odd behavior involving rules which the following script
> demonstrates (in postgresql 8.4.3). Note that at the end of the run, the
> "dud" table contains one row "spam1" when the update rule clearly contains
> two inserts to t
