Re: [PATCH] Allow breaking out of hung connection attempts

On Mon, Jan 09, 2012 at 10:35:50AM +0200, Heikki Linnakangas wrote:
> On 08.01.2012 22:18, Ryan Kelly wrote:
> >@@ -1570,7 +1570,13 @@ do_connect(char *dbname, char *user, char *host, char *port)
> >  		keywords[7] = NULL;
> >  		values[7] = NULL;
> >
> >-		n_conn = PQconnectdbParams(keywords, values, true);
> >+		if (sigsetjmp(sigint_interrupt_jmp, 1) != 0) {
> >+			/* got here with longjmp */
> >+		} else {
> >+			sigint_interrupt_enabled = true;
> >+			n_conn = PQconnectdbParams(keywords, values, true);
> >+			sigint_interrupt_enabled = false;
> >+		}
> >
> >  		free(keywords);
> >  		free(values);
> 
> That assumes that it's safe to longjmp out of PQconnectdbParams at
> any instant. It's not.
I'm guessing because it could result in a resource leak?

> I think you'd need to use the asynchronous connection functions
> PQconnectStartParams() and PQconnectPoll(), and select().
New patch attached.

> 
> -- 
>   Heikki Linnakangas
>   EnterpriseDB   http://www.enterprisedb.com

-Ryan Kelly