Re: droped out precise time calculations in src/interfaces/libpq/fe-connect.c

Tom Lane wrote:
> Joe Conway <mail(at)joeconway(dot)com> writes:
> > [ some convincing test cases that timeout=1 is not good ]
>
> > remains.tv_sec = atoi(conn->connect_timeout);
> > + if (remains.tv_sec == 1)
> > + remains.tv_sec += 1;
> > if (!remains.tv_sec)
> > {
> > conn->status = CONNECTION_BAD;
>
> On pure-paranoia grounds, I'd suggest the logic
>
> + /* force a sane minimum delay */
> + if (remains.tv_sec < 2)
> + remains.tv_sec = 2;
>
> whereupon you could remove the failure check just below.

I think we should fail if they set the timeout to zero, rather than
cover it up by setting the delay to two.

Attached is a patch that implements most of what we have discussed:

use time()
get rid of timeval where not needed
allow restart of select() to properly compute remaining time
add 1 to timeout == 1
pass finish time to pqWaitTimed

Patch applied. I am applying it so it is in CVS and everyone can see
it. I will keep modifying it until everyone likes it. It is just
easier to do it that way when multiple people are reviewing it. They
can jump in and make changes too.

--
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