I checked this code under linux, all works fine, and the memory usage fastly stabilize. But under Win2k, with Php 4.2.0 it's always the same thing.

Please try current CVS HEAD and report back.

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".

socket_set_nonblock ($socket);
		while (1) {
		    sleep(1);
			if (($buf = socket_read($socket, 1024)) !== FALSE ) {
				$buf = NULL;
				unset($buf);
			}
		}
--> that code works fine.
As the original reporter says, if you remove sleep(1) it leaks memory. Approximately 1Mb of memory every few seoncds here.

This is against 4.2.1

Please test against http://snaps.php.net/php4-latest.tar.gz, compile with --enable-debug and if the extension really leaks memory please paste the output after php ends (you will get a leak list [kind of] if there's really a leak).

no output after php terminates. the script is as follows:
<?
error_reporting (E_NONE);
$socket = socket_create (AF_INET, SOCK_STREAM,  SOL_TCP);
$moo = socket_connect ($socket, "www.php.net", "80");
	if ($moo)
	{
		socket_set_nonblock ($socket);
		while ($i<2000000) {
			$i++;
			if (($buf = socket_read($socket, 1024)) !== FALSE ) {
				$buf = NULL;
				unset($buf);
			}
		}
	}
?>

(add a sleep(1) into the while loop and it's fine...)

reclassified where this belongs..

It the same things. I try now whith PHP 4.2.1

(sorry, I can't compile with the last version)

Please try the latest snapshot at http://snaps.php.net/win32/php4-win32-latest.zip instead, this one contains the current development code.

I've just test with PHP 4.2.1 (always under Win2k Pro).

And there is always the same pb...

As suggested, please try http://snaps.php.net/win32/php4-win32-latest.zip

markus, the comment at 7june,2:32 pm was against the latest snapshot. 

PHP did not output anything when it terminated.
The code in that example was 'broken'. 
The fixed version of that code was to add a sleep(1).
Adding a usleep(100000)(i.e. 1sec), still exhibits the broken behaviour. 
Adding a sleep(1) command 'fixes' the memory issue. 

Paul

Yes, sorry... I tried.

But it doesn't work... It's seem to be exactly like before...

This bug has been fixed in CVS. You can grab a snapshot of the
CVS version at http://snaps.php.net/. In case this was a documentation 
problem, the fix will show up soon at http://www.php.net/manual/.
In case this was a PHP.net website problem, the change will show
up on the PHP.net site and on the mirror sites.
Thank you for the report, and for helping us make PHP better.