On May 6, 2008, at 7:36 AM, Lampa wrote:
Hello,
i know but my "hack" should solve this problem. In some way i have 5
vpopmaild connections but only one was active, others are death and
not closed - causing 100% cpu usage.
Try telnet to vpopmaild then login and terminate telnet, then watch
your cpu load on machine with vpopmaild.
Note that Net_Vpopmaild::__destruct() calls quit(). So it *should*
always call it, even when exceptions are thrown. However, I do
occasionally find stale vpopmaild processes running. So either I'm
wrong in the above statement (although I've not been successfully
proving that through testing Net_Vpopmaild), or it could be a
vpopmaild problem.
Regards,
Bill
