I'm having a problem with MySQL timing out after somewhat extensive usage
(nothing compared to what it should be able to handle though).

I'm running NetBSD with PHP 4.0.5 and 11.15 Distrib 3.23.39, for
-netbsd.  This is running on an m68k, which I doubt (but could) is an
issue, using Apache Apache/1.3.20.

I have several Web sites running on this server, using PHP and MySQL.

At first, these sites used one mysql persistent connection each (in a
header file).  When the sites were visited frequently, however, (in
testing, maybe three dozen page views in the space of several minutes),
they would proceed to time out across the board -- iow, all pages using
MySQL would simply time out.

At this time, an unusual number of MySQL sockets were open, due to the
number of children Apache is allowed (one hundred).  It should be noted
that I put an exec timeout on the php of one second, which had no effect
(as waiting for external processes does not qualify as time-outable time).

I then replaced the persistent connections with standard ephemeral
connections, to no avail.  Now, although I have the same problem, the
number of mysql sockets open (according to netstat -an | grep mysql)
hovers around five instead of fifty.

If I leave things alone for a while, a netstat -an reveals five mysql
pipes open, but things return to normal (pages can be loaded with data
from MySQL).

Anyone have any ideas?

Mike Hudack

"You can cage the singer but not the song."
        -- Harry Belafonte

"I like my terminals in blue, please."

PHP Database Mailing List (http://www.php.net/)
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
To contact the list administrators, e-mail: [EMAIL PROTECTED]

Reply via email to