Those are good questions. I suspect that the session within
POE::Component::Client::Ping is remaining active, and so your program
keeps running after your custom session ends.
This may be a bug in POE::Component::Client::Ping with user-provided
sockets. I'd like to try your code and dig into the problem myself.
--
Rocco Caputo - [EMAIL PROTECTED]
On Feb 22, 2007, at 10:32, Michael Hare wrote:
I've done a little more digging and now I'm really confused.
Consider the following '_stop' handler for a Client::Ping session.
sub handler_stop {
print "Session ", $_[SESSION]->ID, " has stopped.\n";
print $poe_kernel->get_active_event() . " - " .
$poe_kernel->get_event_count() . "\n";
#$poe_kernel->stop();
}
---------------------------------
When I bring my own socket, this happens on call of '_stop'
Session 3 has stopped.
_stop - 0
(app spins)
-------------------------------
Shouldn't kernel->run return based on the following comment?
# Run the event loop, only returning when it has no more sessions to
# dispatch events to. Supports two forms.
$poe_kernel->run();
--------------------------------
If I add the kernel->stop(), my app 'seems' to work [as root; haven't
messed with setuid yet]. I'd say 95% of the time, get_event_count
returns 0 in the _stop handler, but sometimes it returns a positive
integer; how could the _stop handler be called if the
get_event_count is
non zero, though?
"_stop is sent to a session when it's about to stop. This usually
occurs
when a session has run out of events to handle and resources to
generate
new events."
Also, POE::Kernel docs say
"stop() has been added as an experimental function to support forking
child kernels with POE::Wheel::Run. We may remove it without notice if
it becomes really icky. If you have good uses for it, please mention
them on POE's mailing list."
So I don't want to become dependent on a feature that is hacking
around
a real problem or may disappear!
-Michael
Michael Hare wrote:
Hello-
Firstly, I want to thank all of the responses I've received thus far.
I'm still learning system level programming so some of these ideas
are
at first foreign.
I'm having problems understanding how to use the 'bring your own
socket'
feature of Client::Ping. I'm investigating using my own sockets
so that
I can open a socket as root and then setuid to the desired user.
If I take working code where Client::Ping provides the socket and
then
just change it so that I provide the socket (and stay the root user
even), after everything is 'done', $poe_kernel->run() never
exits. If I
enable DEBUG I see unrelated packets coming after all of my expected
testing is done. I included a '_stop' handler and that -is-
getting run.
Is this problem because there is still a socket open? If so, is
there a
way to deal with this? Closing the socket would not be ideal
because I
explicitly opened the socket as root and setuid to my desired
'utility'
user in the begging on the program. My goal is to be able to run the
poller in a loop so it would be nice to reuse the socket after a
successful completion of $poe_kernel->run();
I can post the code I'm using if that helps. Sorry if this is a
rather
basic question.
Thanks-
-Michael
