[ProFTPD-committers] [Bug 3619] Exiting sessions don't seem to die properly

[b***@horde.net]

http://bugs.proftpd.org/show_bug.cgi?id=3619

TJ Saunders <***@castaglia.org> changed:

What |Removed |Added
----------------------------------------------------------------------------
Keywords| |Backport
Status|NEW |ASSIGNED
CC| |***@ordal.com,
| |***@secure-computing.net
| |, ***@debian.org,
| |***@maytech.net,
| |***@city-fan.org,
| |***@io.com

[b***@horde.net]

http://bugs.proftpd.org/show_bug.cgi?id=3619

--- Comment #1 from TJ Saunders <***@castaglia.org> 2011-03-15 18:51:46 EDT ---
Created attachment 3576
--> http://bugs.proftpd.org/attachment.cgi?id=3576
Fixes bug

This fixes the issue that I've seen, which is:

pr_event_generate("core.exit", ...)
<exit listener A>
pr_event_generate("core.exit", ...)
<exit listener A>
pr-event_generate("core.exit", ...)

i.e. the same event gets handled by the same listener, over and over.

A more proper fix would have a "memory" of the last N events generated, rather
than just one. But this makes things better now; I will work on improvements.

[b***@horde.net]

http://bugs.proftpd.org/show_bug.cgi?id=3619

TJ Saunders <***@castaglia.org> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED

--- Comment #2 from TJ Saunders <***@castaglia.org> 2011-03-15 19:00:33 EDT ---
Patch committed to CVS, and backported to 1.3.3 branch.

[b***@horde.net]

http://bugs.proftpd.org/show_bug.cgi?id=3619

--- Comment #3 from TJ Saunders <***@castaglia.org> 2011-03-19 14:31:15 EDT ---
Created attachment 3581
--> http://bugs.proftpd.org/attachment.cgi?id=3581
Additional necessary patch

This patch is also necessary; the previous patch created a condition where the
same event, generated twice (without any event loops), would not be properly
published to all listeners again.

[b***@horde.net]

http://bugs.proftpd.org/show_bug.cgi?id=3619

TJ Saunders <***@castaglia.org> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |CLOSED

--- Comment #4 from TJ Saunders <***@castaglia.org> 2011-04-06 03:53:27 EDT ---
Resolved in 1.3.4rc2.