Github user shinrich commented on a diff in the pull request:
https://github.com/apache/trafficserver/pull/766#discussion_r71044678
--- Diff: iocore/net/UnixNetVConnection.cc ---
@@ -1139,8 +1139,8 @@ UnixNetVConnection::mainEvent(int event, Event *e)
(write.vio.mutex && wlock.get_mutex() != write.vio.mutex.get())) {
#ifdef INACTIVITY_TIMEOUT
if (e == active_timeout)
-#endif
e->schedule_in(HRTIME_MSECONDS(net_retry_delay));
+#endif
return EVENT_CONT;
--- End diff --
This appears to be the more critical fix. It looks like a standard
reschedule if you don't get the lock. But it appears that we only do this if
the event is an active timeout (if INACTIVITY_TIMEOUT is set). It seems that
we are not currently building with INACTIVITY_TIMEOUT defined (at least I'm
not). So this change would cause the reschedule to never happen. So if you
don't get the lock the first time, you will never process the event.
I'm not following the description in the JIRA. Is the concern that we are
double freeing an event object?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
