BIND 10 #2790: Lettuce tests timing & missed messages
BIND 10 Development
do-not-reply at isc.org
Fri Mar 15 08:17:22 UTC 2013
#2790: Lettuce tests timing & missed messages
-------------------------------------+-------------------------------------
Reporter: vorner | Owner:
Type: defect | Status: new
Priority: medium | Milestone:
Component: Unclassified | Sprint-20130319
Keywords: | Resolution:
Sensitive: 0 | CVSS Scoring:
Sub-Project: Core | Defect Severity: N/A
Estimated Difficulty: 9 | Feature Depending on Ticket:
Total Hours: 0 | Add Hours to Ticket: 0
| Internal?: 0
-------------------------------------+-------------------------------------
Comment (by vorner):
Hello
Replying to [comment:10 jinmei]:
> The blocking thread is running `ZonemgrRefresh`. On receiving a
> NOTIFY, this thread is eventually invoked and tries to send the
> refresh command to xfrin:
> {{{#!python
> def _send_command(self, module_name, command_name, params):
> """Send command between modules."""
> try:
> self._mccs.rpc_call(command_name, module_name,
params=params)
> }}}
:-(. The threads there and use of the CC is such a mess (and yes, I admit,
I probably caused this bug, by unifying the used mccs over the threads, it
might have used separate connection to msgq before, but that felt wrong ‒
why would one program connect to msgq multiple times?).
I think there are two things that should be done:
* Clarify if accessing CC from multiple threads is allowed, how it is
allowed and if receiving a message should work after we enter the command
main loop.
* Get rid of threads in the zone manager. There does not seem to be need
for them, except to cause bugs and brain damage (we don't really need
performance in zone manager).
But the question now is, both these probably will take some non-trivial
time (though the first one could be reasonably short-time goal). Do we
want to do some work around, like re-introducing separate CC session?
--
Ticket URL: <http://bind10.isc.org/ticket/2790#comment:11>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list