BIND 10 #1924: Msgq undeliverable notifications

[BIND 10 Development]

#1924: Msgq undeliverable notifications
-------------------------------------+-------------------------------------
            Reporter:  vorner        |                        Owner:
                Type:  task          |  jinmei
            Priority:  medium        |                       Status:
           Component:  msgq          |  reviewing
            Keywords:                |                    Milestone:
           Sensitive:  0             |  Sprint-20130219
         Sub-Project:  Core          |                   Resolution:
Estimated Difficulty:  6             |                 CVSS Scoring:
         Total Hours:  0             |              Defect Severity:  N/A
                                     |  Feature Depending on Ticket:  msgq-
                                     |  ng
                                     |          Add Hours to Ticket:  0
                                     |                    Internal?:  0
-------------------------------------+-------------------------------------

Comment (by jinmei):

 Replying to [comment:23 vorner]:

 > > That said...I expect (or hope) we'll clarify these points and revisit
 > > the msgq system at a more fundamental way, and, with that assumption,
 > > I wouldn't argue on this point in the context of this ticket.
 >
 > I hope we get to it soon too. But I'm not sure at which opportunity. Do
 you think it makes sense to create some kind of ticket for it and try to
 push it to a sprint?

 I think creating a ticket makes sense.  I don't have a strong opinion
 about when to work on it; I'm not insisting that should be considered
 an immediate followup task of this one, but I'd rather see it as a bit
 longer term task.

 > However, I discovered a problem. The python file is generated into
 builddir in distcheck, which is different than the srcdir. And python
 insists all the files are in the same directory where the `__init__.py`
 file lives. Would you have an idea how to solve this in an elegant way
 (eg. not by copying all the files into the builddir, for example).

 Is this about that we have python/isc/cc/proto_defs.py under builddir
 while other .py files under srcdir, and distcheck fails in unittests
 when importing isc.cc.proto_defs?

 (I've not tried it myself but) if so...hmm, I'm not sure if it's
 "elegant", but I guess we can use the same trick (or kludge) as that
 used for log_messages:

 - place a "forwarder" module of proto_defs.py under srcdir/.../isc/cc
   that simply imports "work.proto_defs".
 - the real proto_defs.py should go to builddir/.../isc/cc/work/
 - on install, we make sure to install work/proto_defs.py under isc/cc,
   skipping the forwarder proto_defs.py.

 And, on to the latest branch:

 '''util/python/Makefile.am'''

 - It's awkward that the new .py's need to be in EXTRA_DIST while
   gen_wiredata.py doesn't have to (I've read the commit log, btw).
   The latter is located under top_builddir:
 {{{
 .spec.wire:
         $(PYTHON) $(top_builddir)/src/lib/util/python/gen_wiredata.py -o
 $@ $<
 }}}
   i.e., not under top_srcdir.  Could that be related?

-- 
Ticket URL: <http://bind10.isc.org/ticket/1924#comment:24>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development