BIND 10 #1298: don't do xfrin upon notify when address is unknown

BIND 10 Development do-not-reply at isc.org
Tue Nov 8 13:21:53 UTC 2011


#1298: don't do xfrin upon notify when address is unknown
-------------------------------------+-------------------------------------
                   Reporter:  jelte  |                 Owner:  jinmei
                       Type:         |                Status:  reviewing
  defect                             |             Milestone:
                   Priority:  major  |  Sprint-20111108
                  Component:         |            Resolution:
  Unclassified                       |             Sensitive:  0
                   Keywords:         |           Sub-Project:  DNS
            Defect Severity:  N/A    |  Estimated Difficulty:  0
Feature Depending on Ticket:         |           Total Hours:  0
        Add Hours to Ticket:  0      |
                  Internal?:  0      |
-------------------------------------+-------------------------------------
Changes (by jelte):

 * owner:  jelte => jinmei


Comment:

 Replying to [comment:12 jinmei]:
 > Replying to [comment:11 jelte]:
 >
 > - I've made a couple of fixes and pushed them.  Both minor, but the
 >   first one was necessary to compile with clang.

 ok, thanks

 > - format_addrinfo: it doesn't check invalid combination of family
 >   address like this: `format_addrinfo((socket.AF_INET, None, ("::1",
 54)))`
 >   (not sure if we want to catch them in this function though)

 not sure either. It would feel a bit too much to me (what would we do?
 raise an error? if so, why? Would only be something more the caller
 would have to take into account. And if we add code that recognizes
 valid addresses for families, we wouldn't even need the address family
 in the argument in the first place (except then perhaps for said
 checking). So for now let's keep it bogus-in-bogus-out. Perhaps we can
 reconsider in #1379.

 > - test_format_addrinfo: mostly a matter of taste in this context, but
 >   I'd use 2001:db8::XX for testing instead of ::1 or ::2 (especially
 >   the latter)

 oh sure, updated

 > - test_format_addrinfo: I think it's better to use more typical value
 >   for addr[1] (i.e. 'sockettype'), so instead of
 > {{{
 >         self.assertEqual("192.0.2.1:53",
 >                          format_addrinfo((socket.AF_INET, None,
 >                                           ("192.0.2.1", 53))))
 > }}}
 >   do this:
 > {{{
 >         self.assertEqual("192.0.2.1:53",
 >                          format_addrinfo((socket.AF_INET,
 socket.SOCK_STREAM,
 >                                           ("192.0.2.1", 53))))
 > }}}
 >  (it may still make sense to use a bogus value, too, to see if it's
 >   really ignored)
 >

 ok, done

-- 
Ticket URL: <http://bind10.isc.org/ticket/1298#comment:14>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development


More information about the bind10-tickets mailing list