BIND 10 #1288: Update XFROUT to use new interface

[BIND 10 Development]

#1288: Update XFROUT to use new interface
-------------------------------------+-------------------------------------
                   Reporter:         |                 Owner:  jinmei
  jinmei                             |                Status:  reviewing
                       Type:  task   |             Milestone:
                   Priority:  major  |  Sprint-20111122
                  Component:         |            Resolution:
  xfrout                             |             Sensitive:  0
                   Keywords:         |           Sub-Project:  DNS
            Defect Severity:  N/A    |  Estimated Difficulty:  5
Feature Depending on Ticket:         |           Total Hours:  0
        Add Hours to Ticket:  0      |
                  Internal?:  0      |
-------------------------------------+-------------------------------------
Changes (by vorner):

 * owner:  vorner => jinmei


Comment:

 Hello

 Replying to [comment:21 jinmei]:
 > > I'd like to see it resolved before the release, otherwise the user
 will have to reconfigure after we remove the special components. I don't
 care if in a separate ticket or not, though. How large do you guess it
 will be?
 >
 > It won't be too large, but that will be a system wide change and not
 > really specific to xfrout (xfrin will also need to be updated), so I
 > think it's a topic of a separate ticket.  I don't see the point of
 > "will have to reconfigure" part, too, in the context of whether to
 > merge this ticket; first off, if the cleanup requires the user to do
 > something, we already have this problem due to xfrin, so that means
 > whether or now we merge this branch the issue would have already
 > existed.  second, I really don't understand what the user "will have
 > to reconfigure".  Right now, the special hack for xrin/xfrout are
 > hardcoded in the bob.spec and the source code, so, basically as long
 > as the user re-install everything and not heavily customize the
 > component settings such as specifying a special priority for
 > xfrin/xfrout (which I suspect is quite unlikely at least for a while),
 > the change should be transparent.

 I didn't really mean to suggest we should hold of merging this branch
 because of it. I was more interested in if we could do the cleanup before
 the release as well, which would solve the problem with xfrin as well
 (which users didn't have a chance to see yet as well).

 And with the transparentness, it's not completely true. The configuration
 keeps the whole Boss/components section if anything in it was changed. So
 if user changes anything regarding configuration of which processes start
 or how, he will have to change the configuration manually to remove the
 special part in the component. Anyway, it probably isn't so much of a
 problem yet.

 > > Is it possible the Makefile in xfrout test needs that too? Because of
 this failure?
 >
 > Ah, yes, probably.  I updated Makefile.am for xfrout/tests, too.

 All tests now pass, probably all needed workarounds are there already.

 > > Anyway, this error might show that it doesn't send even SERVFAIL. Is
 it possible?
 >
 > That's what the original code does, but it would make more sense to
 > return SERVFAIL in such a case.  I've updated the source to do it.
 > (I slightly hate to add more special cases in an ad hoc manner like
 > this, but as we repeatedly discussed we'll need heavy overhaul of xfr*
 > eventually, and I hope we can make it cleaner at that point)

 Thanks.

 So, I believe this can be merged now.

-- 
Ticket URL: <http://bind10.isc.org/ticket/1288#comment:23>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development