BIND 10 #249: review: auth server can return SERVFAIL with an answer

[BIND 10 Development]

#249: review: auth server can return SERVFAIL with an answer
-------------------------+--------------------------------------------------
 Reporter:  jinmei       |        Owner:  jinmei                                        
     Type:  defect       |       Status:  reviewing                                     
 Priority:  major        |    Milestone:  05. 3rd Incremental Release: Serious Secondary
Component:  data source  |   Resolution:                                                
 Keywords:               |    Sensitive:  0                                             
-------------------------+--------------------------------------------------

Comment(by jinmei):

 Replying to [comment:8 each]:
 > Suggestion for a way to combine both sorts of servfail handling:

 Hmm...I'm not sure if this is the way to go (right now).  We've fixed
 known problems with try-catch + exception, so the explicit check for
 rcode==SERVFAIL is redundant.  For other cases, we are not even sure if
 there's really a scenario that could cause it.

 Although it might be embarrased if this really happened, that's not a type
 of critical bug like a buffer overrun or cache poisoning.  It may not even
 be a protocol violation (RFC1035 doen't seem to say the answer section of
 a SERVFAIL response must be empty, etc).  So, adding an explicit case for
 rcode=SERVFAIL seems to be a kind of premature optimization, which may
 remain in the code for unknown reasons.  I'd be rather make a separate
 ticket where we analyze the other cases of SERVFAIL to determie whether we
 can safely handle all of them as a really exceptional case and if not what
 is the best way to handle them.

 After that analysis, the hybrid approach may turn out to be the best
 solution.

-- 
Ticket URL: <http://bind10.isc.org/ticket/249#comment:9>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development