BIND 10 #604: Evaluate TAHI suite for testing b10-auth

[BIND 10 Development]

#604: Evaluate TAHI suite for testing b10-auth
-------------------------------------+-------------------------------------
                 Reporter:  stephen  |                Owner:  UnAssigned
                     Type:  task     |               Status:  reviewing
                 Priority:  major    |            Milestone:  A-Team-
                Component:           |  Sprint-20110309
  b10-auth                           |           Resolution:
                 Keywords:           |            Sensitive:  0
Estimated Number of Hours:  5.0      |  Add Hours to Ticket:  0
                Billable?:  1        |          Total Hours:  0
                Internal?:  0        |
-------------------------------------+-------------------------------------
Changes (by vorner):

 * owner:  vorner => UnAssigned
 * status:  accepted => reviewing


Comment:

 Hello

 I have been reading trough the web pages and READMEs and such. I think
 that the
 testsuite is of any use for us in the vanilla form. I'm not sure if I
 understand correctly, but from what I read it seams they test network
 routing,
 network stacks and stuff like that in conditions close to real life. With
 DNS,
 they have they own server to send packets against. They describe how to
 make
 the server reconfigure itself for various tests automatically, but there's
 no
 mention of plugging in any other implementation.

 Furthermore, the tests are divided to groups, each group needs a different
 network topology, each test needs a different master file loaded, etc.
 This
 seems like a quite big task to set up (provided that they require a
 specific OS
 version ‒ some freeBSD ‒ for the test master machine). Their tests don't
 run on
 a single machine, they need a lab (which, presumably, could be bunch of
 virtual
 machines).

 So, I don't think we have enough time to use it before the end of Y2.
 However,
 as it contains bunch of various scenarios, it might be of some use in
 future.

 The scenarios are perl sources describing what should be set up and how
 and
 what packets to send and what to expect. I think we can either spend some
 time
 and write the remote end to configure our BIND10 instead of their testing
 mini-server, or (which I believe would be less work, because it wouldn't
 need
 installing several machines, some with patched kernel to inject packets)
 use
 only the descriptions and let a perl DNS library create the packets and
 send
 them, write something to create the master files and load them and make it
 run
 locally ‒ but use their ideas what could be tested.

 So, unless we have someone who wants to spend few days installing freeBSD
 machines and configuring several network adapters on each of them, I
 propose to
 leave this one for now and look into it some time later.

-- 
Ticket URL: <https://bind10.isc.org/ticket/604#comment:4>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development