Sprint planning meeting for sprint 2011-12-06

Date: 2011-11-22



Past Sprint Summary

Closed tickets: 26
Est. of closed tickets: 93
(still using estimated hours, the total fields are too erratic at this point imo)
Tickets left unassigned: 4 (of which 1 is a meta-ticket which has 1 ticket left, and 1 has status new but does seem assigned, but I do not know how much, if any, has been done on this)

We will wait at least one more sprint before we try to use the hours recorded by developers. Burndown page will still be based on estimates.

Past sprint review


  • User stories & requirements for DDNS
  • Design for DDNS (high/low?)
  • Finish data source
  • Boss flexibility
  • IXFR-out

IXFR-out - "serving part" finished - REACHED

Michal: system tests for IXFR-out needed

Boss flexibility - REACHED

Michal will be giving demo tomorrow

Still a few TODO tickets left

Finish data source - NOT REACHED

Some tickets left

Some initial planning for NSEC3

DDNS design - NO WORK

Discussion - we probably can only do this for unsigned zones

DDNS user stories - NO WORK

Plan for this sprint.

Jelte: Comments on last sprint?

Jinmei: Some tickets have been assigned for a long period. What has happened for these tickets?


#1310: Stephen reports he has been busy, but has been progress

#510: Aharen reports no time, but will work in next sprint.

#509: closing NOW

#1386: Dima says no problems, in distcheck now

Next sprint


(proposed in this order of importance)

  • DDNS use-cases and design (still/again)

Larissa: Have results from survey of about 25 people who care a lot about DDNS; extracting out DDNS requests. Also sending e-mail and follow-up phone calls for people who say DDNS is critical for them. Input and review appreciated, there is a wiki page:

This will not stop design.

Jinmei: Seems vague. Should we discuss on list?

Jelte: We could make the ticket a task of "starting & summarizing discussion".

Jinmei: Makes sense to have discussion rather than assigning it to someone specific.

Jinmei: I also think we should do this for user stories.

Larissa: I will take a ticket for this.

  • Finishing and using the socket creator
  • Finishing NSEC3 support in new data source

Shane: This is for the refactored data source, not the red-black tree?

Jinmei: Yes, although adding it to the red-black tree should be straightforward, since it is a port from BIND 9.

  • statistics?

Jinmei: I experimentally suggest we make user visible features, and suggest adding a lot more counters, and tickets based on that suggestion.

Jinmei: We should also more carefully keep track of the status. Maybe in the next sprint planning if there seems to be issues in the progress we should think about how to recover from that.

open discussion on proposed tickets

#1395 python isc.datasrc should be usable without logging

Michal: We had a policy to always throw, if logging was not initalized. This ticket goes against that idea.

Discussion to be done on list, and ticket will be deferred.

#1398 refactor _reply_xfrout_query

Also defered for discussion.

Ticket from current sprint

Tickets for goals


#1347 design and define work needed for DDNS support                   7

(what else?)

-> becomes start/stopping discussion

+ new ticket for writing it down

+ use case ticket (for Larissa)

Finish Socket Creator

#802    Server-side code on boss to handle requests for new sockets    6
#803    Command to list currently used sockets                         4
#804    Client side code to request new sockets (Python)               5
#805    Client side code to request new sockets (C++)                  5

(what else?)

+ actually using it

Jinmei: would not be surprised if we need more work, or if some of the tasks are too big. My suggestion is to start with the things we currently have and see what happens.

Jeremy: My concern is making a stand-alone ticket would be difficult with this design.

Michal: Actually if we want to use the socket creator for both goals, we need to use it for getting all the ports.

Jelte: Need discussion, but we should do these tasks now.

Shane: Do we have an API? Michal: Was a discussion, results should be on the wiki.

NSEC3 support in new data source

Is #1178 about done and were any tickets created from this?

Stephen: We have detailed requirements:

Worried about tasks being linked and cannot be done simultaneously.

Jinmei: My suggestion is to start tasks on this sprint, and based on experience we can discuss how to make more progress (in parallel, or perhaps reorganizing the tasks) on the next sprint.

Stephen: I can breakdown tasks and put them into the sprint, then we can do an estimation just for NSEC3 tasks, we can see which ones we get into the sprint, and maybe cut some out.

Tickets from proposed

#1278  3
#1299  3
#1407  2

#1314  7
#1315  3
#1317  6

#1387  5
#1388  3 (depends on #1292)
#1389  2

#1358  2

Shane: 28 points... Jelte: Don't know how many points we hvae now! Jinmei: Suggest we reserve 10 points for NSEC3

DDNS:           11
Socket Creator: 20
NSEC3:          15
tests+bugs:     28

Statistics tickets:


Too much tickets, so we should work on "base" NSEC3 stuff this sprint (for example closest enclosure).

Any other tickets

Release related discussion

I do not plan to do a release-review or anything of the kind, I just want to reserve some time for last-minute plans surrounding the release.

Jeremy: Finishing up documentation.

Jeremy: Mac OSX server has core dumps when auth runs. Have 30 of them!

Stephen: Details of how to run branches on build bots would be really useful. Jeremy: Okay.

Jelte: Plan to branch tomorrow and release Monday? Jeremy: Yes.

Jeremy: Will push versioning for libraries via libtool, will fix one of our build machines.


Jinmei: What's the status of Likun?

Kevin: Likun invited 3 new staff to join BIND 10 team. These 3 will join the team to do more work on BIND 10.

Michal: Tomorrow is the demo. Will happen after "stand-up" call.

Last modified 7 years ago Last modified on Nov 22, 2011, 6:02:52 PM