wiki:SprintPlanning20120529

pro tip: set your font to monospace for the ticket lists (the gear-like thingy on the top right)

Sprint Planning for sprint ending 2012-06-12


Date: 2012-05-29

Attendees


  • Jelte
  • Jeremy
  • Stephen
  • Shane
  • Aharen
  • Larissa
  • Michal
  • Kambe
  • Fujiwara
  • Mukund
  • Jinmei

Past sprint


Number of tickets closed: 11 Estimated count of closed: 50 Number of tickets left new: 8 Estimated count of new: 51 Number of tickets assigned/in review: 9 Estimated count of active: 48

Current total open defects: 291 (last sprint: 289)

Past Sprint Review


We set out last sprint knowing that we most probably put in too much, so this is a good time to reflect on that.

There were also a few holidays, and two of the tickets I (jelte) did were kinda underestimated IMO (by me as much as by anyone else)

Jelte: Estimate new work higher? Jinmei: I expected this type of thing; how much we would need for it? I was generally not so optimistic that we could complete this work in 1 sprint. Should be less optimistic about these kinds of things (new things). Jelte: Account by estimating, or by planning. (For example 150% of time we think it will take.)

Jinmei: On this particular sprint we had many holidays, and we did not so seriously try to prioritize the DDNS work so we had relatively heavy discussions for design, people worked on non-DDNS tickets, and so on. This is different from the in-memory NSEC case.

Assigned but idle tickets: #1828 - Python warnings; Mukund had no time to work on that this sprint (just needs fewer changes) #1890 - Jeremy can re-reply, initial comments still stand; got buried, will talk with Mukund

Goals:

  • DDNS lots of work done, but not finished
  • finish discussion on meta/containers the discussion seems to have gotten to some form of end
  • Redesign statistics model;

#1972 (mostly) no work done on this

  • something (see mail to dev) with configuration differences jelte got sidetracked by ddns work :/

General review:

  • Maybe not so good at handling defects this time
  • Some are idle, and discussed, others modified quite recently

Next Sprint


Proposed goals:

  • Finish DDNS
  • Finally migrate to new datasources

(- Implement NSID)

  • ?

*Check timeline*

LIST OF TICKETS WE DECIDED ON GOES HERE<<<

(remove 1972, and remove 1851, but move all other existing ones on)

Tickets for goals:

51 points from unassigned tickets in current sprint

DDNS:

1461 Implement DDNS system tests task DDNS 7 1735 ddns_socket file left in build directory defect Unclassified 3 1978 communication to b10-auth from b10-ddns task DDNS 5 2003 support DDNS/TCP response task DDNS 5 2005 DDNS documentation task DDNS 6 1986 b10-auth shouldn't try to forward update requests without ddns running defect b10-auth 5

(perhaps skip 1986 for now, and first discuss how to actually do it)

(31) (26 without 1986)

(+tickets from current sprint, 14)

Datasource conversion work:

(some tickets moved on from current sprint) 1997 Convert static data source to new data source enhancement Unclassified 5

(running total: 82, note that this time i'm including all tickets we move on)

Defects: 2009? (4, recent build error, missed until just now because it was hidden by another one) 1982? (4) 2004? (4)

94

lower priority to: 2003?

END OF LIST OF TICKETS WE DECIDED ON<<<

NSID:

1994 "RFC 5001 (NSID) support, configuration" enhancement Unclassified 4 1995 "RFC 5001 (NSID) support, protocol support" enhancement Unclassified 5 1996 "RFC 5001 (NSID) support, ACL" enhancement Unclassified 5 2000 update libdns++ OPT RDATA so it can handle EDNS options enhancement libdns++ 4 2001 extend libdns++ EDNS class for EDNS options enhancement libdns++ 4 2002 python wrapper for EDNS option extensions enhancement libdns++ 4 2006 create a meta ticket for NSID defect Unclassified meta

(we could conceivably leave out acls, making the total 21 + meta/cleanup) (also even 2002, if we focus on authoritative server, making the total 17 + meta/cleanup)

(26)

TICKETS FROM QUEUES

New: 1991 "boost/numeric/conversion/detail/meta.hpp:30: warning: comparison between 'e defect Unclassified 0 1998 move Formatter classes from DDNS to util task Unclassified 5 1999 detection of googletest defect Unclassified 0 2008 remove equal additions and deletions from diff.py task Unclassified 0

Next-sprint-proposed:

1973 reimplement the current set of statistics counters with new design task statistics 6 1980 Make a way to populate /AUTHORS and the authors.bind code from a single list enhancement build system 3 1981 Investigate ForwardTest? failure (occured just once) defect Unclassified 7 1982 sqlite database upgrade during make install defect build system 4 1983 asiolink::IOMessage should contain the local endpoint defect b10-auth 5 1985 eliminate duplicate hardcoding for ddns UNIX socket file both in auth and dd defect b10-auth 7 1987 'wait for' command in bindctl execute handling (or something similar) task bind-ctl 9 1990 customize EqualityFunc? of python tests enhancement Unclassified 7 1992 "support ""also-notify""" enhancement xfrout 7 1993 simple style checker task Unclassified 6

2004 Message_makeResponse (libdns++ python wrapper) should catch exceptions defect libdns++ 4 2007 redesign in-memory data source structure task data source 11

Jinmei: 1997 is easiest as built-in SQLite database Michal: I think having a master file and loading to memory is slightly better, but that's a detail...

Over at 16:01 UTC.

Last modified 5 years ago Last modified on May 29, 2012, 4:44:02 PM