last testday | index | next testday »

Bug Day September 24th, 2009

How to get help, how to get started, and schedule for Bugdays.

Bug days are conducted to reduce the number of UNCOnfirmed bugs in the bugzilla database, and to improve the quality of the bugs in the database.

Focus

Help shape Thunderbird 3. And help trunk testers who help us by reporting bugs.

This week we'll have two focus :

Regression window wanted

Regressions are defined as something that used to work but doesn't work anymore. It is often valuable to developers to know when the regression happened. They then know who broke the code, why it broke and what they should be careful about when fixing it again. Finding the regression windows means finding the last version of Thunderbird on which the issue was not present and finding the first one in witch it was present. Gary has written a nice tutorial on how to search for regression windows, it explain all what you need to do to participate.

So to help here you'll need :

• a bugzilla account to leave comments on the bugs
• some time to download and test the builds (see gary's Howto above)
• a List of bugs for which the regression-window is wanted .
• Join our chat to tell us which bug you are working on, to avoid two people to work on the same bug at the same time.

QA Wanted

QA wanted is kind of the same idea - there is a bug - but we don't know how to reproduce it. Reproducing the bug is the first step in fixing it. Sometimes figuring :

1. do this
2. then do that
3. then this happens
4. and now click there
5. see the bug

can be tedious - but this is needed in order to be able to fix bugs. So we use QAWANTED bugzilla keyword when we feel there is a bug, but more time needs to be spent investigating it and trying to figure out how the bug shows up.

To help here you'll need :

• a bugzilla account to comment on the bug.
• lists of bugs to work on :
  1. Thunderbird-only bugs that have the qawanted keyword
  2. Seamonkey+Thunderbird bugs that have the qawanted keyword
• Join our chat to tell us which bug you are working on, to avoid two people to work on the same bug at the same time.
• Telling us how you could reproduce, using some Steps To Reproduce (we call these STR), is what we seek.
• But Telling us that you couldn't reproduce the bug is very helpful too (well that is if you've spent more than 1/2 hour trying).

Tips

You can ask help during the event here. Please be patient sometimes we are connected but not in front of our computers.

Before starting, see the Bugdays page for information about what to do and how to get help. Don't be concerned if you don't understand everything, just ask for help.

Searching in bugzilla can be tricky - here are a few tricks that might help.

Results

For this bugday, the times of the sessions, accounting for 30mins [1] of grace period before and after each session, were as follows:

• Session 1
  • Start time: 2009-09-24 0530h PDT
  • End time: 2009-09-24 0830h PDT

• Session 2
  • Start time: 2009-09-24 1130h PDT
  • End time: 2009-09-24 1430h PDT

• Session 3
  • Start time: 2009-09-24 1830h PDT
  • End time: 2009-09-24 2230h PDT

• 17 hour duration
  • All sessions & intermediate rest time.
  • Start time: 2009-09-24 0530h PDT
  • End time: 2009-09-24 2230h PDT

Many thanks to volunteers and developers (total 5 - 10) who turned up for this bug day, and others that I may have possibly missed.

Notes:

• Spreadsheet of bugday results is compiled and updated after each session / bugday.
• The letter Q stands for the query with all the appropriate criteria, which over time may differ from the results that have been logged.
• [1] Session and period time includes 30mins before and after the official time period, so as to include early and late/spill over activity.
• [2] A resolution change may be non-terminal, for example a bug may have been reopened.