QualityAssurance
In this section, we cover the activities of the QA team[1].
Contributing Writer: Adam Williamson
Test Days
There was no Test Day last week.
No Test Day is scheduled on the main track next week. However, the new Fit and Finish[1] Test Day track will be holding its second event[2], on power management and suspend/resume. The Test Day page already includes several test scenarios, and a live CD for testing will soon be available. The Fit and Finish project is a great effort to improve the details of the Fedora project, so please show up to support this event! The Test Day will be held on 2009-07-21 (Tuesday) in IRC #fedora-fit-and-finish (note this is not the same channel where main track Test Days take place).
If you would like to propose a main track Test Day for the Fedora 12 cycle, please contact the QA team via email or IRC, or file a ticket in QA Trac[3].
Weekly meetings
The QA group weekly meeting[1] was held on 2009-07-15. The full log is available[2]. James Laska reported that he had filed tickets to track the creation of the three Debugging pages identified as desirable by Christopher Beland, and would mail the list to try and attract volunteers to work on the pages.
James also noted he is still working on the Goals page[3], using a personal space draft[4], but was not yet ready to go into production with it.
James and Jesse Keating reported on the revisions to the Fedora 12 schedule in terms of QA and release engineering. The latest revised schedules are available: QA[5] and release engineering[6].
James reminded the group about the then-forthcoming Alpha Blocker Bug Day, which would be held on 2009-07-17. Adam Williamson suggested reviewing F12Blocker bugs (which block only the final release) to see if they should be promoted to blocking the Alpha release also. James brought up the question of the criteria for Alpha blocker bugs. After some discussion, there was general agreement to work on the basis of considering only high-severity bugs in critical path components (as defined by the Critical Path Packages Proposal[7]) as Alpha blockers.
James mentioned that the Fedora 12 Test Day schedule is still currently lightly populated, but he and Adam have several events planned which have not yet been set down to specific dates.
Will Woods reported on the progress of the AutoQA project. He has now completed writing the test cases for the Rawhide Acceptance Test Plan[8]. He is now starting to work on writing automated tests for these cases, using autotest. He pointed out that progress information can also be found in AutoQA trac[9].
Finally, the group discussed changing the meeting day and/or time. Adam suggested creating a matrix of possible times and having each interested member fill out the times at which they are available, as has been done by other groups in the past. James offered to create the matrix and notify the mailing list so that people could fill it in once it was ready.
Jóhann Guðmundsson pointed out that the QA group could potentially be affected by the ongoing question about the use of Fedora trademarks in non-official spins, as it frequently generates non-official spins for use in Test Days. The group agreed to monitor this on an ongoing basis.
The Bugzappers group weekly meeting[10] was held on 2009-07-14. The full log is available[11]. Richard June apologized for not having asked Brennan Ashton for an update on the triage metrics project.
The group reviewed Niels Haase's proposed expanded list of priority triage components[12]. Edward Kirk thought that some of the components were not truly critical. The list was tabled for review when Niels could be present at a meeting.
The group discussed the latest version of Matej Cepl's greasemonkey script. It seems to have been deployed by several triagers with no problems so far.
Other topics were tabled due to the absence of several group members for various reasons.
The next QA weekly meeting will be held on 2009-07-22 at 1600 UTC in #fedora-meeting, and the next Bugzappers weekly meeting on 2009-07-21 at 1500 UTC in #fedora-meeting.
- ↑ http://fedoraproject.org/wiki/QA/Meetings
- ↑ http://fedoraproject.org/wiki/QA/Meetings/20090715
- ↑ http://fedoraproject.org/wiki/QA/Goals
- ↑ http://fedoraproject.org/wiki/User:Jlaska/Draft
- ↑ http://poelstra.fedorapeople.org/schedules/f-12/f-12-quality-tasks.html
- ↑ http://poelstra.fedorapeople.org/schedules/f-12/f-12-releng-tasks.html
- ↑ http://fedoraproject.org/wiki/Critical_Path_Packages_Proposal
- ↑ http://fedoraproject.org/wiki/QA:Rawhide_Acceptance_Test_Plan
- ↑ https://fedorahosted.org/autoqa/milestone/israwhidebroken.com
- ↑ http://fedoraproject.org/wiki/BugZappers/Meetings
- ↑ http://meetbot.fedoraproject.org/fedora-meeting/2009-07-14/fedora-meeting.2009-07-14-15.05.log.html
- ↑ http://fedoraproject.org/wiki/User:Arxs/CPCL
F12 Alpha blocker bug review meeting
John Poelstra announced[1] the first blocker bug review meeting for Fedora 12, to be held on 2009-07-17, mainly to review blocker bug status for the upcoming Alpha release. Later, Adam Williamson posted a recap of the meeting[2], recording that it had been well attended and had been able to review the whole F12 Alpha and main blocker lists, remove some from the lists, promote some to block the Alpha release, and check on the development status of several bugs.
Updated list of components for priority triage
Niels Haase announced[1] that he had updated his proposed expansion of the list of priority components for the Bugzappers group to focus on triaging[2], based on the Critical Path Packages Proposal, as previously approved at Bugzappers meetings.
Xfce spin testing
Adam Miller announced[1] that, he would be building a test live image with the Xfce desktop roughly each week, and would appreciate testing and reporting of problems. He also included a link to the first build.
What to do with Fedora 11 target bugs
John Poelstra pointed out[1] that the F11Target bug[2] was still open (and depending on 321 bugs), and asked what people thought should be done about it. Niels Haase suggested[3] moving all that had been triaged to F12Target. Mark McLoughlin suggested[4] having F11Target block F12Target, effectively moving the bugs to F12Target wholesale. Matthias Clasen opined that "I don't think it makes sense to accumulate hundreds of bugs on the target tracker, if they only end up getting pushed from release to release"[5]. Adam Williamson agreed, and suggested[6] just closing the tracker bug, as had been done for Fedora 9 and Fedora 10. No final decision was yet reached.
- ↑ http://www.redhat.com/archives/fedora-test-list/2009-July/msg00259.html
- ↑ http://bugzilla.redhat.com/show_bug.cgi?id=446451
- ↑ http://www.redhat.com/archives/fedora-test-list/2009-July/msg00260.html
- ↑ http://www.redhat.com/archives/fedora-test-list/2009-July/msg00265.html
- ↑ https://www.redhat.com/archives/fedora-test-list/2009-July/msg00267.html
- ↑ http://www.redhat.com/archives/fedora-test-list/2009-July/msg00270.html
QA meeting time/date adjustment
As discussed at the weekly meeting, James Laska announced[1] that he had created a matrix to track possible new times and days for the QA group weekly meeting, and asked everyone interested in attending the meetings to fill out the matrix with the days and times when they are available.
Anaconda triage project progress
Adam Williamson reported[1] on the progress of the ongoing project to integrate anaconda triage into the Bugzappers group and workflow. He thanked Andy Lindeberg for her efforts in joining the mailing list, weekly meetings and IRC channel, and in working to codify the current workflow used to triage anaconda bugs. He recorded that meeting and email discussions had revealed little in the way of fundamental conflicts between the official Bugzappers workflow[2] and the Anaconda workflow[3]. He had therefore modified the components and triagers page[4] to list the Anaconda workflow page as the special instructions for triaging anaconda, and note that additional triagers are now welcome for anaconda if someone has a burning desire to work on it, although Andy is currently covering the area very effectively.
Bugzilla semantics debate
Adam Williamson asked the list[1] about a question he had been discussing with Andy Lindeberg, regarding the semantics of the NEW and ASSIGNED states in Bugzilla. He proposed the use of a keyword (instead of the ASSIGNED state) to indicate a bug has been triaged, and either removing the ASSIGNED state entirely, or noting in the workflow page that it has no real function and is effectively equivalent to NEW. This led to an enthusiastic debate, with many other proposals made, although all seemed to agree that the current state of ASSIGNED meaning that a bug has been triaged is not optimal. No final consensus was yet reached on what changes, if any, to propose to the configuration of Bugzilla and/or the official workflow.