From Fedora Project Wiki

< FWN‎ | Beats

(create 196 draft)
(create 197 qa beat)
Line 10: Line 10:
=== Test Days ===
=== Test Days ===


Last week's Test Day<ref>http://fedoraproject.org/wiki/Test_Day:2009-10-01_Anaconda/Features/StorageFiltering</ref> was on the installer's (Anaconda) storage system. Unfortunately not a huge number of testers were present for this test day, but those who did come managed to test a range of scenarios and file several important bug reports. Thanks to all testers.
Last week's Test Day<ref>http://fedoraproject.org/wiki/Test_Day:2009-10-08_RAID</ref> was on the the use of software RAID arrays with Anaconda. Our hardened group of install testers worked hard to test as many RAID scenarios as possible. We'd like to send a big thanks to [[User:Atorkhov|Alexey Torkhov (greenlion)]] for contributing so much valuable testing.


Next week's Test Day<ref>http://fedoraproject.org/wiki/Test_Day:2009-10-08</ref> on 2009-10-08 will be specifically on the use of software RAID arrays with Anaconda. As always, the Test Day will run all day in the #fedora-test-day IRC channel. This is a nice tightly focused topic with clear test cases, and an important feature for many users, so please come out to the Test Day and help us ensure softward RAID is tested on a wide range of storage hardware and configurations.
Next week's Test Day<ref>https://fedoraproject.org/wiki/Test_Day:2009-10-15</ref> will be on internationalization (also known as i18n)<ref>http://fedoraproject.org/wiki/I18N</ref> - an event which usually has a strong focus on input methods, but can also cover issues like fonts. The Test Day will run all day on Thursday 2009-10-15 in the #fedora-test-day IRC channel. Please come along and help ensure Fedora works just as well no matter what language you use!


No Fit and Finish track Test Day is planned for next week.
No Fit and Finish track Test Day is planned for next week.
Line 22: Line 22:
=== Weekly meetings ===
=== Weekly meetings ===


The QA group weekly meeting<ref>http://fedoraproject.org/wiki/QA/Meetings</ref> was held on 2009-09-28. The full log is available<ref>http://meetbot.fedoraproject.org/fedora-meeting/2009-09-28/fedora-meeting.2009-09-28-16.00.log.html</ref>. [[User:Adamwill|Adam Williamson]] noted that follow-up on the development issues discussed the previous week was impossible with [[User:jkeating|Jesse Keating]] and [[User:dpravec|David Pravec]] absent, so left the topic for the following week.
The QA group weekly meeting<ref>http://fedoraproject.org/wiki/QA/Meetings</ref> was held on 2009-10-05. The full log is available<ref>http://meetbot.fedoraproject.org/fedora-meeting/2009-10-05/fedora-meeting.2009-10-05-16.00.log.html</ref>. [[User:Adamwill|Adam Williamson]] asked [[User:jkeating|Jesse Keating]] if there had been any follow-up on the development process issues discussed at the last meeting but one. Jesse explained that he had repeated to Red Hat management that it was too late to land feature changes in Fedora 12, whatever their desirability for Red Hat products.


[[User:Wwoods|Will Woods]] reported on the progress of the AutoQA project. The major achievement was a working prototype of the planned israwhidebroken.com site, of which Will provided a screenshot<ref>http://wwoods.fedorapeople.org/screenshots/irb.png</ref>. The source code for the web application which controls the page is also available<ref>http://fedorapeople.org/gitweb?p=wwoods/public_git/israwhidebroken.git;a=summary</ref>. The app allows results for manual-only tests to be provided by users authenticated via FAS. Will is planning to extend the app to provide links to logs for the automated tests, and the ability to add bug report links for failed tests. Two new tests<ref>http://fedoraproject.org/wiki/QA:Anaconda_package_install_test_case</ref> <ref>http://fedoraproject.org/wiki/QA:Anaconda_bootloader_setup_test_case</ref> had been added to cover the later stages of Anaconda installation (beyond disk partitioning).  
[[User:Wwoods|Will Woods]] reported on the progress of the AutoQA project. He had implemented a UI for the israwhidebroken.com project, allowing QA group members to submit results for the manual tests which will form a part of it. Implementing a nicer AJAX UI and automating yum and network tests can be done later, so now only backlinks from israwhidebroken.com to autotest need to be implemented before it can go public. [[User:Kparal|Kamil Paral]] mentioned that he is working on a test for comparing the old and new versions of a package each time a new build is uploaded, looking for important changes such as dependency changes, configuration file changes, and file mode changes. He hopes to integrate this test into AutoQA in future.


[[User:Adamwill|Adam Williamson]] summarized upcoming events, and [[User:Jlaska|James Laska]] pointed out the relevant calendar page<ref>http://poelstra.fedorapeople.org/schedules/f-12/f-12-quality-tasks.html</ref>. The beta freeze and beta candidate build process was imminent, and the fourth beta blocker review meeting was due Friday 2009-10-02. Denise Dumas pointed out that some important bugs required testing to confirm prospective fixes, and Adam pointed to the beta blocker bug list<ref>http://bugzilla.redhat.com/showdependencytree.cgi?id=507678&hide_resolved=1</ref> as a reference for these.
The group then looked at the status of Fedora 12 Beta, with regard to the go/no-go decision meeting which would come later that day. There was a broad consensus that the QA team's position would be against shipping the Beta on schedule, due to several blocker bugs still being open and apparently unfixed. The group spent some time discussing individual bugs in detail and agreeing action plans to address several of them.


[[User:Jlaska|James Laska]] started a discussion regarding plans for the then-upcoming Anaconda storage filtering Test Day<ref>http://fedoraproject.org/wiki/Test_Day:2009-10-01_Anaconda/Features/StorageFiltering</ref>. Denise Dumas was in favour of cancelling it as the storage filtering changes had been delayed until Fedora 13. James and [[User:Adamwill|Adam Williamson]] suggested converting it into a more general test day on Anaconda storage issues, and this path was agreed upon. James and Denise agreed to work to ensure Anaconda would be in a testable state for the Test Day.
[[User:Jlaska|James Laska]] summarized upcoming events, including the imminent Beta go/no-go meeting, and the RAID Test Day<ref>http://fedoraproject.org/wiki/Test_Day:2009-10-08_RAID</ref>.


The Bugzappers group weekly meeting<ref>http://fedoraproject.org/wiki/BugZappers/Meetings</ref> was held on 2009-09-29. The full log is available<ref>http://meetbot.fedoraproject.org/fedora-meeting/2009-09-29/fedora-meeting.2009-09-29-15.25.log.html</ref>. [[User:Tk009|Edward Kirk]] reported that he had started work on moving action items from previous meetings into Trac<ref>http://fedorahosted.org/triage/</ref>, as had been discussed the previous week, and would continue to work on it. [[User:Adamwill|Adam Williamson]] encouraged him to ask for help if he felt it was too much work to complete on his own.
The Bugzappers group weekly meeting<ref>http://fedoraproject.org/wiki/BugZappers/Meetings</ref> was held on 2009-10-06. The full log is available<ref>http://meetbot.fedoraproject.org/fedora-meeting/2009-10-06/fedora-meeting.2009-10-06-15.00.log.html</ref>. The group noted that [[User:Tk009|Edward Kirk]] was absent, so his plans to improve Triage Day events could not be discussed.


[[User:Tk009|Edward Kirk]] also provided a draft of the meeting organization SOP<ref>https://fedoraproject.org/wiki/User:Tk009/meeting-sop</ref> that he had begun work on. The group felt it was a good start. Edward wondered where the page should end up, and [[User:Adamwill|Adam Williamson]] suggested it should be a page of its own, linked from the main Meetings page<ref>http://fedoraproject.org/wiki/BugZappers/Meetings</ref>, as other SOP pages are. Edward promised to work on finalizing the page for the next meeting.
[[User:Rjune|Richard June]] gave an update on the kernel triage project. He had found some bugs where useful triage could be conducted, and would contact [[User:linville|John Linville]] to ensure the information requests he was sending out were sensible. [[User:Adamwill|Adam Williamson]] noted that the amount of time it had taken to find a few bugs where triage could be useful might suggest the kernel triage project overall was not worth pursuing, but Richard said it was still too early to tell and he would continue to investigate different types of bugs.  


[[User:Tk009|Edward Kirk]] said that he had emailed Brennan Ashton for an update on the triage metrics project, but had not received a reply. He said he was working on a plan to move the project forward given Brennan's erratic availability.
[[User:Tk009|Edward Kirk]] also provided a draft of the meeting organization SOP<ref>https://fedoraproject.org/wiki/User:Tk009/meeting-sop</ref> that he had begun work on. The group felt it was a good start. Edward wondered where the page should end up, and [[User:Adamwill|Adam Williamson]] suggested it should be a page of its own, linked from the main Meetings page<ref>http://fedoraproject.org/wiki/BugZappers/Meetings</ref>, as other SOP pages are. Edward promised to work on finalizing the page for the next meeting. Jeff Hann volunteered to help Richard with the project, and they agreed to co-ordinate after the meeting.


[[User:Tk009|Edward Kirk]] asked if it was yet time to implement the previously discussed change to triage policy, which was intended to begin when Fedora 13 became active. [[User:Adamwill|Adam Williamson]] noted that branching had occurred on the development side, but Rawhide was still tracking Fedora 12 rather than Fedora 13, and the policy change should take effect when Rawhide began tracking Fedora 13.
[[User:Adamwill|Adam Williamson]] started a discussion on [[User:Johannbg|Jóhann Guðmundsson's]] proposal to standardize the Wiki pages containing instructions on debugging problems in various Fedora components. The group generally agreed that it was a good idea, and Jóhann agreed to work on a proposed standard template for these pages, to be discussed and refined on the mailing list.


[[User:Tk009|Edward Kirk]] reminded the group that housekeeping tasks<ref>http://fedoraproject.org/wiki/BugZappers/HouseKeeping</ref> for the Fedora 13 release would be due soon. He and [[User:poelstra|John Poelstra]] already had planned to divide the tasks up between themselves.
The next QA weekly meeting will be held on 2009-10-12 at 1600 UTC in #fedora-meeting, and the next Bugzappers weekly meeting on 2009-10-13 at 1500 UTC in #fedora-meeting.
 
[[User:Rudchenkos|Sergey Rudchenko]] asked about what to do with Fedora 10 bugs as the end of support for Fedora 10 approached. [[User:Tk009|Edward Kirk]] and [[User:Adamwill|Adam Williamson]] suggested that he ask the reporters of Fedora 10 bugs to see if they could be reproduced on Fedora 11 or 12, and move the bugs to one of those releases if they could.
 
The next QA weekly meeting will be held on 2009-10-05 at 1600 UTC in #fedora-meeting, and the next Bugzappers weekly meeting on 2009-10-06 at 1500 UTC in #fedora-meeting.


<references/>
<references/>


=== Blocker bug criteria ===
=== Wiki debugging page standardization ===


[[User:atorkhov|Alexey Torkhov]] asked<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00650.html</ref> about the Wiki page on release criteria<ref>http://fedoraproject.org/wiki/QA/ReleaseCriteria</ref>, asking 'If I've found a bug that (I think) breaks MUST rule should I add it to both F12Beta and F12Blocker trackers? And if it breaks SHOULD rule, it should be added only to F12Blocker tracker?' [[User:Adamwill|Adam Williamson]] replied<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00671.html</ref> that Alexey's interpretation was correct as far as the criteria defined on that page went, but in practice bugs outside of those criteria were still considered Beta and final release blockers, as it was very hard comprehensively to codify all possible release criteria.
Following discussion in the meeting, [[User:Johannbg|Jóhann Guðmundsson]] submitted his proposal<ref>http://www.redhat.com/archives/fedora-test-list/2009-October/msg00112.html</ref> for standardizing Wiki pages which explain how to debug and file proper bug reports on Fedora components. The proposal was enthusiastically received, with additional suggestions coming from Giovanni Tirloni<ref>http://www.redhat.com/archives/fedora-test-list/2009-October/msg00113.html</ref>, [[User:Beland|Christopher Beland]]<ref>http://www.redhat.com/archives/fedora-test-list/2009-October/msg00115.html</ref>, [[User:Adamwill|Adam Williamson]]<ref>http://www.redhat.com/archives/fedora-test-list/2009-October/msg00123.html</ref> and others. A template page<ref>http://fedoraproject.org/wiki/Template:How_to_debug</ref> is now in place and being refined, and some debugging pages have already been revised to use the template. The group will convert other pages one by one.


<references/>
<references/>


=== Beta test compose Delta ISOs ===
=== Fedora 12 Beta release process ===
 
Andre Robatino reported<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00676.html</ref> that he had been unable to generate Delta ISOs from the Alpha release to the Beta test compose due to some technical problems.
 
<references/>
 
=== Xfce Test Day recap ===
 
[[User:Kevin|Kevin Fenzi]] provided a recap<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00691.html</ref> of the previous week's Xfce Test Day<ref>http://fedoraproject.org/wiki/Test_Day:2009-09-24_XFCE</ref>, thanking participants and providing a list of bugs which had been filed.
 
<references/>
 
=== Wiki Test Results name space ===
 
[[User:Jlaska|James Laska]] forwarded<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00692.html</ref> a ticket<ref>http://fedorahosted.org/fedora-infrastructure/ticket/1664</ref> about the creation of a Test_Results wiki namespace for the purpose of filing test results separately from Test Day events (the name space will allow people who have not signed the CLA to file results, as they currently are able to on Test Day pages). [[User:Adamwill|Adam Williamson]] replied<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00693.html</ref> worrying about the implied suggestion that the Test_Day namespace be removed and all Test Day pages moved to the Test_Results namespace, which he thought would be a bad change. [[User:Johannbg|Jóhann Guðmundsson]] agreed<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00705.html</ref>.
 
<references/>
 
=== Proposed removal of Anaconda from Test Day CDs ===
 
[[User:Kparal|Kamil Paral]] proposed<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00708.html</ref> the removal of Anaconda from the standard build configuration for Test Day live images, on the basis that it is rarely used as part of Test Day testing and dependency problems with the anaconda package sometimes cause problems in the generation of the images. Several replies felt the change was unnecessary, and the availability of Anaconda on the Test Day CDs probably helped get more people testing Rawhide installation. In the end it was agreed that adding instructions for removing anaconda from the build to the Test Day live CD creation instructions page<ref>http://fedoraproject.org/wiki/QA/Test_Days/Live_Image</ref> would be sufficient to address the problem.
 
<references/>


=== Installation testing results ===
[[User:Adamwill|Adam Williamson]] represented the QA group at the 2009-10-05 release engineering meeting<ref>http://meetbot.fedoraproject.org/fedora-meeting/2009-10-05/fedora-meeting.2009-10-05-18.02.log.txt</ref>. During the discussion of whether or not to release the Fedora 12 Beta according to the original schedule, Adam presented the QA group's position that the Beta release, and all subsequent events, should be delayed by a week due to the fact that all release blocker bugs had not been resolved. Other groups present agreed, and the meeting resulted in an agreement to delay the release due to this wide consensus.


[[User:Liam|Liam Li]] provided a report<ref>http://www.redhat.com/archives/fedora-test-list/2009-September/msg00721.html</ref> on testing conducted by his team of Red Hat testers on installation using the pre-Beta and Beta test compose images. He summarized the bugs they had encountered, and asked for help in completing the test cases listed on the test matrix<ref>http://fedoraproject.org/wiki/Test_Results:Fedora_12_Beta_TC_Install</ref> which the team had not been able to cover.
As a result, the group continued to work on aiding the process of resolving release blocker bugs during the week, and a final blocker bug review meeting was held on Friday 2009-10-09. [[User:Jlaska|James Laska]] provided a recap of this meeting<ref>http://www.redhat.com/archives/fedora-test-list/2009-October/msg00157.html</ref>.


<references/>
<references/>

Revision as of 02:52, 10 October 2009

QualityAssurance

In this section, we cover the activities of the QA team[1].

Contributing Writer: Adam Williamson

Test Days

Last week's Test Day[1] was on the the use of software RAID arrays with Anaconda. Our hardened group of install testers worked hard to test as many RAID scenarios as possible. We'd like to send a big thanks to Alexey Torkhov (greenlion) for contributing so much valuable testing.

Next week's Test Day[2] will be on internationalization (also known as i18n)[3] - an event which usually has a strong focus on input methods, but can also cover issues like fonts. The Test Day will run all day on Thursday 2009-10-15 in the #fedora-test-day IRC channel. Please come along and help ensure Fedora works just as well no matter what language you use!

No Fit and Finish track Test Day is planned for next week.

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[4].

Weekly meetings

The QA group weekly meeting[1] was held on 2009-10-05. The full log is available[2]. Adam Williamson asked Jesse Keating if there had been any follow-up on the development process issues discussed at the last meeting but one. Jesse explained that he had repeated to Red Hat management that it was too late to land feature changes in Fedora 12, whatever their desirability for Red Hat products.

Will Woods reported on the progress of the AutoQA project. He had implemented a UI for the israwhidebroken.com project, allowing QA group members to submit results for the manual tests which will form a part of it. Implementing a nicer AJAX UI and automating yum and network tests can be done later, so now only backlinks from israwhidebroken.com to autotest need to be implemented before it can go public. Kamil Paral mentioned that he is working on a test for comparing the old and new versions of a package each time a new build is uploaded, looking for important changes such as dependency changes, configuration file changes, and file mode changes. He hopes to integrate this test into AutoQA in future.

The group then looked at the status of Fedora 12 Beta, with regard to the go/no-go decision meeting which would come later that day. There was a broad consensus that the QA team's position would be against shipping the Beta on schedule, due to several blocker bugs still being open and apparently unfixed. The group spent some time discussing individual bugs in detail and agreeing action plans to address several of them.

James Laska summarized upcoming events, including the imminent Beta go/no-go meeting, and the RAID Test Day[3].

The Bugzappers group weekly meeting[4] was held on 2009-10-06. The full log is available[5]. The group noted that Edward Kirk was absent, so his plans to improve Triage Day events could not be discussed.

Richard June gave an update on the kernel triage project. He had found some bugs where useful triage could be conducted, and would contact John Linville to ensure the information requests he was sending out were sensible. Adam Williamson noted that the amount of time it had taken to find a few bugs where triage could be useful might suggest the kernel triage project overall was not worth pursuing, but Richard said it was still too early to tell and he would continue to investigate different types of bugs.

Edward Kirk also provided a draft of the meeting organization SOP[6] that he had begun work on. The group felt it was a good start. Edward wondered where the page should end up, and Adam Williamson suggested it should be a page of its own, linked from the main Meetings page[7], as other SOP pages are. Edward promised to work on finalizing the page for the next meeting. Jeff Hann volunteered to help Richard with the project, and they agreed to co-ordinate after the meeting.

Adam Williamson started a discussion on Jóhann Guðmundsson's proposal to standardize the Wiki pages containing instructions on debugging problems in various Fedora components. The group generally agreed that it was a good idea, and Jóhann agreed to work on a proposed standard template for these pages, to be discussed and refined on the mailing list.

The next QA weekly meeting will be held on 2009-10-12 at 1600 UTC in #fedora-meeting, and the next Bugzappers weekly meeting on 2009-10-13 at 1500 UTC in #fedora-meeting.

Wiki debugging page standardization

Following discussion in the meeting, Jóhann Guðmundsson submitted his proposal[1] for standardizing Wiki pages which explain how to debug and file proper bug reports on Fedora components. The proposal was enthusiastically received, with additional suggestions coming from Giovanni Tirloni[2], Christopher Beland[3], Adam Williamson[4] and others. A template page[5] is now in place and being refined, and some debugging pages have already been revised to use the template. The group will convert other pages one by one.

Fedora 12 Beta release process

Adam Williamson represented the QA group at the 2009-10-05 release engineering meeting[1]. During the discussion of whether or not to release the Fedora 12 Beta according to the original schedule, Adam presented the QA group's position that the Beta release, and all subsequent events, should be delayed by a week due to the fact that all release blocker bugs had not been resolved. Other groups present agreed, and the meeting resulted in an agreement to delay the release due to this wide consensus.

As a result, the group continued to work on aiding the process of resolving release blocker bugs during the week, and a final blocker bug review meeting was held on Friday 2009-10-09. James Laska provided a recap of this meeting[2].