I updated https://wiki.mozilla.org/B2G/QA/Triage#Summary_of_Keywords_and_Flags to include more information on what was called out in the below email. The decision on these rules was made internally within the QA team. I realize now that this wasn't well communicated publicly when we determined these rules, which is probably why there's confusion here.
----- Original Message ----- > From: "Tim Chien" <[email protected]> > To: "Jason Smith" <[email protected]> > Cc: "dev-b2g" <[email protected]>, "dev-gaia" > <[email protected]> > Sent: Wednesday, September 10, 2014 11:58:20 PM > Subject: Re: QA Keywords - Focusing on Using One Keyword to Focus on the > Immediate Need > > Jason, > > Thank for sending out the note here. > > -- Should the rule here be document somewhere on the wiki instead of > burying in the mailing list? > -- AFAIK many of the rules was being enforced months before this note. > Who made the decision on these rules and why developers wasn't > properly communicated (and consulted) beforehand? > > I am not saying the rules doesn't make sense -- they make a lot of > sense given out QA resource constraints -- I just want to make sure we > don't have frustration coming out of hidden rules in the future. > > > On Sat, Sep 6, 2014 at 7:43 AM, Jason Smith <[email protected]> wrote: > > Hi Everyone, > > > > I've received feedback from our test team that some people are adding > > multiple QA keywords onto a bug, instead of adding a single keyword for > > the immediate need, which is causing confusion to the test team on what > > the immediate need of the QA request. The process is documented here - > > https://wiki.mozilla.org/B2G/QA/Triage#Summary_of_Keywords_and_Flags, but > > in summary: > > > > * Only one keyword should be used to focus on the immediate QA need. > > ** Example: We aim to get branch checks first before we do a window, as > > branch checks helps us confirm an issue is truly a regression before we > > start to bisect this. In this case, we start with qawanted here to do > > branch checks. After branch checks are complete, the tester will add > > regressionwindow-wanted to work on getting a window for a particular bug > > if it's a blocking issue. > > ** Exception is for qaurgent - that's used in conjunction with a QA keyword > > to bump the priority of the QA request. It should only be used for bugs > > with critical impact at a bare bones level (e.g. smoketest, all automation > > is down due to X bug). > > * We only focus on doing regression windows on blocker bugs only due to > > resource cost requirements needed to do a window. If you would like to > > have a window for a particular bug, then make sure to figure out if the > > issue is a blocker first. If it is a blocking issue, then feel free to add > > the regressionwindow-wanted keyword. > > > > Feel free to reach out to Tony and I on any questions you have. > > > > Sincerely, > > Jason Smith > > _______________________________________________ > > dev-gaia mailing list > > [email protected] > > https://lists.mozilla.org/listinfo/dev-gaia > > > > -- > Tim Guan-tin Chien, Engineering Manager and Front-end Lead, Firefox > OS, Mozilla Corp. (Taiwan) > _______________________________________________ dev-b2g mailing list [email protected] https://lists.mozilla.org/listinfo/dev-b2g
