Blocker Tracking App Deadlines and Schedules for F19

Fri Mar 1 05:53:38 UTC 2013

As we start getting closer to the time where we're all busy testing
Fedora 19, it's time to start looking at what we want to get wrapped up
before then.

Looking at the Fedora 19 schedule [1], we have:

 2013-03-12: F19 Branch from Rawhide
 2013-04-02: Feature Freeze

[1] https://fedoraproject.org/wiki/Releases/19/Schedule

The way I'm thinking, the drop-dead-date is 2013-04-02. I don't
remember a release where we weren't mostly consumed by testing tasks by
the freeze date and even though we don't _think_ F19 is going to be a
huge problem - that doesn't mean that we won't be busy :). I don't
want to find ourselves in a situation where we really should be working
on F19 but need to be wrapping up devel/integration work on any of the
tools.

The way I see it, we have some critical stuff left to do and some more
nice-to-have stuff that won't cause too many problems if left
unfinished. I've created a list at the end of this mail listing groups
of tickets in order of priority.

I'm not worried about (1) and (2) on that list. (1) is pretty much done
and (2) should be pretty quick. (3) could be a problem but should be
workable before F19 branches.

That leaves (4) and (5) as the larger items which I don't think _have_
to be done before F19 but it would be really nice if they were done.
Martin and I should know more about the likelihood of their completion
in the next couple of days.

(6) is a bunch of stuff that would be nice to have but I'm not going to
lose any sleep if they aren't done in time. They would be great small
things to pick up if anyone else is interested, though :)

Tim

== Critical to Finish Before F19 ==

1. Finish in-progress work
  - https://fedorahosted.org/fedora-qa/ticket/334
    * Implement Blocker/FE Proposal Page
  - https://fedorahosted.org/fedora-qa/ticket/332
    * Add Fedora 19 Milestones

  These are either mostly done or close to being done - code reviews
  have been filed and code should be in develop soon

2. fixing some small bugs reported during F18
  - https://fedorahosted.org/fedora-qa/ticket/314
    * Bug summaries are not updated
  - https://fedorahosted.org/fedora-qa/ticket/330
    * Blocker Components are not updated when a bug is reassigned

  These bugs were found during the F18 test cycle and shouldn't be too
  hard to actually get done. Since they were reported by users, the
  issues were noticed and thus really need to be fixed before F19
  testing gets into full swing.

3. packaging and making sure we're able to use SSL for F19
  - https://fedorahosted.org/fedora-qa/ticket/337
    * Package Blocker Tracking App for Fedora and EPEL
  - https://fedorahosted.org/fedora-qa/ticket/359
    * Create Ansible Playbooks for Deployment and Updating

  Now that we're using FAS logins for blocker proposals, we need to be
  able to use SSL so that passwords aren't transmitted in the clear.
  While using a self-signed cert is an option, it is kind of a last
  resort that decreases the usability of the site.

  I've been talking to the infra folks and the best way forward is if
  we can get the tracking app to the point where we are hosted in the
  production infrastructure. This means that we should be packaged in
  EPEL and have Ansible scripts for the deployment and updating of the
  app.

  I already have the Ansible playbooks mostly written - they mostly
  need some cleanup before submitting them. I'm not exactly sure how to
  handle alembic updates with playbooks and the package but I think
  that's something which could be figured out with packaging.

  The potential snag here is the CSS - right now that is being compiled
  from SCSS using sass/compass and the zurb-foundation framework. None
  of these things are currently packaged for EPEL6 (sass and compass
  are at least packaged for fedora so unless there are problems with
  the ruby version in EL6, those shouldn't be difficult). It's possible
  that CSS could be considered content instead of code which would
  allow us to continue distributing the compiled css instead of needing
  to compile it at build time.

== Would be VERY nice to have but not absolutely critical for F19 ==

4. improving the blocker submission so that it's either not so slow or
   not vulnerable to refresh during submission
  - https://fedorahosted.org/fedora-qa/ticket/356
    * Blocker Proposal is Slow

   I haven't filed any detailed tickets for this yet because while I
   have ideas on how I want to fix this, I'm not sure if they would
   be practical or if there is enough time before F19 to actually get
   them done.

5. Bugzilla auth integration
  - https://fedorahosted.org/fedora-qa/ticket/347
    * Implement Bugzilla Account Association
  - https://fedorahosted.org/fedora-qa/ticket/348
    * Store Bugzilla Email Association in Database
  - https://fedorahosted.org/fedora-qa/ticket/349
    * Implement Bugzilla Email Verification
  - https://fedorahosted.org/fedora-qa/ticket/350
    * Integration Bugzilla Email Association with Blocker Proposal

  At the moment, Martin is planning to do this - not sure if it'll be
  done in time but we're going to be discussing that tomorrow.

== Nice to have for F19 but not absolutely critical ==

6. misc. cleanup (minifying css and js, reworking templates so that
   only the js _needed_ is loaded on every page, adding some links)
  - https://fedorahosted.org/fedora-qa/ticket/357
    * Improve asset management
  - https://fedorahosted.org/fedora-qa/ticket/358
    * Optimize Template CSS and JS includes
  - https://fedorahosted.org/fedora-qa/ticket/333
    * Add visible Link to IRC Meeting Prep
  - https://fedorahosted.org/fedora-qa/ticket/343
    * Add cgit link to Main Page
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: not available
URL: <http://lists.fedoraproject.org/pipermail/qa-devel/attachments/20130228/a865532b/attachment.sig>