<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">Hi Julien,<br>
<br>
Could that be the cause of why
<a class="moz-txt-link-freetext" href="https://bugzilla.mozilla.org/show_bug.cgi?id=823189">https://bugzilla.mozilla.org/show_bug.cgi?id=823189</a> is occurring?
We currently can complete an update process for a packaged app,
but if you try to launch it, it will fail.<br>
<br>
As for the smoke tests - I'll followup with Tony about that. It's
probably worthwhile to consider given that many bad
bugs/regressions always turn up during update testing.<br>
<pre class="moz-signature" cols="72">Sincerely,
Jason Smith
Desktop QA Engineer
Mozilla Corporation
<a class="moz-txt-link-freetext" href="https://quality.mozilla.com">https://quality.mozilla.com</a></pre>
On 12/28/2012 12:25 PM, Julien Wajsberg wrote:<br>
</div>
<blockquote cite="mid:50DDD5EE.7020904@mozilla.com" type="cite">
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
Thanks Jason, this makes sense.<br>
<br>
However, working on this bug, I saw another more serious bug that
I fixed in the same patch. Basically, it was preventing from
updating any app that was previously installed with an ETag on the
package file. That is, every packaged app now (and even
preinstalled ones as iirc we plan to ship them with prefilled ETag
values).<br>
<br>
I'm not sure how this could be integrated in the smoketests, but
it should. :-)<br>
(or maybe it is already and this couldn't be checked because of
the previous bug; if this is the case everything's fine)<br>
<br>
Le 28/12/2012 18:18, Jason Smith a écrit :
<blockquote cite="mid:50DDD471.4010401@mozilla.com" type="cite">
<meta content="text/html; charset=UTF-8"
http-equiv="Content-Type">
<div class="moz-cite-prefix">Hi Alex,<br>
<br>
+Packaged Apps<br>
+Julien<br>
<br>
Let me take a shot at explaining this. Basically, we landed
support for requiring etags to be served for the packaged app
zip archive. However, putting this requirement in ended up
breaking all of packaged app test cases that were <b>not</b>
on marketplace, as none of them were serving the archives with
etags. As a result, the error seen was a "bad" state the phone
could enter where the download of the packaged app could spin
forever right now if you installed a packaged app that had a
zip archive not serving an etag - which is what that bug ended
up being about. It is no longer a smoketest blocker though
given that this is an error case now being handled poorly, not
a primary happy path case.<br>
<br>
The smoke test for that feature will now have to change (note
to QA that you should change this):<br>
<br>
<b>Install a Packaged App from Marketplace</b><br>
<br>
1. Go to <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://marketplace-dev.allizom.org/app/test-webapi-permissions">https://marketplace-dev.allizom.org/app/test-webapi-permissions</a><br>
2. Install the packaged app<br>
<br>
Result - Verify the packaged app was successfully downloaded
with the correct icon and app name.<br>
<br>
Due to <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=825209">https://bugzilla.mozilla.org/show_bug.cgi?id=825209</a>,
this will not work right now as marketplace is not serving
etags for zip archives. So technically bug 825209 is a
smoketest blocker right now, but that won't block pushing a
new build to dogfooders given that this is not a client-side
bug, it's a marketplace bug - which is fixed on the server
side.<br>
<br>
Make sense?<br>
<pre class="moz-signature" cols="72">Sincerely,
Jason Smith
Desktop QA Engineer
Mozilla Corporation
<a moz-do-not-send="true" class="moz-txt-link-freetext" href="https://quality.mozilla.com">https://quality.mozilla.com</a></pre>
On 12/28/2012 11:56 AM, akeybl wrote:<br>
</div>
<blockquote
cite="mid:jykr1i13rirsw5k79l745fxc.1356713330644@email.android.com"
type="cite">
<meta http-equiv="Content-Type" content="text/html;
charset=UTF-8">
<div>Hi Marcia and Jason,</div>
<div><br>
</div>
Aren't important Marketplace apps like Twitter already
packaged apps? Can you confirm whether or not this is a
regression? The bug isn't clear to me.
<div><br>
</div>
<div>If it's not a regression and doesn't affect important
Marketplace apps, then the 12/26 build may be a good choice.</div>
<div><br>
</div>
<div>-Alex</div>
<br>
Marcia Knous <a moz-do-not-send="true"
class="moz-txt-link-rfc2396E"
href="mailto:marcia@mozilla.org"><marcia@mozilla.org></a>
wrote:<br>
Naoki was asking about this yesterday, so I thought I would
mention that <br>
prior to 12-25 the builds suffered from <br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=821810">https://bugzilla.mozilla.org/show_bug.cgi?id=821810</a>,
which is a blocker. <br>
According to the comments that started working again on 12-25.<br>
<br>
We started smoketesting again on 12-26. Since then, the large
issues <br>
continue to be (including TEF results from the 12-28 build
today)<br>
<br>
*<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=824695">https://bugzilla.mozilla.org/show_bug.cgi?id=824695</a>
- Fix bustage from <br>
requiring etags for packaged apps to keep the phone in a
consistent <br>
state during download of a packaged app (Since this is
packaged apps <br>
only is this a blocker?)<br>
* Facebook button in the Contacts page is disabled. -<br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=825017">https://bugzilla.mozilla.org/show_bug.cgi?id=825017</a><br>
*[NEW] [CONTACTS] Contact details are not shown correctly- <br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=825166">https://bugzilla.mozilla.org/show_bug.cgi?id=825166</a>
(Doesn't seem to <br>
affect English)<br>
* <a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=822913">https://bugzilla.mozilla.org/show_bug.cgi?id=822913</a>
- Still seen in <br>
daily builds, but not seen unless you use the Video app to
record quite <br>
a bit.<br>
<br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://docs.google.com/spreadsheet/ccc?key=0AqDJxnTd7VDLdDRKOE1uanE3M3FtMDBOVjRIV253S1E#gid=54">https://docs.google.com/spreadsheet/ccc?key=0AqDJxnTd7VDLdDRKOE1uanE3M3FtMDBOVjRIV253S1E#gid=54</a>
<br>
links to the smoketest results from the week so far.<br>
<br>
<br>
<br>
<br>
akeybl wrote:<br>
> Hi Marcia and B2G QA,<br>
><br>
> It's been a while since we've pushed a new stable build
to our test drivers, so we're looking for the best build from
the past week or so.<br>
><br>
> Can you recommend a build from the last week that we can
push? I think bug 824695 makes 12/26 a non-starter for now.<br>
><br>
> Would be great to hear back tomorrow morning so that we
can ask RelEng to push a new build out tomorrow during the
day. I'd be happy to test the stable->stable update once
pushed.<br>
><br>
> -AlexMarcia Knous<a moz-do-not-send="true"
class="moz-txt-link-rfc2396E"
href="mailto:marcia@mozilla.org"><marcia@mozilla.org></a>
wrote:Hello all - Just an update that since I am on the East
Coast I grabbed<br>
> an earlier build out of the "latest" directory this
morning (3 hour head<br>
> start :)). That build still had<br>
> <a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=818933">https://bugzilla.mozilla.org/show_bug.cgi?id=818933</a>,
but the latest<br>
> build does not exhibit that issue.<br>
><br>
> Both <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=822913">https://bugzilla.mozilla.org/show_bug.cgi?id=822913</a>
and<br>
> <a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=824695">https://bugzilla.mozilla.org/show_bug.cgi?id=824695</a>
still remain as<br>
> issues in the 12-26-12 build on my device. The subject
line contains the<br>
> updated count of pass/fail. Sorry for the noise.<br>
><br>
> marcia<br>
><br>
> Marcia Knous wrote:<br>
>> Hi Everyone,<br>
>><br>
>> 33 out of 41 tests passed for the 12/26/2012 build.
Here are the<br>
>> major issues that prevented the smoke tests from
passing:<br>
>><br>
>> Smoketest:<br>
>> <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://docs.google.com/spreadsheet/ccc?key=0AqDJxnTd7VDLdDRKOE1uanE3M3FtMDBOVjRIV253S1E#gid=54">https://docs.google.com/spreadsheet/ccc?key=0AqDJxnTd7VDLdDRKOE1uanE3M3FtMDBOVjRIV253S1E#gid=54</a><br>
>><br>
>><br>
>> Tests were performed with:<br>
>><br>
>> - gecko: e261861b0270<br>
>> - gaia: b4bcf70792f4<br>
>><br>
>> - 2 reboots<br>
>> - 2 crashes<br>
>><br>
>> Bugs Breaking Smoke Tests*<br>
>><br>
>> [EXISTING] <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=822913">https://bugzilla.mozilla.org/show_bug.cgi?id=822913</a>
- Camera<br>
>> - video recording - "Video not recorded. An error
prevented Camera from<br>
>> recording the video<br>
>> [EXISTING] <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=818933">https://bugzilla.mozilla.org/show_bug.cgi?id=818933</a>
-<br>
>> [Camera] Taking a photo from lockscreen when Camera
has already been<br>
>> launched from the Homescreen causes a hang<br>
>> [NEW] <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=824695">https://bugzilla.mozilla.org/show_bug.cgi?id=824695</a>
- Installation<br>
>> of packaged apps is broken again on 12/26/2012 build<br>
>><br>
>> New Regressions Not Breaking Smoke Tests<br>
>><br>
>> [NEW] <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://bugzilla.mozilla.org/show_bug.cgi?id=824795">https://bugzilla.mozilla.org/show_bug.cgi?id=824795</a>
- Clicking the<br>
>> back button in Settings launches the Camera<br>
>><br>
>><br>
>> Please address these issues as soon as possible to
get these tests passing.<br>
>><br>
>> Sincerely,<br>
>><br>
>> The QA Team<br>
>><br>
>><br>
>> _______________________________________________<br>
>> dev-gaia mailing list<br>
>> <a moz-do-not-send="true"
class="moz-txt-link-abbreviated"
href="mailto:dev-gaia@lists.mozilla.org">dev-gaia@lists.mozilla.org</a><br>
>> <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://lists.mozilla.org/listinfo/dev-gaia">https://lists.mozilla.org/listinfo/dev-gaia</a><br>
> _______________________________________________<br>
> dev-gaia mailing list<br>
> <a moz-do-not-send="true"
class="moz-txt-link-abbreviated"
href="mailto:dev-gaia@lists.mozilla.org">dev-gaia@lists.mozilla.org</a><br>
> <a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://lists.mozilla.org/listinfo/dev-gaia">https://lists.mozilla.org/listinfo/dev-gaia</a><br>
</blockquote>
<br>
</blockquote>
<br>
</blockquote>
<br>
</body>
</html>