865274 - email with image attachment is never sent

Status: RESOLVED WORKSFORME

(Firefox OS Graveyard :: Gaia::E-Mail, defect)

Description

[Dietrich Ayala (:dietrich)]

+++ This bug was initially created as a clone of Bug #865270 +++

(This bug for "email never sent", and bug 865270 for a "cancel" action.)

Unagi, v1-train, 4/24

STR:

1. from gallery share an image via email
2. enter TO: address and hit send
3. see the "sending email" screen with spinner

Expected: eventually sends and goes back to Email or Gallery (i don't know)

Actual: says "sending email" forever, and email never received

Comment 1

[Andrew Sutherland [:asuth] (he/him)]

is there a logcat available?
https://wiki.mozilla.org/Gaia/Email/RequiredBugInfo

Comment 2

[Jeni]

Here are the regression window results for when this issue started occurring: 

b2g build 20130410070209 pass
b2g build 20130417070204 pass
b2g build 20130417230203 Pass
b2g build 20130418070205 Fail
b2g build 20130418113917 Fail

Comment 3

[Andrew Sutherland [:asuth] (he/him)]

(In reply to Andrew Sutherland (:asuth) from comment #1)
> is there a logcat available?
> https://wiki.mozilla.org/Gaia/Email/RequiredBugInfo

When reproducing, please collect logcat logs!

Comment 4

[Jeni]

Attachment: Log showing issue

(In reply to Andrew Sutherland (:asuth) from comment #3)
> (In reply to Andrew Sutherland (:asuth) from comment #1)
> > is there a logcat available?
> > https://wiki.mozilla.org/Gaia/Email/RequiredBugInfo
> 
> When reproducing, please collect logcat logs!

Here is the log for the issue.

Comment 5

[Tracy Walker [:tracy]]

I was able to attache and send one of the wall paper images on Leo with:

Gecko  http://hg.mozilla.org/releases/mozilla-b2g18/rev/0c71cbc5fe0c
Gaia   a7b0810580afc734f3d5e441914fe895f9c1923e
BuildID 20130508230207
Version 18.0

Comment 6

[Angela Hubenya]

removing qawanted per comment 4

Comment 7

[Andrew Sutherland [:asuth] (he/him)]

(In reply to Jeni from comment #4)
> Here is the log for the issue.

What build is this log from?

The regression window matches up with when we broke mozTCPSocket, which makes complete sense.  Dietrich's bug is filed the day after we fixed mozTCPSocket, but he doesn't provide a build id, so it's very possible his build in fact still didn't have the fix.

Based on Tracy's comment 5 and the fact that we had a known total failure of TCPSocket nearly exactly matching the time window, I'm resolving this WFM.

It's possible I meant to probably blame the problem on the TCPSocket error after confirming the content of dietrich's logs given the ambiguity of the timestamp, but dietrich never provided them.  (I probably should have used needinfo.)

Comment 8

[Jeni]

(In reply to Andrew Sutherland (:asuth) from comment #7)
> (In reply to Jeni from comment #4)
> > Here is the log for the issue.
> 
> What build is this log from?

build 20130418070205