800110 - Mirror mozbase -> m-c for bug 800097

Status: RESOLVED FIXED

(Testing :: Mozbase, defect)

Description

[Andrew Halberstadt [:ahal]]

Sync time!

Comment 1

[Jeff Hammel]

bug 799605, the previous mirroring is still in the works, so i don't think we should land on inbound until that makes it to m-c (presumedly tonight), but we can go ahead and generate the diff and test

Comment 2

[Jeff Hammel]

Attachment: da60c88b8c7eace5d24b250395268b7af040127a

Comment 3

[Andrew Halberstadt [:ahal]]

Comment on attachment 670423 [details] [diff] [review]
da60c88b8c7eace5d24b250395268b7af040127a

Review of attachment 670423 [details] [diff] [review]:
-----------------------------------------------------------------

r+, assuming those process handler changes aren't the ones that were supposed to be reverted.

Comment 4

[Jeff Hammel]

(In reply to Andrew Halberstadt [:ahal] from comment #3)
> Comment on attachment 670423 [details] [diff] [review]
> da60c88b8c7eace5d24b250395268b7af040127a
> 
> Review of attachment 670423 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> r+, assuming those process handler changes aren't the ones that were
> supposed to be reverted.

No, this time I *actually* finished (and tested) my patch before pushing ;) [I actually have no idea how i screwed up before; probably forgot a `git checkout <branch>` at some point in time]

Comment 5

[Jeff Hammel]

pushed to try: https://tbpl.mozilla.org/?tree=Try&rev=e3b8c1b61504

If we're lucky, we'll hit this night's inbound

Comment 6

[Jeff Hammel]

Traceback (most recent call last):
  File "mochitest/runtestsremote.py", line 14, in <module>
    from remoteautomation import RemoteAutomation
  File "/builds/tegra-357/test/build/tests/mochitest/remoteautomation.py", line 13, in <module>
    from devicemanager import NetworkTools, DMError
  File "/builds/tegra-357/test/build/tests/mochitest/devicemanager.py", line 13, in <module>
    from Zeroconf import Zeroconf, ServiceBrowser
ImportError: No module named Zeroconf
program finished with exit code 1

Comment 7

[Jeff Hammel]

Long story short, this is because we don't run on the slaves using the full python packages; we just shove all the things in test.zip and run from there.  So we need to change the definition of what all the things is:

http://mxr.mozilla.org/mozilla-central/search?string=mozdevice&find=&findi=&filter=^[^\0]*%24&hitlimit=&tree=mozilla-central

Comment 8

[Andrew Halberstadt [:ahal]]

See also: https://bugzilla.mozilla.org/show_bug.cgi?id=800116

Unfortunately fixing this means moving all fennec automation to mozharness (or doing something hacky).

Comment 9

[Jeff Hammel]

Attachment: fixes to Makefiles

to be landed concurrently/merged with attachment 670423 [details] [diff] [review]

Comment 10

[William Lachance (:wlach)]

Comment on attachment 670481 [details] [diff] [review]
fixes to Makefiles

Yup, this is what we need

Comment 11

[Jeff Hammel]

Attachment: combined diff, for posterity

Comment 12

[Jeff Hammel]

pushed combined diff to try: https://tbpl.mozilla.org/?tree=Try&rev=05a63389adfc

Comment 13

[Jeff Hammel]

This looks green, pushed to inbound: https://hg.mozilla.org/integration/mozilla-inbound/rev/64b6e5be0426

Comment 14

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/64b6e5be0426

Comment 15

[Jeff Hammel]

tagged https://github.com/mozilla/mozbase/zipball/mozilla-central

Comment 16

[Jonathan Griffin (:jgriffin)]

Pushed to ash as aurora-try: https://tbpl.mozilla.org/?tree=Ash&rev=18b5195e134c

Comment 17

[Jonathan Griffin (:jgriffin)]

Ash run looks ok, landed on aurora: https://hg.mozilla.org/releases/mozilla-aurora/rev/d48279987712