Adding an example multi-machine test for Chromoting. This example starts a test controller machine …

Adding an example multi-machine test for Chromoting. This example starts a test controller machine which then starts a dedicated task machine to run the browser_tests_launcher on. The output is then streamed back to the test controller to be logged and the output (pass or fail) is mirrored on the test controller. This is a simple example of how to write and run multi-machine tests using the Legion framework.

BUG=467620
Committed: https://crrev.com/36da2435cdfe6cfd3775aa63c4bf36424f7980cc
Cr-Commit-Position: refs/heads/master@{#320945}

[Mike Meade, 2015-03-16 18:18:46]

mmeade@chromium.org changed reviewers:
+ anandc@chromium.org, andrewrs@chromium.org

[M-A Ruel, 2015-03-16 23:23:55]

maruel@chromium.org changed reviewers:
+ maruel@chromium.org

[M-A Ruel, 2015-03-16 23:23:56]

lgtm

[anandc, 2015-03-16 23:24:47]

Thanks Mike. PTAL.

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/integrat...
File testing/chromoting/integration_tests.gyp (right):

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/integrat...
testing/chromoting/integration_tests.gyp:34: 'dependencies': [
Would adding a dependency here like:
'./integration_tests.gyp:chromoting_intergration_tests_run',

Or maybe even just:
chromoting_intergration_tests_run

work?
Then you just have to include the new sources here.

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
File testing/chromoting/multi_machine_example/example_test_controller.py
(right):

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
testing/chromoting/multi_machine_example/example_test_controller.py:59: def
TaskAbsPath(self, path):
Add an Args section for 'path'?

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
testing/chromoting/multi_machine_example/example_test_controller.py:100: """Runs
the integration tests (via browser_tests_launcher.py."""
Missing ), or remove (

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
testing/chromoting/multi_machine_example/example_test_controller.py:120: #
Output the test's stdout and stderr in semi-realtime.
It would help to explain a bit why it is not fully realtime.
Is this semi-realtime because of the 1s sleep?

[Mike Meade, 2015-03-16 23:35:50]

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/integrat...
File testing/chromoting/integration_tests.gyp (right):

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/integrat...
testing/chromoting/integration_tests.gyp:34: 'dependencies': [
On 2015/03/16 23:24:46, anandc wrote:
> Would adding a dependency here like:
> './integration_tests.gyp:chromoting_intergration_tests_run',
> 
> Or maybe even just:
> chromoting_intergration_tests_run
> 
> work?
> Then you just have to include the new sources here.

As per M-A we would need to refactor out the common items into a gypi file. This
may or may not be the right way to go long term. Since its only a few lines my
feeling is to leave it this way unless it grows any larger. The new tests may
just get added to the sources section of this new target instead of adding new
targets. In that case we would probably just rename this to
chromoting_multi_machine_tests or something.

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
File testing/chromoting/multi_machine_example/example_test_controller.py
(right):

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
testing/chromoting/multi_machine_example/example_test_controller.py:59: def
TaskAbsPath(self, path):
On 2015/03/16 23:24:46, anandc wrote:
> Add an Args section for 'path'?

Done.

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
testing/chromoting/multi_machine_example/example_test_controller.py:100: """Runs
the integration tests (via browser_tests_launcher.py."""
On 2015/03/16 23:24:46, anandc wrote:
> Missing ), or remove (

Good catch.

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/multi_ma...
testing/chromoting/multi_machine_example/example_test_controller.py:120: #
Output the test's stdout and stderr in semi-realtime.
On 2015/03/16 23:24:46, anandc wrote:
> It would help to explain a bit why it is not fully realtime.
> Is this semi-realtime because of the 1s sleep?

I added it. There are 2 RPC calls (Poll and ReadOutput) and the sleep which adds
latency. Even if I didn't include the sleep it wouldn't be true realtime due to
the RPC call overhead.

[anandc, 2015-03-16 23:52:10]

LGTM % one comment.

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/integrat...
File testing/chromoting/integration_tests.gyp (right):

https://codereview.chromium.org/1012883002/diff/1/testing/chromoting/integrat...
testing/chromoting/integration_tests.gyp:34: 'dependencies': [
On 2015/03/16 23:35:50, Mike Meade wrote:
> On 2015/03/16 23:24:46, anandc wrote:
> > Would adding a dependency here like:
> > './integration_tests.gyp:chromoting_intergration_tests_run',
> > 
> > Or maybe even just:
> > chromoting_intergration_tests_run
> > 
> > work?
> > Then you just have to include the new sources here.
> 
> As per M-A we would need to refactor out the common items into a gypi file.
This
> may or may not be the right way to go long term. Since its only a few lines my
> feeling is to leave it this way unless it grows any larger. The new tests may
> just get added to the sources section of this new target instead of adding new
> targets. In that case we would probably just rename this to
> chromoting_multi_machine_tests or something.

OK, could you add a TODO to that effect please?

[Mike Meade, 2015-03-17 00:15:12]

The CQ bit was checked by mmeade@chromium.org

[Mike Meade, 2015-03-17 00:15:13]

The patchset sent to the CQ was uploaded after l-g-t-m from maruel@chromium.org,
anandc@chromium.org
Link to the patchset: https://codereview.chromium.org/1012883002/#ps40001
(title: "Addressing final comments")

[Mike Meade, 2015-03-17 00:15:15]

The CQ bit was unchecked by mmeade@chromium.org

[Mike Meade, 2015-03-17 15:35:02]

The CQ bit was checked by mmeade@chromium.org

[commit-bot: I haz the power, 2015-03-17 15:35:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1012883002/40001

[commit-bot: I haz the power, 2015-03-17 15:55:40]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-03-17 15:55:41]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[Mike Meade, 2015-03-17 17:13:34]

The CQ bit was checked by mmeade@chromium.org

[Mike Meade, 2015-03-17 17:13:34]

The patchset sent to the CQ was uploaded after l-g-t-m from anandc@chromium.org,
maruel@chromium.org
Link to the patchset: https://codereview.chromium.org/1012883002/#ps60001
(title: "Adding the executable bit to the test controller")

[commit-bot: I haz the power, 2015-03-17 17:14:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1012883002/60001

[commit-bot: I haz the power, 2015-03-17 18:25:21]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-03-17 18:26:18]

Patchset 4 (id:??) landed as
https://crrev.com/36da2435cdfe6cfd3775aa63c4bf36424f7980cc
Cr-Commit-Position: refs/heads/master@{#320945}