metrics: Shutdown LoginState after CrOSMetricsProvider tests

metrics: Shutdown LoginState after CrOSMetricsProvider tests

BUG=chromium:379140
TEST=unit tests pass

Signed-off-by: Simon Que <sque@chromium.org>;

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=274012

[Simon Que, 2014-05-30 21:33:58]

FYI I'm creating this because the CQ has already started trying the other CL:
https://codereview.chromium.org/310493002/

If that fails, I'll merge this one with that one.

[Alexei Svitkine (slow), 2014-05-30 21:46:56]

lgtm

Any idea why this sort of thing wasn't caught when the initial refactoring
commits landed? Are some tests simply not being run on the main waterfall on
some platforms?

[Simon Que, 2014-05-30 21:48:00]

On 2014/05/30 21:46:56, Alexei Svitkine wrote:
> lgtm
> 
> Any idea why this sort of thing wasn't caught when the initial refactoring
> commits landed? Are some tests simply not being run on the main waterfall on
> some platforms?

It's possible that the ordering of tests is non-deterministic. Still, with so
many tests, I'd think at least one of them should catch it.

[Simon Que, 2014-05-30 21:48:31]

The CQ bit was checked by sque@chromium.org

[commit-bot: I haz the power, 2014-05-30 21:49:51]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/sque@chromium.org/302153003/1

[commit-bot: I haz the power, 2014-05-31 02:27:58]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  android_dbg_triggered_tests on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/android_dbg_triggered...)

[commit-bot: I haz the power, 2014-05-31 04:20:44]

Change committed as 274012

[blundell, 2014-06-03 11:44:49]

On 2014/05/30 21:48:00, Simon Que wrote:
> On 2014/05/30 21:46:56, Alexei Svitkine wrote:
> > lgtm
> > 
> > Any idea why this sort of thing wasn't caught when the initial refactoring
> > commits landed? Are some tests simply not being run on the main waterfall on
> > some platforms?
> 
> It's possible that the ordering of tests is non-deterministic. Still, with so
> many tests, I'd think at least one of them should catch it.

The reason is due to the test retry logic: when a test fails it is retried at
the end of the run, and when these tests are run in isolation they succeed fine.
:(

The one fyi bot that was reporting the failures presumably doesn't do retries.