unittest flake: resttest.go: unexpected timeout from result channel

[jsafrane]

From https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/pr-logs/pull/31978/kubernetes-pull-test-unit-integration/44935/

2016-09-15 01:47:51.549894 I | etcdserver: published {Name:node4735261938323256311 ClientURLs:[unix://127.0.0.1:21012.7913.sock]} to cluster c9c2d39b7c65124f
2016-09-15 01:47:51.598332 W | etcdserver: apply entries took too long [11.933395ms for 1 entries]
2016-09-15 01:47:51.599521 W | etcdserver: avoid queries with large range/delete range!
2016-09-15 01:48:23.871044 I | integration: terminating node4735261938323256311 (unix://localhost:node4735261938323256311.sock.bridge)
2016-09-15 01:48:23.878286 I | integration: terminated node4735261938323256311 (unix://localhost:node4735261938323256311.sock.bridge)
--- FAIL: TestWatch (32.41s)
    resttest.go:1251: unexpected timeout from result channel

The test usually takes 2-3 seconds, here it took 32 s. It happened when testing pr #31978, IMO it's unrelated to the bug.

[pwittrock]

I am encountering this a lot and it is getting painful.

[calebamiles]

@saad-ali, this issue should probably be closed. The associated PR has already been merged and the PR test grid was green except for GKE tests which were non blocking

[saad-ali]

ACK. Closing