test(discovery): fixes admin controller test intermittent failures

[guido9j]

Fixes com.netflix.spinnaker.orca.controllers.AdminControllerTest intermittent failures. The failure seems to be caused by a start up issue. The DiscoveryActivator is not initialized when the test is checking for its initial status. This fix adds wait/sleep, so DiscoveryActivator can be fully initialized.

Here is the test log showing failure:
TEST-com.netflix.spinnaker.orca.controllers.AdminControllerTest.xml.log

Here is the partial output of test when it encounters issue, and successful:

2021-12-09T04:09:24.5943329Z 2021-12-09 04:08:58.103 INFO 7552 --- [ Test worker] o.s.t.web.servlet.TestDispatcherServlet : Completed initialization in 239 ms
2021-12-09T04:09:24.5945652Z 2021-12-09 04:08:58.580 WARN 7552 --- [ events-6] c.n.s.k.d.NoDiscoveryStatusPublisher : No service discovery client is available, assuming application is UP
2021-12-09T04:09:24.5948049Z 2021-12-09 04:08:58.606 INFO 7552 --- [ Test worker] c.n.s.o.c.AdminControllerTest$Fixture : Started AdminControllerTest.Fixture in 32.505 seconds (JVM running for 44.683)

// DiscoveryActivator not ready when test starts
2021-12-09T04:09:24.5949692Z discoveryActivator.enabled = false. Sleep 1sec

2021-12-09T04:09:24.5951472Z 2021-12-09 04:08:58.788 INFO 7552 --- [ events-1] c.n.s.k.d.DiscoveryStatusListener : Instance status has changed to UP in service discovery
2021-12-09T04:09:24.5953626Z 2021-12-09 04:08:58.788 INFO 7552 --- [ events-19] c.n.s.q.discovery.DiscoveryActivator : Instance is UP... DiscoveryActivator starting
2021-12-09T04:09:24.5955891Z 2021-12-09 04:08:58.826 INFO 7552 --- [ scheduler-6] s.k.p.u.r.r.RemotePluginInfoReleaseCache : Cached 0 remote plugin configurations.

// After sleeping 1 second, DiscoveryActivator is ready
2021-12-09T04:09:24.5957448Z discoveryActivator.enabled = true. Start

2021-12-09T04:09:24.5959311Z 2021-12-09 04:08:59.838 INFO 7552 --- [ events-4] c.n.s.q.discovery.DiscoveryActivator : Instance is OUT_OF_SERVICE... DiscoveryActivator stopping
2021-12-09T04:09:24.5961573Z 2021-12-09 04:08:59.838 INFO 7552 --- [ events-7] c.n.s.k.d.DiscoveryStatusListener : Instance status has changed to OUT_OF_SERVICE in service discovery

[guido9j]

@dbyron-sf here is a possible fix for that intermittent test failing. There is probably a better way to deal with this, but this is working as far as I can tell.

[dbyron-sf]

I'm going to re-run the job again, and I agree there's likely a better fix, but this feels like progress. Thanks!

[dbyron-sf]

@Mergifyio backport release-1.27.x

[mergify]

backport release-1.27.x

✅ Backports have been created

• #4249 test(discovery): fixes admin controller test intermittent failures (backport #4202) has been created for branch release-1.27.x