Removing KubernetesPipelineTest.runWithSlaveConnectTimeout

[jglick]

Removing a test added in #421 (and lightly edited in #447).

In #503 I saw that it had flaked, which I could reproduce as follows:

diff --git a/src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/KubernetesPipelineTest.java b/src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/KubernetesPipelineTest.java
index 4b88e385..1ff45bf7 100644
--- a/src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/KubernetesPipelineTest.java
+++ b/src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/KubernetesPipelineTest.java
@@ -344,6 +344,7 @@ public class KubernetesPipelineTest extends AbstractKubernetesPipelineTest {
     public void runWithSlaveConnectTimeout() throws Exception {
         r.waitForMessage("podTemplate", b);
 
+        Thread.sleep(60_000);
         PodTemplate timeoutTemplate = cloud.getAllTemplates().stream().filter(x -> name.getMethodName().equals(x.getLabel())).findAny().orElse(null);
 
         assertNotNull(timeoutTemplate);

This is because it was waiting for podTemplate to run and then grabbing metadata, but that metadata will soon be deleted. That aspect is easily correctable using SemaphoreStep. But then I started to see the assertLogNotContains fail. Why should the log not contain this message? We are setting a 10s timeout on agent connection, and either the agent connects in <10s or it connects in >10s; we cannot predict that in the test. So then I was going to delete the log assertion and just use a 101s timeout (distinct from DEFAULT_SLAVE_JENKINS_CONNECTION_TIMEOUT), but wondered what the whole point of the test was to begin with? Presumably to verify that some configuration was taking effect. But looking again at #421 I see that it was patching KubernetesDeclarativeAgent, normally tested in KubernetesDeclarativeAgentTest, yet this test was using a Scripted Pipeline—not even touching the src/main/ code being patched! So it does not seem to have had much if any value.

[Vlatombe]

lgtm