integration test: connection timed out; no servers could be reached

[tstromberg]

Example: https://storage.googleapis.com/minikube-builds/logs/3161/Linux-VirtualBox.txt

--- FAIL: TestFunctional/DNS (101.69s)
util.go:185: Error ;; connection timed out; no servers could be reached

		command terminated with exit code 1
		 running command [exec busybox-qf5m4 nslookup kubernetes]. Return code: exit status 1
	util.go:185: Error ;; connection timed out; no servers could be reached
		
		command terminated with exit code 1
		 running command [exec busybox-qf5m4 nslookup kubernetes]. Return code: exit status 1
	util.go:185: Error ;; connection timed out; no servers could be reached
		
		command terminated with exit code 1
		 running command [exec busybox-qf5m4 nslookup kubernetes]. Return code: exit status 1
	cluster_dns_test.go:62: running nslookup in pod:Temporary Error: Error running command. Error  exit status 1. Output: ;; connection timed out; no servers could be reached
		
		command terminated with exit code 1
		
		Temporary Error: Error running command. Error  exit status 1. Output: ;; connection timed out; no servers could be reached
		
		command terminated with exit code 1
		
		Temporary Error: Error running command. Error  exit status 1. Output: ;; connection timed out; no servers could be reached

[tstromberg]

git bisect points to this change, after which the test no longer waits for the pods to be running:

0a5efe1

I'll see what I can do to extend the timeout, or better yet, improve the error message.