communicators/winrm: respect boot_timeout when fetching winrm_info

We gained a ton of improvemnts to WinRM error handling in
#4943, but we also got one bug.

The new code raises an exception when `winrm_info` does not return right
away. This was preventing us from catching the retry/timout logic that's
meant to wait until boot_timeout for the WinRM communicator to be ready.

This restores the proper behavior by rescuing the WinRMNotReady
exception and continuing to retry until the surrounding timeout fires.

plugins/communicators/winrm/communicator.rb

@@ -30,7 +30,12 @@ def wait_for_ready(timeout)
           # Wait for winrm_info to be ready
           winrm_info = nil
           while true
-            winrm_info = Helper.winrm_info(@machine)
+            winrm_info = nil
+            begin
+              winrm_info = Helper.winrm_info(@machine)
+            rescue Errors::WinRMNotReady
+              @logger.debug("WinRM not ready yet; retrying until boot_timeout is reached.")
+            end
             break if winrm_info
             sleep 0.5
           end

test/unit/plugins/communicators/winrm/communicator_test.rb

@@ -5,10 +5,11 @@
 describe VagrantPlugins::CommunicatorWinRM::Communicator do
   include_context "unit"
 
-  let(:winrm) { double("winrm", timeout: 1) }
+  let(:winrm) { double("winrm", timeout: 1, host: nil, port: 5986, guest_port: 5986) }
   let(:config) { double("config", winrm: winrm) }
-  let(:machine) { double("machine", config: config) }
-
+  let(:provider) { double("provider") }
+  let(:ui) { double("ui") }
+  let(:machine) { double("machine", config: config, provider: provider, ui: ui) }
   let(:shell) { double("shell") }
 
   subject do
@@ -22,6 +23,37 @@
     allow(shell).to receive(:password).and_return('password')
   end
 
+  describe ".wait_for_ready" do
+    context "with no winrm_info capability and no static config (default scenario)" do
+      before do
+        # No default providers support this capability
+        allow(provider).to receive(:capability?).with(:winrm_info).and_return(false)
+
+        # Get us through the detail prints
+        allow(ui).to receive(:detail)
+        allow(shell).to receive(:host)
+        allow(shell).to receive(:port)
+        allow(shell).to receive(:username)
+        allow(shell).to receive(:config) { double("config", transport: nil)}
+      end
+
+      context "when ssh_info requires a multiple tries before it is ready" do
+        before do
+          allow(machine).to receive(:ssh_info).and_return(nil, {
+            host: '10.1.2.3',
+            port: '22',
+          })
+          # Makes ready? return true
+          allow(shell).to receive(:powershell).with("hostname").and_return({ exitcode: 0 })
+        end
+
+        it "retries ssh_info until ready" do
+          expect(subject.wait_for_ready(2)).to eq(true)
+        end
+      end
+    end
+  end
+
   describe ".ready?" do
     it "returns true if hostname command executes without error" do
       expect(shell).to receive(:powershell).with("hostname").and_return({ exitcode: 0 })