installer/*: Add --continue-on-error to destroy workflow

This would let us skip destroying things like worker machines. Useful
if the API never came up. As those resources can't exist yet.

I'm currently not passing the continue-on-error down into the individual
steps, but we could do that later if we need more granularity.

Author: wking

installer/cmd/tectonic/main.go

@@ -16,8 +16,9 @@ var (
 	clusterInstallCommand = kingpin.Command("install", "Create a new Tectonic cluster")
 	clusterInstallDirFlag = clusterInstallCommand.Flag("dir", "Cluster directory").Default(".").ExistingDir()
 
-	clusterDestroyCommand = kingpin.Command("destroy", "Destroy an existing Tectonic cluster")
-	clusterDestroyDirFlag = clusterDestroyCommand.Flag("dir", "Cluster directory").Default(".").ExistingDir()
+	clusterDestroyCommand   = kingpin.Command("destroy", "Destroy an existing Tectonic cluster")
+	clusterDestroyDirFlag   = clusterDestroyCommand.Flag("dir", "Cluster directory").Default(".").ExistingDir()
+	clusterDestroyContOnErr = clusterDestroyCommand.Flag("continue-on-error", "Log errors, but attempt to continue cleaning up the cluster.  THIS MAY LEAK RESOURCES, because you may not have enough state left after a partial removal to be able to perform a second destroy.").Default("false").Bool()
 
 	convertCommand    = kingpin.Command("convert", "Convert a tfvars.json to a Tectonic config.yaml")
 	convertConfigFlag = convertCommand.Flag("config", "tfvars.json file").Required().ExistingFile()
@@ -34,7 +35,7 @@ func main() {
 	case clusterInstallCommand.FullCommand():
 		w = workflow.InstallWorkflow(*clusterInstallDirFlag)
 	case clusterDestroyCommand.FullCommand():
-		w = workflow.DestroyWorkflow(*clusterDestroyDirFlag)
+		w = workflow.DestroyWorkflow(*clusterDestroyDirFlag, *clusterDestroyContOnErr)
 	case convertCommand.FullCommand():
 		w = workflow.ConvertWorkflow(*convertConfigFlag)
 	}

installer/pkg/workflow/destroy.go

@@ -22,9 +22,12 @@ const (
 // DestroyWorkflow creates new instances of the 'destroy' workflow,
 // responsible for running the actions required to remove resources
 // of an existing cluster and clean up any remaining artefacts.
-func DestroyWorkflow(clusterDir string) Workflow {
+func DestroyWorkflow(clusterDir string, contOnErr bool) Workflow {
 	return Workflow{
-		metadata: metadata{clusterDir: clusterDir},
+		metadata: metadata{
+			clusterDir: clusterDir,
+			contOnErr:  contOnErr,
+		},
 		steps: []step{
 			readClusterConfigStep,
 			generateTerraformVariablesStep,

installer/pkg/workflow/workflow.go

@@ -1,6 +1,9 @@
 package workflow
 
-import "github.com/openshift/installer/installer/pkg/config"
+import (
+	log "github.com/Sirupsen/logrus"
+	"github.com/openshift/installer/installer/pkg/config"
+)
 
 // metadata is the state store of the current workflow execution.
 // It is meant to carry state for one step to another.
@@ -12,6 +15,7 @@ type metadata struct {
 	cluster        config.Cluster
 	configFilePath string
 	clusterDir     string
+	contOnErr      bool
 }
 
 // step is the entrypoint of a workflow step implementation.
@@ -28,11 +32,18 @@ type Workflow struct {
 
 // Execute runs all steps in order.
 func (w Workflow) Execute() error {
+	var firstError error
 	for _, step := range w.steps {
 		if err := step(&w.metadata); err != nil {
-			return err
+			if !w.metadata.contOnErr {
+				return err
+			}
+			if firstError == nil {
+				firstError = err
+			}
+			log.Warn(err)
 		}
 	}
 
-	return nil
+	return firstError
 }