WIP: Power VS: Create LB DNS records in IBM DNS for private clusters

.ci-operator.yaml

@@ -1,4 +1,4 @@
 build_root_image:
   name: release
   namespace: openshift
-  tag: rhel-8-release-golang-1.17-openshift-4.11
+  tag: rhel-8-release-golang-1.18-openshift-4.12

OWNERS

@@ -2,7 +2,8 @@
 # This file just uses aliases defined in OWNERS_ALIASES.
 
 approvers:
-  - installer-approvers
+  - installer-team
+  - sdodson
 reviewers:
-  - installer-reviewers
+  - installer-team
 component: Installer

OWNERS_ALIASES

@@ -1,24 +1,15 @@
 # See the OWNERS_ALIASES docs: https://git.k8s.io/community/contributors/guide/owners.md#OWNERS_ALIASES
 
 aliases:
-  installer-approvers:
-    - jhixson74
-    - jstuever
-    - patrickdillon
-    - staebler
-    - sdodson
-    - smarterclayton
-  installer-reviewers:
+  installer-team:
     - AnnaZivkovic
     - barbacbd
     - jhixson74
     - jstuever
-    - kirankt
     - patrickdillon
     - r4f4
     - rna-afk
     - sadasu
-    - staebler
   libvirt-approvers:
     - praveenkumar
     - cfergeau
@@ -27,12 +18,18 @@ aliases:
     - cfergeau
   openstack-approvers:
     - EmilienM
+    - MaysaMacedo
+    - dulek
+    - gryf
     - mandre
     - mdbooth
     - pierreprinetti
     - stephenfin
   openstack-reviewers:
     - EmilienM
+    - MaysaMacedo
+    - dulek
+    - gryf
     - mandre
     - mdbooth
     - pierreprinetti
@@ -47,12 +44,14 @@ aliases:
     - Prashanth684
     - mjturek
   vsphere-approvers:
+    - bostrt
     - dav1x
     - jcpowermac
     - patrickdillon
     - jstuever
     - rvanderp3
   vsphere-reviewers:
+    - bostrt
     - jcpowermac
     - patrickdillon
     - jstuever
@@ -73,9 +72,11 @@ aliases:
     - sadasu
   aws-approvers:
     - jstuever
+    - mtulio
     - patrickdillon
   aws-reviewers:
     - jstuever
+    - mtulio
     - patrickdillon
   gcp-approvers:
     - jstuever
@@ -84,12 +85,12 @@ aliases:
     - jstuever
     - patrickdillon
   azure-approvers:
-    - fabianofranz
     - jhixson74
+    - mtulio
   azure-reviewers:
-    - fabianofranz
     - jhixson74
     - m1kola
+    - mtulio
   ovirt-approvers:
     - janosdebugs
     - eslutsky
@@ -103,20 +104,24 @@ aliases:
   ibmcloud-approvers:
     - bobbyradford
     - hasueki
+    - rvanderp3
   ibmcloud-reviewers:
     - bobbyradford
     - hasueki
+    - rvanderp3
   nutanix-approvers:
+    - adiantum
+    - rvanderp3
     - thunderboltsid
     - yanhua121
-    - adiantum
   nutanix-reviewers:
+    - adiantum
+    - rvanderp3
     - thunderboltsid
     - yanhua121
-    - adiantum
   # Keep in sync with the `coreos-approvers` and `coreos-reviewers` lists in https://github.com/openshift/release/blob/master/OWNERS_ALIASES
   coreos-approvers:
-    - miabbott
+    - c4rt0
     - cgwalters
     - bgilbert
     - dustymabe
@@ -126,13 +131,16 @@ aliases:
     - lucab
     - sohankunkerkar
     - travier
-    - saqibali-2k
     - HuijingHei
     - jmarrero
     - aaradhak
     - ravanelli
+    - gursewak1997
+    - prestist
+    - marmijo
+    - RishabhSaini
   coreos-reviewers:
-    - miabbott
+    - c4rt0
     - cgwalters
     - bgilbert
     - dustymabe
@@ -142,8 +150,29 @@ aliases:
     - lucab
     - sohankunkerkar
     - travier
-    - saqibali-2k
     - HuijingHei
     - jmarrero
     - aaradhak
     - ravanelli
+    - gursewak1997
+    - prestist
+    - marmijo
+    - RishabhSaini
+  agent-reviewers:
+    - andfasano
+    - bfournie
+    - celebdor
+    - dhellmann
+    - lranjbar
+    - pawanpinjarkar
+    - rwsu
+    - zaneb
+  agent-approvers:
+    - andfasano
+    - bfournie
+    - celebdor
+    - dhellmann
+    - lranjbar
+    - pawanpinjarkar
+    - rwsu
+    - zaneb

README.md

@@ -49,7 +49,8 @@ Example output:
 ```sh
 INFO Waiting 10m0s for the openshift-console route to be created...
 INFO Install complete!
-INFO To access the cluster as the system:admin user when using 'oc', run 'export KUBECONFIG=/path/to/installer/auth/kubeconfig'
+INFO To access the cluster as the system:admin user when using 'oc', run
+    export KUBECONFIG=/path/to/installer/auth/kubeconfig
 INFO Access the OpenShift web-console here: https://console-openshift-console.apps.${CLUSTER_NAME}.${BASE_DOMAIN}:6443
 INFO Login to the console with user: kubeadmin, password: 5char-5char-5char-5char
 ```

cmd/openshift-install/agent.go

@@ -0,0 +1,91 @@
+package main
+
+import (
+	"github.com/spf13/cobra"
+
+	"github.com/openshift/installer/cmd/openshift-install/agent"
+	"github.com/openshift/installer/pkg/asset"
+	"github.com/openshift/installer/pkg/asset/agent/agentconfig"
+	"github.com/openshift/installer/pkg/asset/agent/image"
+	"github.com/openshift/installer/pkg/asset/agent/manifests"
+	"github.com/openshift/installer/pkg/asset/agent/mirror"
+	"github.com/openshift/installer/pkg/asset/kubeconfig"
+)
+
+func newAgentCmd() *cobra.Command {
+	agentCmd := &cobra.Command{
+		Use:   "agent",
+		Short: "Commands for supporting cluster installation using agent installer",
+		RunE: func(cmd *cobra.Command, args []string) error {
+			return cmd.Help()
+		},
+	}
+
+	agentCmd.AddCommand(newAgentCreateCmd())
+	agentCmd.AddCommand(agent.NewWaitForCmd())
+	return agentCmd
+}
+
+var (
+	agentConfigTarget = target{
+		// TODO: remove template wording when interactive survey has been implemented
+		name: "Agent Config Template",
+		command: &cobra.Command{
+			Use:   "agent-config-template",
+			Short: "Generates a template of the agent config manifest used by the agent installer",
+			Args:  cobra.ExactArgs(0),
+		},
+		assets: []asset.WritableAsset{
+			&agentconfig.AgentConfig{},
+		},
+	}
+
+	agentManifestsTarget = target{
+		name: "Cluster Manifests",
+		command: &cobra.Command{
+			Use:   "cluster-manifests",
+			Short: "Generates the cluster definition manifests used by the agent installer",
+			Args:  cobra.ExactArgs(0),
+		},
+		assets: []asset.WritableAsset{
+			&manifests.AgentManifests{},
+			&mirror.RegistriesConf{},
+			&mirror.CaBundle{},
+		},
+	}
+
+	agentImageTarget = target{
+		name: "Image",
+		command: &cobra.Command{
+			Use:   "image",
+			Short: "Generates a bootable image containing all the information needed to deploy a cluster",
+			Args:  cobra.ExactArgs(0),
+		},
+		assets: []asset.WritableAsset{
+			&image.AgentImage{},
+			&kubeconfig.AgentAdminClient{},
+		},
+	}
+
+	agentTargets = []target{agentConfigTarget, agentManifestsTarget, agentImageTarget}
+)
+
+func newAgentCreateCmd() *cobra.Command {
+
+	cmd := &cobra.Command{
+		Use:   "create",
+		Short: "Commands for generating agent installation artifacts",
+		Args:  cobra.ExactArgs(0),
+		RunE: func(cmd *cobra.Command, args []string) error {
+			return cmd.Help()
+		},
+	}
+
+	for _, t := range agentTargets {
+		t.command.Args = cobra.ExactArgs(0)
+		t.command.Run = runTargetCmd(t.assets...)
+		cmd.AddCommand(t.command)
+	}
+
+	return cmd
+}

cmd/openshift-install/agent/OWNERS

@@ -0,0 +1,7 @@
+# See the OWNERS docs: https://git.k8s.io/community/contributors/guide/owners.md
+# This file just uses aliases defined in OWNERS_ALIASES.
+
+approvers:
+  - agent-approvers
+reviewers:
+  - agent-reviewers

cmd/openshift-install/agent/waitfor.go

@@ -0,0 +1,89 @@
+package agent
+
+import (
+	"github.com/pkg/errors"
+	"github.com/sirupsen/logrus"
+	"github.com/spf13/cobra"
+
+	agentpkg "github.com/openshift/installer/pkg/agent"
+)
+
+const (
+	exitCodeInstallConfigError = iota + 3
+	exitCodeInfrastructureFailed
+	exitCodeBootstrapFailed
+	exitCodeInstallFailed
+)
+
+// NewWaitForCmd create the commands for waiting the completion of the agent based cluster installation.
+func NewWaitForCmd() *cobra.Command {
+	cmd := &cobra.Command{
+		Use:   "wait-for",
+		Short: "Wait for install-time events",
+		Args:  cobra.ExactArgs(0),
+		RunE: func(cmd *cobra.Command, args []string) error {
+			return cmd.Help()
+		},
+	}
+
+	cmd.AddCommand(newWaitForBootstrapCompleteCmd())
+	cmd.AddCommand(newWaitForInstallCompleteCmd())
+	return cmd
+}
+
+func newWaitForBootstrapCompleteCmd() *cobra.Command {
+	return &cobra.Command{
+		Use:   "bootstrap-complete",
+		Short: "Wait until the cluster bootstrap is complete",
+		Args:  cobra.ExactArgs(0),
+		Run: func(cmd *cobra.Command, args []string) {
+			assetDir := cmd.Flags().Lookup("dir").Value.String()
+			logrus.Debugf("asset directory: %s", assetDir)
+			if len(assetDir) == 0 {
+				logrus.Fatal("No cluster installation directory found")
+			}
+			cluster, err := agentpkg.WaitForBootstrapComplete(assetDir)
+			if err != nil {
+				logrus.Debug("Printing the event list gathered from the Agent Rest API")
+				cluster.PrintInfraEnvRestAPIEventList()
+				err2 := cluster.API.OpenShift.LogClusterOperatorConditions()
+				if err2 != nil {
+					logrus.Error("Attempted to gather ClusterOperator status after wait failure: ", err2)
+				}
+				logrus.Info("Use the following commands to gather logs from the cluster")
+				logrus.Info("openshift-install gather bootstrap --help")
+				logrus.Error(errors.Wrap(err, "Bootstrap failed to complete: "))
+				logrus.Exit(exitCodeBootstrapFailed)
+			}
+		},
+	}
+}
+
+func newWaitForInstallCompleteCmd() *cobra.Command {
+	return &cobra.Command{
+		Use:   "install-complete",
+		Short: "Wait until the cluster installation is complete",
+		Args:  cobra.ExactArgs(0),
+		Run: func(cmd *cobra.Command, args []string) {
+			assetDir := cmd.Flags().Lookup("dir").Value.String()
+			logrus.Debugf("asset directory: %s", assetDir)
+			if len(assetDir) == 0 {
+				logrus.Fatal("No cluster installation directory found")
+			}
+			cluster, err := agentpkg.WaitForInstallComplete(assetDir)
+			if err != nil {
+				logrus.Debug("Printing the event list gathered from the Agent Rest API")
+				cluster.PrintInfraEnvRestAPIEventList()
+				err2 := cluster.API.OpenShift.LogClusterOperatorConditions()
+				if err2 != nil {
+					logrus.Error("Attempted to gather ClusterOperator status after wait failure: ", err2)
+				}
+				logrus.Error(`Cluster initialization failed because one or more operators are not functioning properly.
+				The cluster should be accessible for troubleshooting as detailed in the documentation linked below,
+				https://docs.openshift.com/container-platform/latest/support/troubleshooting/troubleshooting-installations.html`)
+				logrus.Exit(exitCodeInstallFailed)
+			}
+			cluster.PrintInstallationComplete()
+		},
+	}
+}