Merge branch 'master' into feat/cb-cluster-via-cao

.github/workflows/build-and-test.yml

@@ -38,15 +38,26 @@ jobs:
           ./cbdinocluster-linux init --auto
       - name: Setup Cluster
         run: |
-          CBDC_ID=$(./cbdinocluster-linux allocate simple:7.0.0)
+          CBDC_ID=$(./cbdinocluster-linux -v allocate simple:7.0.0)
           echo "CBDC_ID=$CBDC_ID" >> "$GITHUB_ENV"
       - name: Run Tests
         timeout-minutes: 10
         run: |
-          ./cbdinocluster-linux connstr $CBDC_ID
-          CBDC_MGMT=$(./cbdinocluster-linux mgmt $CBDC_ID)
+          # validate getting the connection string
+          ./cbdinocluster-linux -v connstr $CBDC_ID
+          # validate getting management addresses
+          CBDC_MGMT=$(./cbdinocluster-linux -v mgmt $CBDC_ID)
           echo $CBDC_MGMT
+          # validate cluster is accessible
           curl $CBDC_MGMT
+          # pick a random node to mess with
+          NODE_IP=$(./cbdinocluster-linux -v ip $CBDC_ID)
+          # validate blocking traffic to the node
+          ./cbdinocluster-linux -v chaos block-traffic $CBDC_ID $NODE_IP
+          # validate we can still access that node
+          curl http://$NODE_IP:8091
+          # ensure we can unblock traffic
+          ./cbdinocluster-linux -v chaos allow-traffic $CBDC_ID $NODE_IP
       - name: Cleanup Cluster
         if: always()
         run: |

.gitignore

@@ -1,3 +1,3 @@
 /data/
-/cbdynclusterd
-cleanup_registry/.venv
+/cbdinocluster*
+cleanup_registry/.venv

README.md

@@ -66,6 +66,50 @@ environments.
 ./cbdinocluster rm {{CLUSTER_ID}}
 ```
 
+#### Create a bucket named `default`
+
+```
+./cbdinocluster buckets add {{CLUSTER_ID}} default --ram-quota-mb=100 --flush-enabled=true
+```
+
+#### Create a collection in the default scope on the bucket named `default`
+
+```
+./cbdinocluster collections add {{CLUSTER_ID}} default _default test
+```
+
+#### Using prefixes to match cluster and container IDs
+
+To avoid constantly copying and pasting IDs in terminal, it is recommended to use only unique prefix.
+
+```
+~ $ cbdinocluster ps
+...
+Clusters:
+  4f9e6625-6f48-4866-bab9-076ae5f57052 [State: ready, Timeout: none, Deployer: docker]
+    a20294bf-44fc-4a6f-b019-a005d2fbf16b                       192.168.107.130      af3dcbd5...
+    86da5182-9f3c-450e-8b69-1c65fc646f7b                       192.168.107.128      b6b7ab97...
+    51024e57-660d-497d-8878-42ca7781ab4c                       192.168.107.129      2ca9e448...
+```
+
+Note that the we have only one cluster, and its ID starts with `4`, so it is enough to refert to the cluster with
+cbdinocluster commands.
+
+```
+~ $ cbdinocluster connstr 4
+...
+couchbase://192.168.107.130,192.168.107.128,192.168.107.129
+```
+
+The same work with docker itself (first node ID starts with `a`):
+
+```
+~ $ docker exec -ti a /usr/bin/bash
+root@af3dcbd5b2a4:/# ps aux | grep memcached
+couchba+     384  0.0  0.4 1762336 27268 ?       SLsl 20:39   0:00 /opt/couchbase/bin/memcached -C /opt/couchbase/var/lib/couchbase/config/memcached.json
+root         664  0.0  0.0   2976  1536 pts/0    S+   20:49   0:00 grep --color=auto memcached
+```
+
 ### Advanced Usage
 
 #### Resetting Colima

clusterdef/nodegroup.go

@@ -18,6 +18,7 @@ type NodeGroup struct {
 }
 
 type DockerNodeGroup struct {
+	EnvVars map[string]string `yaml:"env,omitempty"`
 }
 
 type CloudNodeGroup struct {

cmd/allocate.go

@@ -70,6 +70,7 @@ var allocateCmd = &cobra.Command{
 		ctx := helper.GetContext()
 		config := helper.GetConfig(ctx)
 
+		dryRun, _ := cmd.Flags().GetBool("dry-run")
 		defStr, _ := cmd.Flags().GetString("def")
 		defFile, _ := cmd.Flags().GetString("def-file")
 		purpose, _ := cmd.Flags().GetString("purpose")
@@ -107,6 +108,10 @@ var allocateCmd = &cobra.Command{
 
 		logger.Info("deploying definition", zap.Any("def", def))
 
+		if dryRun {
+			return
+		}
+
 		var deployer deployment.Deployer
 		if def.Deployer == "" {
 			deployer = helper.GetDefaultDeployer(ctx)
@@ -134,6 +139,7 @@ var allocateCmd = &cobra.Command{
 func init() {
 	rootCmd.AddCommand(allocateCmd)
 
+	allocateCmd.Flags().Bool("dry-run", false, "Disables the actual allocate and simply does a dry-run.")
 	allocateCmd.Flags().String("def", "", "The cluster definition you wish to provision.")
 	allocateCmd.Flags().String("def-file", "", "The path to a file containing a cluster definition to provision.")
 	allocateCmd.Flags().String("purpose", "", "The purpose for allocating this cluster")

cmd/chaos-allowtraffic.go

@@ -7,7 +7,7 @@ import (
 
 var chaosAllowTrafficCmd = &cobra.Command{
 	Use:   "allow-traffic",
-	Short: "Allows inter-node traffic to a specific node",
+	Short: "Allows all traffic to a specific node",
 	Args:  cobra.MinimumNArgs(2),
 	Run: func(cmd *cobra.Command, args []string) {
 		helper := CmdHelper{}

cmd/chaos-blocktraffic.go

@@ -1,13 +1,14 @@
 package cmd
 
 import (
+	"github.com/couchbaselabs/cbdinocluster/deployment"
 	"github.com/spf13/cobra"
 	"go.uber.org/zap"
 )
 
 var chaosBlockTrafficCmd = &cobra.Command{
-	Use:   "block-traffic",
-	Short: "Blocks inter-node traffic to a specific node",
+	Use:   "block-traffic [nodes/clients/all]",
+	Short: "Blocks a type of traffic to a specific node",
 	Args:  cobra.MinimumNArgs(2),
 	Run: func(cmd *cobra.Command, args []string) {
 		helper := CmdHelper{}
@@ -17,7 +18,22 @@ var chaosBlockTrafficCmd = &cobra.Command{
 		_, deployer, cluster := helper.IdentifyCluster(ctx, args[0])
 		node := helper.IdentifyNode(ctx, cluster, args[1])
 
-		err := deployer.BlockNodeTraffic(ctx, cluster.GetID(), node.GetID())
+		blockType := deployment.BlockNodeTrafficNodes
+		if len(args) > 2 {
+			switch args[2] {
+			case "nodes":
+				blockType = deployment.BlockNodeTrafficNodes
+			case "clients":
+				blockType = deployment.BlockNodeTrafficClients
+			case "all":
+				blockType = deployment.BlockNodeTrafficAll
+			default:
+				logger.Fatal("unexpected traffic type",
+					zap.String("type", args[2]))
+			}
+		}
+
+		err := deployer.BlockNodeTraffic(ctx, cluster.GetID(), node.GetID(), blockType)
 		if err != nil {
 			logger.Fatal("failed to block node traffic", zap.Error(err))
 		}

cmd/chaos-pausenode.go

@@ -0,0 +1,29 @@
+package cmd
+
+import (
+	"github.com/spf13/cobra"
+	"go.uber.org/zap"
+)
+
+var chaosPauseNodeCmd = &cobra.Command{
+	Use:   "pause-node",
+	Short: "Pauses a particular node in the cluster.",
+	Args:  cobra.MinimumNArgs(2),
+	Run: func(cmd *cobra.Command, args []string) {
+		helper := CmdHelper{}
+		logger := helper.GetLogger()
+		ctx := helper.GetContext()
+
+		_, deployer, cluster := helper.IdentifyCluster(ctx, args[0])
+		node := helper.IdentifyNode(ctx, cluster, args[1])
+
+		err := deployer.PauseNode(ctx, cluster.GetID(), node.GetID())
+		if err != nil {
+			logger.Fatal("failed to pause node", zap.Error(err))
+		}
+	},
+}
+
+func init() {
+	chaosCmd.AddCommand(chaosPauseNodeCmd)
+}

cmd/chaos-unpausenode.go

@@ -0,0 +1,29 @@
+package cmd
+
+import (
+	"github.com/spf13/cobra"
+	"go.uber.org/zap"
+)
+
+var chaosUnpauseNodeCmd = &cobra.Command{
+	Use:   "unpause-node",
+	Short: "Unpauses a particular node in the cluster.",
+	Args:  cobra.MinimumNArgs(2),
+	Run: func(cmd *cobra.Command, args []string) {
+		helper := CmdHelper{}
+		logger := helper.GetLogger()
+		ctx := helper.GetContext()
+
+		_, deployer, cluster := helper.IdentifyCluster(ctx, args[0])
+		node := helper.IdentifyNode(ctx, cluster, args[1])
+
+		err := deployer.UnpauseNode(ctx, cluster.GetID(), node.GetID())
+		if err != nil {
+			logger.Fatal("failed to resume node", zap.Error(err))
+		}
+	},
+}
+
+func init() {
+	chaosCmd.AddCommand(chaosUnpauseNodeCmd)
+}

cmd/init.go

@@ -470,7 +470,7 @@ var initCmd = &cobra.Command{
 						}
 					}
 					if dockerNetwork == "" {
-						dockerNetwork = "default"
+						dockerNetwork = "bridge"
 					}
 
 					dockerNetwork = readString(

cmd/ip.go

@@ -0,0 +1,33 @@
+package cmd
+
+import (
+	"fmt"
+
+	"github.com/couchbaselabs/cbdinocluster/deployment"
+	"github.com/spf13/cobra"
+)
+
+var ipCmd = &cobra.Command{
+	Use:   "ip [flags] cluster [node]",
+	Short: "Gets the IP of a node in the cluster",
+	Args:  cobra.MinimumNArgs(1),
+	Run: func(cmd *cobra.Command, args []string) {
+		helper := CmdHelper{}
+		ctx := helper.GetContext()
+
+		_, _, cluster := helper.IdentifyCluster(ctx, args[0])
+
+		var node deployment.ClusterNodeInfo
+		if len(args) >= 2 {
+			node = helper.IdentifyNode(ctx, cluster, args[1])
+		} else {
+			node = cluster.GetNodes()[0]
+		}
+
+		fmt.Printf("%s\n", node.GetIPAddress())
+	},
+}
+
+func init() {
+	rootCmd.AddCommand(ipCmd)
+}

cmd/root.go

@@ -9,13 +9,7 @@ import (
 // rootCmd represents the base command when called without any subcommands
 var rootCmd = &cobra.Command{
 	Use:   "cbdinocluster",
-	Short: "A brief description of your application",
-	Long: `A longer description that spans multiple lines and likely contains
-examples and usage of using your application. For example:
-
-Cobra is a CLI library for Go that empowers applications.
-This application is a tool to generate the needed files
-to quickly create a Cobra application.`,
+	Short: "provides tooling for quickly creating, modifying and destroying couchbase clusters.",
 }
 
 func Execute() {

deployment/caodeploy/deployer.go

@@ -48,13 +48,23 @@ func NewDeployer(opts *NewDeployerOptions) (*Deployer, error) {
 	}, nil
 }
 
+// PauseNode implements deployment.Deployer.
+func (*Deployer) PauseNode(ctx context.Context, clusterID string, nodeID string) error {
+	return errors.New("caodeploy does not support pausing node")
+}
+
+// UnpauseNode implements deployment.Deployer.
+func (*Deployer) UnpauseNode(ctx context.Context, clusterID string, nodeID string) error {
+	return errors.New("caodeploy does not support unpausing node")
+}
+
 // AllowNodeTraffic implements deployment.Deployer.
 func (*Deployer) AllowNodeTraffic(ctx context.Context, clusterID string, nodeID string) error {
 	return errors.New("caodeploy does not support allowing node traffic")
 }
 
 // BlockNodeTraffic implements deployment.Deployer.
-func (*Deployer) BlockNodeTraffic(ctx context.Context, clusterID string, nodeID string) error {
+func (*Deployer) BlockNodeTraffic(ctx context.Context, clusterID string, nodeID string, blockType deployment.BlockNodeTrafficType) error {
 	return errors.New("caodeploy does not support blocking node traffic")
 }
 

deployment/clouddeploy/deployer.go

@@ -1141,7 +1141,7 @@ func (d *Deployer) DeleteCollection(ctx context.Context, clusterID string, bucke
 	return errors.New("clouddeploy does not support deleting collections")
 }
 
-func (d *Deployer) BlockNodeTraffic(ctx context.Context, clusterID string, nodeID string) error {
+func (d *Deployer) BlockNodeTraffic(ctx context.Context, clusterID string, nodeID string, blockType deployment.BlockNodeTrafficType) error {
 	return errors.New("clouddeploy does not support traffic control")
 }
 
@@ -1160,3 +1160,11 @@ func (d *Deployer) ListImages(ctx context.Context) ([]deployment.Image, error) {
 func (d *Deployer) SearchImages(ctx context.Context, version string) ([]deployment.Image, error) {
 	return nil, errors.New("clouddeploy does not support image search")
 }
+
+func (d *Deployer) PauseNode(ctx context.Context, clusterID string, nodeID string) error {
+	return errors.New("clouddeploy does not support node pausing")
+}
+
+func (d *Deployer) UnpauseNode(ctx context.Context, clusterID string, nodeID string) error {
+	return errors.New("clouddeploy does not support node pausing")
+}

deployment/deployer.go

@@ -64,6 +64,14 @@ type Image struct {
 	Name   string
 }
 
+type BlockNodeTrafficType string
+
+const (
+	BlockNodeTrafficClients BlockNodeTrafficType = "clients"
+	BlockNodeTrafficNodes   BlockNodeTrafficType = "nodes"
+	BlockNodeTrafficAll     BlockNodeTrafficType = "all"
+)
+
 type Deployer interface {
 	ListClusters(ctx context.Context) ([]ClusterInfo, error)
 	NewCluster(ctx context.Context, def *clusterdef.Cluster) (ClusterInfo, error)
@@ -87,9 +95,11 @@ type Deployer interface {
 	CreateCollection(ctx context.Context, clusterID string, bucketName, scopeName, collectionName string) error
 	DeleteScope(ctx context.Context, clusterID string, bucketName, scopeName string) error
 	DeleteCollection(ctx context.Context, clusterID string, bucketName, scopeName, collectionName string) error
-	BlockNodeTraffic(ctx context.Context, clusterID string, nodeID string) error
+	BlockNodeTraffic(ctx context.Context, clusterID string, nodeID string, blockType BlockNodeTrafficType) error
 	AllowNodeTraffic(ctx context.Context, clusterID string, nodeID string) error
 	CollectLogs(ctx context.Context, clusterID string, destPath string) ([]string, error)
 	ListImages(ctx context.Context) ([]Image, error)
 	SearchImages(ctx context.Context, version string) ([]Image, error)
+	PauseNode(ctx context.Context, clusterID string, nodeID string) error
+	UnpauseNode(ctx context.Context, clusterID string, nodeID string) error
 }