FIX: Use OpenSSH and prevent false-positive process detection

go.mod

@@ -8,7 +8,6 @@ require (
 	github.com/spf13/cobra v1.9.1
 	github.com/spf13/viper v1.21.0
 	go.yaml.in/yaml/v3 v3.0.4
-	golang.org/x/crypto v0.50.0
 	golang.org/x/term v0.42.0
 )
 

go.sum

@@ -49,8 +49,6 @@ github.com/subosito/gotenv v1.6.0 h1:9NlTDc1FTs4qu0DDq7AEtTPNw6SVm7uBMsUCUjABIf8
 github.com/subosito/gotenv v1.6.0/go.mod h1:Dk4QP5c2W3ibzajGcXpNraDfq2IrhjMIvMSWPKKo0FU=
 go.yaml.in/yaml/v3 v3.0.4 h1:tfq32ie2Jv2UxXFdLJdh3jXuOzWiL1fo0bu/FbuKpbc=
 go.yaml.in/yaml/v3 v3.0.4/go.mod h1:DhzuOOF2ATzADvBadXxruRBLzYTpT36CKvDb3+aBEFg=
-golang.org/x/crypto v0.50.0 h1:zO47/JPrL6vsNkINmLoo/PH1gcxpls50DNogFvB5ZGI=
-golang.org/x/crypto v0.50.0/go.mod h1:3muZ7vA7PBCE6xgPX7nkzzjiUq87kRItoJQM1Yo8S+Q=
 golang.org/x/sys v0.43.0 h1:Rlag2XtaFTxp19wS8MXlJwTvoh8ArU6ezoyFsMyCTNI=
 golang.org/x/sys v0.43.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/term v0.42.0 h1:UiKe+zDFmJobeJ5ggPwOshJIVt6/Ft0rcfrXZDLWAWY=

internal/cluster/delete.go

@@ -83,7 +83,7 @@ func verifyAllStopped(
 ) error {
 	for _, server := range topo.Servers {
 		pidFile := pidFilePath(server.Address, topo.Path, version)
-		cmd := fmt.Sprintf("pgrep -f %q > /dev/null 2>&1", pidFile)
+		cmd := processRunningCommand(pidFile)
 		if err := ssh.Run(server.Host(), cmd); err == nil {
 			return fmt.Errorf(
 				"cache server %q is still running; stop the cluster before delete",

internal/cluster/server.go

@@ -3,6 +3,7 @@ package cluster
 import (
 	"fmt"
 	"path"
+	"regexp"
 	"strings"
 
 	"github.com/jam2in/arcusctl/internal/topology"
@@ -21,6 +22,11 @@ func pidFilePath(
 	return path.Join(installPath, fmt.Sprintf("memcached-%s.pid", listenPort(serverAddress)))
 }
 
+func processRunningCommand(pidFile string) string {
+	pattern := fmt.Sprintf("[m]emcached.*-P %s", regexp.QuoteMeta(pidFile))
+	return fmt.Sprintf("pgrep -f %q > /dev/null 2>&1", pattern)
+}
+
 func listenPort(address string) string {
 	parts := strings.SplitN(address, ":", 2)
 	if len(parts) < 2 {

internal/cluster/status.go

@@ -57,7 +57,7 @@ func processStatus(
 	version string,
 ) string {
 	pidFile := pidFilePath(server.Address, topoPath, version)
-	cmd := fmt.Sprintf("pgrep -f %q > /dev/null 2>&1", pidFile)
+	cmd := processRunningCommand(pidFile)
 	if err := ssh.Run(server.Host(), cmd); err == nil {
 		return "running"
 	}

internal/ssh/ssh.go

@@ -5,114 +5,39 @@ import (
 	"fmt"
 	"os"
 	"os/exec"
-	"os/user"
-	"path/filepath"
-
-	gossh "golang.org/x/crypto/ssh"
 )
 
-func currentUser() (string, error) {
-	u, err := user.Current()
-	if err != nil {
-		return "", err
-	}
-
-	return u.Username, nil
-}
-
-func newClient(host string) (*gossh.Client, error) {
-	username, err := currentUser()
-	if err != nil {
-		return nil, err
-	}
-
-	homeDir, err := os.UserHomeDir()
-	if err != nil {
-		return nil, err
-	}
-
-	keyPath := filepath.Join(homeDir, ".ssh", "id_rsa")
-	key, err := os.ReadFile(keyPath)
-	if err != nil {
-		return nil, err
-	}
-
-	signer, err := gossh.ParsePrivateKey(key)
-	if err != nil {
-		return nil, err
-	}
-
-	config := &gossh.ClientConfig{
-		User: username,
-		Auth: []gossh.AuthMethod{
-			gossh.PublicKeys(signer),
-		},
-		HostKeyCallback: gossh.InsecureIgnoreHostKey(),
-	}
-
-	return gossh.Dial("tcp", host+":22", config)
+func withOptions(args ...string) []string {
+	return append([]string{
+		"-o", "BatchMode=yes",
+		"-o", "ConnectTimeout=10",
+	}, args...)
 }
 
 func Run(host string, command string) error {
-	client, err := newClient(host)
-	if err != nil {
-		return err
-	}
-	defer client.Close()
-
-	session, err := client.NewSession()
-	if err != nil {
-		return err
-	}
-	defer session.Close()
-
-	session.Stdout = os.Stdout
-	session.Stderr = os.Stderr
-	return session.Run(command)
+	cmd := exec.Command("ssh", withOptions(host, command)...)
+	cmd.Stdout = os.Stdout
+	cmd.Stderr = os.Stderr
+	return cmd.Run()
 }
 
 func Copy(localPath string, host string, remotePath string) error {
-	username, err := currentUser()
-	if err != nil {
-		return err
-	}
-
-	homeDir, err := os.UserHomeDir()
-	if err != nil {
-		return err
-	}
-
-	dest := fmt.Sprintf("%s@%s:%s", username, host, remotePath)
-	cmd := exec.Command("scp",
-		"-i", filepath.Join(homeDir, ".ssh", "id_rsa"),
-		"-o", "StrictHostKeyChecking=no",
-		localPath,
-		dest)
+	dest := fmt.Sprintf("%s:%s", host, remotePath)
+	cmd := exec.Command("scp", withOptions(localPath, dest)...)
 	cmd.Stdout = os.Stdout
 	cmd.Stderr = os.Stderr
 	return cmd.Run()
 }
 
 func FileExists(host string, remotePath string) (bool, error) {
-	client, err := newClient(host)
-	if err != nil {
-		return false, err
-	}
-	defer client.Close()
-
-	session, err := client.NewSession()
-	if err != nil {
-		return false, err
-	}
-	defer session.Close()
-
-	err = session.Run(fmt.Sprintf("test -f %s", remotePath))
+	cmd := exec.Command("ssh", withOptions(host, fmt.Sprintf("test -f %s", remotePath))...)
+	err := cmd.Run()
 	if err == nil {
 		return true, nil
 	}
 
-	var exitErr *gossh.ExitError
-	if errors.As(err, &exitErr) && exitErr.ExitStatus() == 1 {
+	var exitErr *exec.ExitError
+	if errors.As(err, &exitErr) && exitErr.ExitCode() == 1 {
 		return false, nil
 	}
 

internal/zk/delete.go

@@ -59,7 +59,7 @@ func verifyTopology(servers []topology.ZKServer, topoPath string) error {
 func verifyAllStopped(servers []topology.ZKServer, topoPath string) error {
 	for _, server := range servers {
 		confPath := zkConfigPath(topoPath, server.MyID)
-		cmd := fmt.Sprintf("pgrep -f 'QuorumPeerMain.*%s' > /dev/null 2>&1", confPath)
+		cmd := fmt.Sprintf("pgrep -f '[Q]uorumPeerMain.*%s' > /dev/null 2>&1", confPath)
 		if err := ssh.Run(server.Host(), cmd); err == nil {
 			return fmt.Errorf("server %s (myid=%d) is still running. stop the ensemble before delete",
 				server.Host(), server.MyID)