Executing remote commands with SSH

Run commands on a remote server without opening an interactive shell. Syntax: ssh [options] user@hostname "command"

Basic syntax

ssh user@hostname "command"
ssh user@hostname command  # Command is optional

Return to local shell immediately after execution.

Simple examples

ssh user@hostname "ls -la /tmp"
ssh user@hostname "pwd"
ssh user@hostname "whoami"

Multiple commands

Use semicolons or && (depends on success):

ssh user@hostname "cd /tmp && ls -la"
ssh user@hostname "uptime; df -h"

Or with newlines in a multi-line string:

ssh user@hostname << 'EOF'
cd /tmp
ls -la
whoami
EOF

Capture output locally

ssh user@hostname "ls -la /tmp" > local-file.txt
ssh user@hostname "cat /var/log/syslog" | grep "error"

Useful patterns

Check if service is running:

ssh user@hostname "systemctl is-active nginx"

Run with custom port and key:

ssh -i ~/.ssh/id_ed25519 -p 2222 user@hostname "df -h"

Execute script on remote:

ssh user@hostname < local-script.sh

Run with sudo (prompts for password):

ssh user@hostname "sudo systemctl restart nginx"

Background task (agent/daemon):

ssh user@hostname "nohup long-running-command > output.log 2>&1 &"

Important notes

Command runs non-interactively — no interactive prompts
No shell history recorded on remote
Input/output fully controlled by local shell
Permissions depend on the ssh user
Environment variables may differ from interactive shell

SSH key authentication

# With specific key
ssh -i ~/.ssh/custom_key user@hostname "ls"

# Uses default key (~/.ssh/id_rsa or id_ed25519) if `-i` omitted
ssh user@hostname "ls"

Timeout and error handling

# Timeout after 5 seconds
ssh -o ConnectTimeout=5 user@hostname "sleep 10"

# Fail if command exits with error (useful in scripts)
ssh user@hostname "false" && echo "Success" || echo "Failed"