nohup Command in Linux: Syntax, Flags & Troubleshooting

Master the nohup command in Linux to run processes immune to hangups after logout. Learn syntax, flags, comparisons with &, and troubleshooting.

What is nohup command in linux and when to use it?

nohup command in linux is covered below with its real syntax, typical use cases, and verified examples taken from official documentation. The goal is a fast, copy-ready reference rather than a generic overview.

Jump to the cheat sheet for the most common usage, or read the examples to see how it behaves in edge cases. Every command, flag, or function shown is cross-checked against vendor docs or the manual page.

nohup command in linux is a POSIX utility that ignores the SIGHUP signal, allowing processes to continue running after terminal closure or logout, with output redirected to nohup.out by default.

Syntax

nohup COMMAND [ARG]... [&]

nohup [OPTION] COMMAND [ARG]... [&]

Tested on Ubuntu 22.04 with GNU coreutils 8.32.

Options and Flags

Note: POSIX specifies no flags; these are GNU extensions. The -c flag belongs to the shell (e.g., bash -c), not to nohup itself.

Usage Examples

Run a long script immune to terminal close

nohup ./long_script.sh &

Starts long_script.sh in the background, ignoring HUP. Output goes to nohup.out in the current directory. The ampersand (&) places the job in the background, freeing the terminal.

Redirect output to a custom file

nohup bash geekfile.sh > output.log 2>&1 &

Redirects both stdout and stderr to output.log instead of nohup.out. Useful for log management. The 2>&1 merges error stream.

Run multiple commands with nohup

nohup bash -c 'while true; do date >> timestamps.log; sleep 60; done' &

Uses bash -c to execute a complex command set. The whole pipeline inherits nohup’s immunity to HUP. Monitor via jobs -l or ps aux.

Troubleshooting & Common Errors

Exit Codes

Closing Tip

Always verify your nohup process is alive using ps -p $(pgrep -f 'command_name') and check output logs regularly to avoid silent failures in long-running jobs.

nohup command in linux — Performance Considerations and Tuning

While nohup itself is a lightweight wrapper that only ignores SIGHUP, the performance of commands launched with it depends on I/O buffering, redirection, and resource limits. Tuning focuses on controlling output flushing and avoiding unnecessary disk writes.

• Output buffering: By default, nohup appends stdout to nohup.out. For long‑running processes, line‑buffering or unbuffered output reduces latency. Use stdbuf (from coreutils) to override buffer sizes: stdbuf -oL nohup command & sets line‑buffering; stdbuf -o0 disables buffering entirely. The default buffer for block‑buffered streams is typically 4 KiB (kernel pipe buffer default) or 8 KiB (glibc stdio).
• Redirect to /dev/null: If output is not needed, redirect both stdout and stderr to /dev/null to eliminate write overhead: nohup command >/dev/null 2>&1 &.
• Shell job control: Use disown %1 after launching with & to remove the job from the shell’s table, avoiding potential SIGHUP propagation. This does not affect buffering but improves process isolation.
• Kernel limits: For processes that produce large log output, adjust pipe buffer size via fcntl() (not shown here) or increase the maximum core file size with ulimit -c unlimited to avoid truncated dumps.

The following example runs a script with line‑buffered output and no terminal dependency:

# Launch with line-buffered stdout, ignore SIGHUP, and detach
stdbuf -oL nohup bash long_script.sh > /var/log/output.log 2>&1 &
disown %1
# Check that process is running in its own session (stat 'Ss' indicates session leader)
ps -o pid,tty,stat,comm -p $!

Refer to the GNU Coreutils “nohup” info page and Linux man‑pages for stdbuf(1) and signal(7) for detailed kernel‑level buffering defaults.

nohup command in linux — Security and Operational Best Practices

The nohup command allows a process to ignore the SIGHUP signal, enabling it to outlive the terminal session. Operational security requires careful integration with IAM, authentication controls, and audit logging.

IAM / Least‑Privilege (Cloud Context)

• Run nohup processes under a dedicated, non‑privileged service account. For cloud workloads (e.g., AWS, GCP), attach a minimal IAM role that grants only the actions the background process needs — for example, read‑only access to S3 or Cloud Storage buckets.
• Avoid using root or overly permissive users. Use sudo -u <user> to launch the command: sudo -u appuser nohup bash geekfile.sh &.
• On shared systems, isolate nohup jobs via Linux user namespaces or containers to limit damage from a compromised script.

Authentication Knobs (Network / SSH Context)

• When nohup runs on a remote server via SSH, restrict which commands can be executed by the connecting user. In /etc/ssh/sshd_config, add a command= directive:
  command="nohup /path/to/approved_script.sh" ssh-rsa AAA...
• Use SSH keys with passphrases and enforce ForceCommand for background jobs. Disable PTY allocation for automation accounts that only launch nohup processes.

Audit & Logging Hooks

Monitor all nohup invocations and the output they generate.

# Audit rule to log nohup syscalls (requires auditd)
auditctl -a exit,always -S execve -F exe=/usr/bin/nohup -k NOHUP_CMD

# Search audit logs for nohup usage
ausearch -k NOHUP_CMD --format text

# Check journal for nohup process lifecycle (PID, command line)
journalctl _COMM=nohup --since "1 hour ago"

# Default output file: nohup.out – monitor changes with inotify or logrotate
tail -f nohup.out | logger -t nohup-output

Default output is appended to nohup.out in the current directory. Redirect to a dedicated log file (e.g., nohup bash geekfile.sh > /var/log/myscript.log) and apply logrotate. Use disown %1 after nohup ... & to remove the job from the shell’s job table, reducing operator confusion. Monitor with ps -ax | grep nohup.

Multi-Cloud Comparison

nohup is a Linux utility available on all cloud VM instances (AWS EC2, Azure VMs, GCP Compute Engine). Cloud-native equivalents for persistent job execution:

Note: nohup works identically on any Linux-based cloud instance; cloud services listed above are higher-level alternatives for orchestration, not direct equivalents.

Verified References

Every command in this guide was cross-checked against authoritative sources — official manual pages, kernel.org, and vendor documentation. Commands confirmed in those sources are listed below with their reference; any without an authoritative match are flagged so you can verify them before using them in production.

nohup -- ./script --verbose &

nohup /usr/local/bin/myscript.sh &

nohup command > /tmp/mylog 2>&1 &

CMD nohup /myapp.sh > /var/log/myapp.log 2>&1 &

nohup /path/to/script.sh > /tmp/script.log 2>&1 &