Check Process Running Shell Script

Monitoring tools and cron jobs often need a single question: is mysqld running? This script wraps pgrep with clear logging and exit codes so failures propagate to your alert pipeline.

What this script does

Accepts process name or pattern as argument
Uses pgrep -x for exact match when possible
Prints matching PIDs on success
Exits 1 when no process found (monitoring-friendly)
Optional restart hook placeholder for advanced use

Prerequisites

procps package provides pgrep
Run as user that can see target processes
Name matches the real process (nginx vs nginx: master)

Step 1: Save the script

sudo nano /usr/local/bin/process-check.sh
sudo chmod +x /usr/local/bin/process-check.sh

Step 2: Full script (scroll to read)

process-check.sh

#!/usr/bin/env bash
set -euo pipefail

NAME="${1:?usage: process-check.sh PROCESS_NAME}"
LOG="/var/log/process-check.log"

log(){ echo "[$(date '+%F %T')] $*" | tee -a "$LOG"; }

if pgrep -x "$NAME" >/dev/null 2>&1; then
  pids=$(pgrep -x "$NAME" | tr '\n' ' ')
  log "OK: $NAME running (PIDs: $pids)"
  exit 0
fi

# fallback: partial match (e.g. mysqld vs mariadbd)
if pgrep -f "$NAME" >/dev/null 2>&1; then
  pids=$(pgrep -f "$NAME" | head -5 | tr '\n' ' ')
  log "OK: pattern '$NAME' matched (PIDs: $pids)"
  exit 0
fi

log "FAIL: $NAME not running"
exit 1

Scroll inside the box to read the full script.

Step 3: Configure settings

Different from service watchdog script — this is a lightweight probe
Pair with email alert wrapper
Prefer systemctl is-active for systemd units when applicable

Shell script checking if a process is running with pgrep

Step 4: Test manually

./process-check.sh sshd

./process-check.sh nonexistent-daemon; echo exit=$?

Schedule with cron

sudo crontab -e

Add:

*/2 * * * * /usr/local/bin/process-check.sh mysqld || /usr/local/bin/alert-on-fail.sh /usr/local/bin/process-check.sh mysqld

Check if Process Is Running Shell Script (pgrep, Nagios, Cron)