cephrecover

#!/bin/bash
#Creator: nathan.loosevelt@grandblanctwpmi.gov
#Date Created: 9/17/2023
#This script is made to remove a ceph node from maintenance mode after a reboot and to autostart any services that often fail to start.

Red='\033[1;31m'
BBlue='\033[1;34m'
Coloroff='\033[0m'


  echo ""
  echo -e "${BBlue}This script will verify/start critical services and remove the node from maintenance mode after a reboot.${Coloroff}"
#Check if nmbd.service is running
  echo "" 
  echo "Checking if critical services are running. If not, the script will attempt to start them..."
if systemctl is-active --quiet nmbd.service; then
  echo "The nmbd.service is already running."
else
  # Start nmbd.service if it's not running
  echo "nmbd.service is not running. Starting it now..."
  sudo systemctl start nmbd.service

  # Check the status again to confirm if it started successfully
  if systemctl is-active --quiet nmbd.service; then
    echo "nmbd.service has been successfully started."
  else
    echo "Failed to start nmbd.service. Check the service status and logs for more information."
  fi
fi

# Run Ceph commands to unset maintenance flags commands
echo ""
ceph osd unset noout
ceph osd unset norecover
ceph osd unset norebalance

# Prompt employee to verify cluster returns to a healthy state. Run command to open an update ceph status window.
echo ""
echo -e "${Red}Verify the cluster returns to a healthy state by running:  ${Coloroff}watch -n1 ceph -s ${Red}. This command is live, to exit the command hit Ctrl+C.${Coloroff}"