Search code examples
linuxsystemctl

How to check how many times linux service restarted

Is there a way to check how many times linux service restarted in specific time range? Currently using systemctl status nameOfService | grep -Po ".*; \K(.*)(?= ago)" to extract uptime, but it's not enough for diagnostics.

Solution

  • If you have logs from that period, you can indeed extract the "Stopped"/"Started" messages from the logs. On nowadays system with systemd, many systems use journalctl, but many systems use something else - you'll have to inspect your specific system on how to query the logs.

    Your script could look something along:

     journalctl --since=<date> --until=<another data> UNIT=<service name> SYSLOG_IDENTIFIER=systemd |
 awk '
      /Started/{ started = 1 }
      /Stopped/{ if (started) { started = 0; restarted++; } }
      END {
          print "It was restarted " restarted "times."
      }
 '

    The regex here are just an example. Better use journalctl -o json and parse the output with jq.