| Hello,
the problem with FQDN is, that getting it often requires a working DNS - if the DNS or network is down, monit start/reload is then slow as it tries the FQDN lookup will timeout. If monit is responsible for starting the network interface and/or DNS, it'll be problem on each start (monit will pause on FQDN lookup before starting services, making the startup time longer).
There were few issues related to it:
The $HOST now expands to the hostname only and doesn't block - it is possible to override the hostname using "check system <name>".
Best regards, Martin
Hello, I've upgraded from monit-5.17.1 to monit 5.20.0, but now the config line: check system $HOST
is expanded only to "hostname -s", instead of "hostname -f" as before. This can be seen by the logfile on the server "mmonit": Nov 23 17:13:45 mmonit monit[7402]: 'mmonit.as-infodienste.de' Monit 5.17.1 stopped
Nov 23 17:13:52 mmonit monit[31741]: /usr/local/monit.d/mmonit.monit:8: Service name conflict, mmonit already defined '/run/mmonit/mmonit.pid'
Nov 23 17:13:52 mmonit /etc/init.d/monit[31740]: start-stop-daemon: failed to start `/usr/bin/monit'
Nov 23 17:13:52 mmonit /etc/init.d/monit[31724]: ERROR: monit failed to start
Nov 23 17:22:47 mmonit monit[3567]: Starting Monit 5.20.0 daemon with http interface at [*]:10443
Nov 23 17:22:47 mmonit monit[3567]: Monit start delay set to 15s
Nov 23 17:23:02 mmonit monit[3570]: 'mmonit' Monit 5.20.0 started
In this case it was especially bad, because I have defined a service "mmonit" that now conflicts with the hostname "mmonit". This also influences everything else: The mmonit Dashboard, the monit mails... How can I get the old behaviour back? Best regards, Marcus Mülbüsch -- To unsubscribe: https://lists.nongnu.org/mailman/listinfo/monit-general
|