health-checker, health-checker.service — Checks that all important services of a system did come up correctly after an update and reboot.
health-checker checks during boot, if the system is coming up correct. In error case, the action depends on what did happen before. If this is the first boot after a transactional update, an automatic rollback to the last known working snapshot is made. If the snapshot was already rebooted successfull before, a reboot is tried. If this does not help, some sevices are shutdown and an admin has to repair the system.
If the boot was successfull, the current snapshot is marked as known to be working.
Plugins are used for performing the tests and new plugins can be
written. They are stored in the /usr/libexec/health-checker
or /usr/local/libexec/health-checker (individual user plugins) directory
and are executed either with the option check or
stop.
If called with "check", the plugin should verify that the service works and
return 0, else 1. If one plugin returns "1", health-check
will evaluate what to do. If neither a rollback or reboot is usefull, all
plugins will be called with the "stop" option. In this case, the plugins
should shutdown the services are far as possible and usefull.
If telem-record-gen(1) is installed telemetry events are created additional to logging with syslog(3).