I’m using node exporter with systemd enabled and measure this metrics:
(time() - (node_systemd_unit_start_time_seconds))
I’m trying to setup alerts on service restart or inactive, so I’m using service unit uptime value and create alert with condition OUTSIDE RANGE:
Initially some systemd services already inactive so I get allert about it. But when I stop or restart working service and it’s uptime relatively become above 60seconds or below 1562655700 grafana don’t send neither ok alert nor warning alert.
Is there way to get around this.
PS: When I testing rule - the values of service I stop/restart is outside range