NTP alerts coinciding with reboots of ppc64le host
Multiple alerts for qa-power8-5-kvm as visible on
where the alerts seem to coincide with reboots that I triggered as an experiment. It seems that the NTP offset is exceeding a threshold as there are many values missing due to reboot. Likely we can live with a longer pending time to prevent this problem.
- AC1: NTP offset alert does not trigger on repeated reboots
- File ntp_offset_fine_within_12m.png ntp_offset_fine_within_12m.png added
- Status changed from In Progress to Resolved
shows how the machine qa-power8-5-kvm is offline for 9 minutes and returns with a nearly 1s clock offset but returning to a sane level after in total 12m so the new alert threshold of 20m is not exceeded.