Hi,
Seeing some errors on metricbeat v6.4.2, on collection and sending of CPU metrics.
My configuration is largely the same as what is provided here, other than swapping out appropriate environment config. https://www.elastic.co/guide/en/beats/metricbeat/current/running-on-kubernetes.html
I'm seeing this across many metricbeat instances, but here are kibana (so please read in opposite order) logs for one specifically.
|2018-11-04T01:17:24.776Z|ERROR|elasticsearch/client.go:356|Failed to encode event: unsupported float value: NaN|
|---|---|---|---|
|2018-11-04T01:17:24.749Z|ERROR|elasticsearch/client.go:356|Failed to encode event: unsupported float value: NaN|
|2018-11-04T01:17:23.748Z|ERROR|cpu/helper.go:118|Error calculating CPU time change for docker module: new stats value (0) is lower than the old one(705949450000000)|
|2018-11-04T01:17:23.748Z|ERROR|cpu/helper.go:118|Error calculating CPU time change for docker module: new stats value (0) is lower than the old one(9256415)|
|2018-11-04T01:17:23.747Z|ERROR|cpu/helper.go:118|Error calculating CPU time change for docker module: new stats value (0) is lower than the old one(895185196301)|
|2018-11-04T01:17:23.747Z|ERROR|cpu/helper.go:118|Error calculating CPU time change for docker module: new stats value (0) is lower than the old one(562900000000)|
|2018-11-04T01:17:23.747Z|ERROR|cpu/helper.go:118|Error calculating CPU time change for docker module: new stats value (0) is lower than the old one(705949450000000)|
|2018-11-04T01:17:23.747Z|ERROR|cpu/helper.go:118|Error calculating CPU time change for docker module: new stats value (0) is lower than the old one(241520000000)|
Best,
Justin