使用
journalctl -f -u kubelet
一直报警告
kubelet[192184]: W0317 10:59:11.494878 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:28.485263 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:28.486748 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:31.467736 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:32.399734 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:41.492686 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:41.492732 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:41.492813 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:41.493864 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:44.007218 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:56.669451 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
kubelet[192184]: W0317 10:59:58.495169 192184 conversion.go:110] Could not get instant cpu stats: cumulative stats decrease
该如何解决?
这个错误属于kubelet 内置的cadvisor,cadvisor 统计和监控节点的资源使用情况。kubelet试图从cadvisor获取cpu指标。
一些版本问题,会导致资源统计错误,引起这个错误:
1、升级kubelet
2、重新启动
systemctl restart kubelet
你的答案