Description of problem: atomic-openshift-node.service logs show the following message: ~~~ Failed to get system container stats for "/system.slice/atomic-openshift-node.service": failed to get cgroup stats for "/system.slice/atomic-openshift-node.service": failed to get container info for "/system.slice/atomic-openshift-node.service": unknown container "/system.slice/atomic-openshift-node.service" container_manager_linux.go:793] CPUAccounting not enabled for pid: xxxx container_manager_linux.go:796] MemoryAccounting not enabled for pid: xxxx ~~~ Version-Release number of selected component (if applicable): ~~~ oc v3.11.154 kubernetes v1.11.0+d4cacc0 features: Basic-Auth GSSAPI Kerberos SPNEGO Server https://lb.guachen.ocp:443 openshift v3.11.154 kubernetes v1.11.0+d4cacc0 ~~~ How reproducible: Always Steps to Reproduce: 1. install cluster 2. ssh to node 3. journalctl -u atomic-openshift-node.service Actual results: CPU/Memory Accounting is not enabled by default Expected results: CPU/Memory Accounting is enabled by default Additional info: ~~~ # cat /etc/systemd/system.conf.d/origin-accounting.conf [Manager] DefaultCPUAccounting=yes DefaultMemoryAccounting=yes DefaultBlockIOAccounting=yes # systemctl show docker|grep Account CPUAccounting=no BlockIOAccounting=no MemoryAccounting=no TasksAccounting=no ~~~ CPU/Memory Accounting should be enabled by default.
Did you restart docker, and then restart Kubelet? Both restarts are necessary to enable accounting.
Yes, I tried both restart docker and atom-openshift-node, but it didn't work. I found that the system had to be rebooted to take effect.
For systemd to detect the new configuration file the following command will pick that up: `systemctl daemon-reexec`.