iptables-restore high cpu usage in OpenShift
Issue
OpenShift nodes have a very high load average and top
output reports the iptables-restore
process is using most CPU. For example:
# top
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
99226 root 20 0 62180 18080 976 R 100.0 0.0 0:05.58 iptables-restor
121061 root 20 0 6014480 520156 38272 S 41.5 0.1 18136:02 openshift
103066 root 20 0 38.003g 1.948g 17444 S 19.8 0.4 14173:02 dockerd-current
89157 root 20 0 2674184 58408 3480 S 15.2 0.0 3435:29 rhel-push-plugi
...
Environment
- Red Hat OpenShift Container Platform 3.5 and earlier
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.