Kubelet Service is Unstable due to Invalid vm/Overcommit_memory Kernel Flag

Problem

• kubelet service is unstable showing the below error in the log:

xxxxxxxxxx

Failed to start ContainerManager invalid kernel flag: vm/overcommit_memory, expected value: 1, actual value: 0

Environment

• Platform9 Edge Cloud v-5.3.0

Cause

• In one of the master node, the vm.overcommit_memory parameter in the /etc/sysctl.conf file set to "0"

xxxxxxxxxx

#sysctl vm.overcommit_memory

vm.overcommit_memory = 0

Resolution

• Replace the value of above parameter in the /etc/sysctl.conffile with 1

xxxxxxxxxx

# overcommit memory

vm.overcommit_memory=0 >>>>>>>>>>>>>> Replace this with 1

• Run below command to persist the change made above:

xxxxxxxxxx

# sysctl -p

• After making the change, confirm if the value is reflected for the parameter:

xxxxxxxxxx

# sysctl vm.overcommit_memory

• Check pf9-kubelet.service status, if it is still stuck in failed state, restart once and confirm if it becomes healthy.

xxxxxxxxxx

# systemctl status pf9-kubelet

# systemctl restart pf9-kubelet