Troubleshooting time synchronization on Managed Service for Kubernetes cluster nodes
Issue description
System time displayed incorrectly in the Linux console, as well as in container and Managed Service for Kubernetes cluster pod logs.
Solution
You can specify your own NTP server in the subnet settings using the YC CLI or with Terraform. This will configure the cloud DHCP server to include your NTP server information and propagate it to all nodes in the cloud subnet.
Note
This operation does not automatically delete or recreate nodes in your Managed Service for Kubernetes cluster. To apply these NTP settings within the Managed Service for Kubernetes cluster itself, use one of these two options:
- Renew the DHCP configuration on each node by running
sudo dhclient -v -r && sudo dhclient. - Reboot the cluster's worker nodes at a time convenient for you.
Either option may cause brief (up to a few minutes) network unavailability for services within the cluster while the nodes re-acquire their network configuration.
If the issue persists
If the above actions did not help, create a support ticket. Provide the following information in your ticket:
- Managed Service for Kubernetes cluster ID.
- Managed Service for Kubernetes cluster event log:
kubectl get eventsoutput.