Graceful Pod Shutdown

This means that if the pod’s main container attempts to make any new network calls after the proxy has received the TERM signal, those network calls will fail. This also has implications for clients of the terminating pod and for job resources.

To mitigate this, use the flag with to delay the Linkerd proxy’s handling of the TERM signal for a given number of seconds using a hook. This delay gives slow clients additional time to receive the endpoints update before beginning graceful shutdown. To achieve max benefit from the option, the main container should have its own hook with the sleep command inside which has a smaller period than is set for the proxy sidecar. And none of them must be bigger than configured for the entire pod.

Job Resources

Pods which are part of a job resource run until all of the containers in the pod complete. However, the Linkerd proxy container runs continuously until it receives a TERM signal. This means that job pods which have been injected will continue to run, even once the main container has completed.