How to understand error in kubelet logs in the worker nodes?

Question

I see these errors in the kubelet.log on my worker nodes

Feb 11 19:39:41 my-node-ip kubelet[4358]: I0711 19:39:41.666680    4358 log.go:198] http: TLS handshake error from 172.16.4.71:58965: EOF
Feb 11 19:39:42 my-node-ip kubelet[4358]: I0711 19:39:42.121386    4358 log.go:198] http: TLS handshake error from 172.16.4.40:21053: EOF
Feb 11 19:39:45 my-node-ip kubelet[4358]: I0711 19:39:45.001122    4358 log.go:198] http: TLS handshake error from 172.16.4.71:36777: EOF
Feb 11 19:39:45 my-node-ip kubelet[4358]: I0711 19:39:45.455301    4358 log.go:198] http: TLS handshake error from 172.16.4.40:31905: EOF
Feb 11 19:39:48 my-node-ip kubelet[4358]: I0711 19:39:48.333620    4358 log.go:198] http: TLS handshake error from 172.16.4.71:1877: EOF

• What does the error "http: TLS handshake error from 172.16.4.71:58965: EOF" indicate?
• What could be these ips 172.16.4.71 and 172.16.4.40 and port correspond to?

What should I understand by this error message?

Answer 1

As mentioned in this document:

Whenever an SSL/TLS Handshake fails, it’s mostly due to certain things going on with the server, website, and the configuration of its installed SSL/TLS. Presently the culprit is TLS configuration as support for SSL 3.0 is deprecated. However, there’s a distinct possibility that a client-side error can be the reason behind the SSL/TLS Handshake Failed error. And, some of the common ones are like incorrect system time or browser updates.

Also this is an known issue as mentioned in the github link

EOF errors seem to be related to a Go bug and appear on kubernetes 1.22, 1.23 and 1.24. This is not affecting any functional issues and these are generated from core kubernetes. We need to wait until a fix is suggested by kubernetes.

Can you also try this work around to disable the TLS Handshake errors by following this document.