Hi! I'm trying to figure out if I can get the original client IP when using kubernetes services/load balancer.
My test is a docker container running a simple TCP server that returns to the client the IP addresses of the socket (written in c, using getsockname/getpeername). On an on-prem cluster, I get results similar to this - server = 172.30.182.87:5555 client = 10.73.68.170:60155 server = 172.30.195.26:5555 client = 10.73.68.170:25024 server = 172.30.195.26:5555 client = 10.73.68.170:60157 server = 172.30.195.26:5555 client = 10.73.68.170:47226 Which are what I expected - 2 pods and the same IP address for the client (not the actual one, but that's a separate issue!). But when doing the same on an AWS cluster I get the following - server = 100.96.3.4:5555 client = 172.20.41.6:10397 server = 100.96.3.4:5555 client = 100.96.3.1:55627 server = 100.96.2.3:5555 client = 172.20.49.85:52491 server = 100.96.3.4:5555 client = 172.20.41.6:10403 server = 100.96.3.4:5555 client = 100.96.3.1:55633 What are these client IPs? the 100.x.x.x are the service endpoints, but I don't understand where the 172.x.x.x came from. And why do I get different client IPs when I run it from the same machine? Thanks in advance to anyone who can shed some light on this:) Gail -- You received this message because you are subscribed to the Google Groups "Kubernetes user discussion and Q&A" group. To unsubscribe from this group and stop receiving emails from it, send an email to kubernetes-users+unsubscr...@googlegroups.com. To post to this group, send email to kubernetes-users@googlegroups.com. Visit this group at https://groups.google.com/group/kubernetes-users. For more options, visit https://groups.google.com/d/optout.
