Hello Andy,
I'm seeing some issues, maybe it's some misconfiguration on my side.
But it looks like that when I try Piko with replicas > 1 there are some TLS errors - in particular when the proxy connects to one node and the upstream to the other:
{"level":"warn","ts":"2025-09-30T08:03:13.613Z","subsystem":"proxy","msg":"httputil: ReverseProxy read error during body copy: read tcp 10.0.28.94:44020->10.0.13.236:8000: read: connection reset by peer"}
In this case the proxy-side-connection is connected to10.0.28.94 and the upstream-side-connection to 10.0.13.236.
I've TLS configured in the proxy server. It looks like it's terminating the TLS and then forwarding to the node with the upstream.
Have you seen this before/have any recommendation here?
(I'm using Kubernetes/the helm charts)
Many thanks!
Hello Andy,
I'm seeing some issues, maybe it's some misconfiguration on my side.
But it looks like that when I try Piko with
replicas > 1there are some TLS errors - in particular when the proxy connects to one node and the upstream to the other:In this case the proxy-side-connection is connected to
10.0.28.94and the upstream-side-connection to10.0.13.236.I've TLS configured in the proxy server. It looks like it's terminating the TLS and then forwarding to the node with the upstream.
Have you seen this before/have any recommendation here?
(I'm using Kubernetes/the helm charts)
Many thanks!