Consul envoy sidecar proxy - cluster related configuration

when i check envoy proxy admin
in “/clusters” url , i can check cluster infomation

i can see “cx_active” value
i think this value is for active connection count
active conneciton is remained because of keep-alive time
i want to remove keep-alive configuration of envoy cluster

is there some way to do this in consul ?

echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::cx_active::3
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::cx_connect_fail::1
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::cx_total::46
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::rq_active::3
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::rq_error::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::rq_success::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::rq_timeout::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::rq_total::45
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::hostname::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::health_flags::healthy
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::weight::1
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::region::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::zone::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::sub_zone::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::canary::false
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::priority::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::success_rate::-1
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot}::local_origin_success_rate::-1
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::cx_active::5
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::cx_connect_fail::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::cx_total::64
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::rq_active::5
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::rq_error::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::rq_success::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::rq_timeout::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::rq_total::64
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::hostname::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::health_flags::healthy
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::weight::1
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::region::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::zone::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::sub_zone::
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::canary::false
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::priority::0
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::success_rate::-1
echo-dc2-hc.default.dc2.internal.aaaaa.consul::{ip:prot2}::local_origin_success_rate::-1

You can try to remove the keepalive settings with the property override extension. Here’s an example config.

Kind = "service-defaults"
Name = "downstream-service"
Protocol = "http"
EnvoyExtensions = [
  {
    Name = "builtin/property-override"
    Arguments = {
      ProxyType = "connect-proxy"
      Patches = [
        {
          ResourceFilter = {
            ResourceType = "cluster"
            TrafficDirection = "outbound"
            Service = {
              Name = "upstream-svc"
            }
          }
          Op = "remove"
          Path =  "/upstream_connection_options/tcp_keepalive"
        }
      ]
    }
  }
]

Refer to https://www.envoyproxy.io/docs/envoy/latest/api-v3/config/cluster/v3/cluster.proto#config-cluster-v3-upstreamconnectionoptions for a list of supported fields.