-
Trying to make simple api calls from sales force, AgentForce and they die at the gateway, We get a 400 error and the only logs i see is this {
"searchResults": "Error: 400 Bad Request"
} {
"bytes_received": 0,
"bytes_sent": 11,
"cluster_name": "example-cluster",
"container_name": "envoy",
"duration": 0,
"entity.guid": "MjM5MTMxMnxFWFR8U0VSVklDRXwtNTQ3NTc2NzA3NDY0MjI5NTg0",
"entity.guids": "MjM5MTMxMnxFWFR8U0VSVklDRXwtNTQ3NTc2NzA3NDY0MjI5NTg0",
"entity.name": "envoy",
"entity.type": "SERVICE",
"method": "null",
"newrelic.source": "api.logs",
"pod_labels.app.kubernetes.io/component": "proxy",
"pod_labels.app.kubernetes.io/managed-by": "envoy-gateway",
"pod_labels.app.kubernetes.io/name": "envoy",
"pod_labels.gateway.envoyproxy.io/owning-gateway-name": "acme-gateway",
"pod_labels.gateway.envoyproxy.io/owning-gateway-namespace": "envoy-gateway-system",
"pod_labels.kubernetes.acme.com/role": "server",
"pod_labels.kubernetes.acme.com/service": "envoy-gateway",
"pod_labels.kubernetes.acme.com/team": "acme",
"pod_labels.pod-template-hash": "7fc447488d",
"pod_name": "acme-gateway-7fc447488d-qtth2",
"pod_namespace": "envoy-gateway-system",
"pod_node_name": "nodename",
"pod_owner": "ReplicaSet/acme-gateway-7fc447488d",
"response_code": 400,
"response_flags": "DownstreamProtocolError",
"route_name": "null",
"service_name": "envoy",
"source_type": "kubernetes_logs",
"start_time": "2025-05-20T15:28:07.326Z",
"timestamp": 1747754888299,
"upstream_host": "null",
"upstream_ssl_cipher": "null",
"user_agent": "null",
"vhost": "null",
"x-envoy-original-path": "null",
"x-forwarded-for": "null",
"x-request-id": "null"
} When making the same call to nginx, it works and i get the following {
"bytes_sent": "458",
"cluster_name": "example-cluster",
"container_name": "nginx-ingress-controller",
"duration": "0.511",
"entity.guid": "MjM5MTMxMnxFWFR8U0VSVklDRXwtNzk2ODczMjAzNDcyODAzNDk1MQ",
"entity.guids": "MjM5MTMxMnxFWFR8U0VSVklDRXwtNzk2ODczMjAzNDcyODAzNDk1MQ",
"entity.name": "ingress-nginx",
"entity.type": "SERVICE",
"http_referrer": "",
"http_user_agent": "SFDC-Callout/63.0",
"ingress": "default/agent-ecosystem-gateway",
"method": "POST",
"newrelic.source": "api.logs",
"path": "/aeg/api/v1/aura-search",
"pod_name": "nginx-ingress-controller-74677fcd8-d64sl",
"pod_node_name": "nodename",
"remote_addr": "",
"request_length": "2400",
"request_proto": "HTTP/1.1",
"request_query": "",
"service_name": "ingress-nginx",
"source_type": "kubernetes_logs",
"ssl_cipher": "TLS_AES_256_GCM_SHA384",
"status": "401",
"time": "2025-05-20T14:57:53+00:00",
"timestamp": 1747753081323,
"traceparent": "",
"upstream": "default-agent-ecosystem-gateway-5000",
"upstream_connect_time": "0.001",
"upstream_header_time": "0.511",
"upstream_response_time": "0.512",
"upstream_server": "10.101.71.167:5000",
"vhost": "example.com",
"x_auto_test": "",
"x_forward_for": "IPADDDRESS"
} Is this a cipher mismatch? I've tried setting some client traffic policys to adjust it but with no luck |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 10 replies
-
any idea @rudrakhp ? |
Beta Was this translation helpful? Give feedback.
-
@davem-git can you share your gateway configurations (both EG and NGINX)? Access log has no info, so Envoy wasn't able to parse anything from the request:
I think there is a mismatch between the protocol used by the caller and your gateway. |
Beta Was this translation helpful? Give feedback.
I think we figured it out. enableing HTTP debug was hard to read so i missed