Gloo Edge Port Reference

Gloo Edge and Gloo Edge Enterprise both deploy containers that listen on certain ports for incoming traffic. This document lists out the pods and services that make up Gloo Edge and Gloo Edge Enterprise, and the ports which these pods and services listen on. It is also possible to set up mutual TLS (mTLS) for communication between Gloo Edge resources. The addition of mTLS changes the ports and traffic flows slightly, which is addressed in this document as well.

This document is specific to the pods and services deployed in a Kubernetes environment. Deploying Gloo Edge using the HashiCorp stack is also supported, and the port mappings and services should be the same.

It is possible to customize some port settings by providing custom values to the Helm chart that installs Gloo Edge open-source and Gloo Edge Enterprise. The port reference below is for an installation of Gloo Edge that uses the default settings in the Helm chart.


Gloo Edge Open-source

Gloo Edge open-source software is the free, open-source version of Gloo Edge. The installation process uses a Helm chart to create the necessary custom resource definitions (CRDs), deployments, services, pods, etc. The services and pods listen on specific ports to enable communication between the components that make up Gloo Edge and outside sources that will consume Upstream resources through Gloo Edge.

What’s included

A standard installation of Gloo Edge includes four primary components:

Pods and ports

The four primary components are instantiated using pods and services. The following table lists the deployed pods and ports in use by each pod, as well as the optional access-log pod if Access Logging has been enabled.

Pod Port Usage
gateway 8443 Validation
gloo 9977 xDS Server
gloo 9988 Validation
gloo 9979 WASM cache
gateway-proxy 8080 HTTP
gateway-proxy 8443 HTTPS
gateway-proxy 19000 Envoy admin
access-log 8083 Access logging

The discovery pod does not listen on any ports as it uses outbound connections only.

Services and ports

The following table lists the services backed by the deployed pods.

Service Port Target Target Port Usage
gateway 443 gateway 8443 Validation
gloo 9977 gloo 9977 xDS Server
gloo 9988 gloo 9988 Validation
gloo 9979 gloo 9979 WASM cache
gloo 9966 gloo 9966 Proxy Debug gRPC
gateway-proxy 80 gateway-proxy 8080 HTTP
gateway-proxy 443 gateway-proxy 8443 HTTPS
access-log 8083 access-log 8083 Access logging

Gloo Edge Enterprise

Gloo Edge Enterprise adds many pods and services to provide the extra functionality included in the paid offering. More information on what is included in Gloo Edge Enterprise can be found on the Gloo Edge product page.

What’s included

At a high level, the following additional components are available in Gloo Edge Enterprise.

The Prometheus server and Grafana dashboard are optional components. If you have an existing instance of either, they can be used instead. More information is available in the Observability section of the docs.

Pods and ports

The Gloo Edge Enterprise components are instantiated using pods and services. The following table lists the deployed pods and ports in use by each pod.

Pod Port Usage
gloo-fed-console 8090 UI server
gloo-fed-console 10101 API Server
gloo-fed-console 8081 healthcheck
extauth 8083 External authentication
grafana 80 Grafana (unused)
grafana 3000 Grafana UI
prometheus-kube-state-metrics 8080 Kubernetes metric collection
prometheus-server 9090 Prometheus server
rate-limit 18081 Rate-limiting
redis 6379 Rate-limiting

There is an observability pod that automatically configures dashboards on the Grafana instance. It does not accept inbound traffic, so it is not included in the table above.

Services and ports

The following table lists the services backed by the deployed pods.

Service Port Target Target Port Usage
gloo-fed-console 8090 UI server
gloo-fed-console 10101 API Server
gloo-fed-console 8081 healthcheck
extauth 8083 extauth 8083 External authentication
glooe-grafana 80 grafana 3000 Grafana UI
glooe-prometheus-kube-state-metrics-v2 80 prometheus-kube-state-metrics 8080 Kubernetes metric collection
glooe-prometheus-server 80 prometheus-server 9090 Prometheus server
rate-limit 18081 rate-limit 18081 Rate-limiting
redis 6379 redis 6379 Rate-limiting

mTLS considerations

Gloo Edge supports the use of mutual TLS (mTLS) communication between the Gloo Edge pod and other services, including the Envoy proxy, Extauth server, and Rate-limiting server. Enabling mTLS includes the addition of sidecars for multiple pods, Envoy proxy for TLS termination and SDS for certificate rotation and management. More information on the details of mTLS implementation are available in the mTLS doc.

Updated pods

The following pods are updated to support mTLS:

The additional Envoy sidecar has an admin port listening on 8081 for each pod.

Updated traffic flow

The Envoy sidecar on the Gloo Edge, Extauth, and Rate-limit pods will be intercepting the inbound traffic for each pod and performing the TLS decryption before passing the traffic to the main container. This does not alter the ports being used by the pods and services, but it does create additional ports that are used internally within the pod for communication. For instance, the Gloo Edge pod continues to listen on 9977 as the xDS server. Internally, the Gloo Edge container is listening on 127.0.0.1:9999 for xDS requests. The Envoy sidecar in the pod accepts requests on 9977, decrypts the request, and sends it to port 9999 on the localhost for processing.


Summary and next steps

This document provides the ports being used by pods and services on a default installation of Gloo Edge and Gloo Edge Enterprise. Some of these ports can be customized, and additional components can be added that introduce more pods and services. To better understand the architecture of Gloo Edge, we recommend reading the following docs: