1. Documentation
×

Configuring the cluster-wide proxy

Production environments can deny direct access to the internet and instead have an HTTP or HTTPS proxy available. You can configure OKD to use a proxy by modifying the Proxy object for existing clusters or by configuring the proxy settings in the install-config.yaml file for new clusters.

After you enable a cluster-wide egress proxy for your cluster on a supported platform, Fedora CoreOS (FCOS) populates the status.noProxy parameter with the values of the networking.machineNetwork[].cidr, networking.clusterNetwork[].cidr, and networking.serviceNetwork[] fields from your install-config.yaml file that exists on the supported platform.

As a postinstallation task, you can change the networking.clusterNetwork[].cidr value, but not the networking.machineNetwork[].cidr and the networking.serviceNetwork[] values. For more information, see "Configuring the cluster network range".

For installations on Amazon Web Services (AWS), Google Cloud Platform (GCP), Microsoft Azure, and OpenStack, the status.noProxy parameter is also populated with the instance metadata endpoint, 169.254.169.254.

Example of values added to the status: segment of a Proxy object by FCOS

apiVersion: config.openshift.io/v1
kind: Proxy
metadata:
  name: cluster
# ...
networking:
  clusterNetwork: (1)
  - cidr: <ip_address_from_cidr>
    hostPrefix: 23
  network type: OVNKubernetes
  machineNetwork: (2)
  - cidr: <ip_address_from_cidr>
  serviceNetwork: (3)
  - 172.30.0.0/16
# ...
status:
  noProxy:
  - localhost
  - .cluster.local
  - .svc
  - 127.0.0.1
  - <api_server_internal_url> (4)
# ...
1 Specify IP address blocks from which pod IP addresses are allocated. The default value is 10.128.0.0/14 with a host prefix of /23.
2 Specify the IP address blocks for machines. The default value is 10.0.0.0/16.
3 Specify IP address block for services. The default value is 172.30.0.0/16.
4 You can find the URL of the internal API server by running the oc get infrastructures.config.openshift.io cluster -o jsonpath='{.status.etcdDiscoveryDomain}' command.

If your installation type does not include setting the networking.machineNetwork[].cidr field, you must include the machine IP addresses manually in the .status.noProxy field to make sure that the traffic between nodes can bypass the proxy.

Prerequisites

Review the sites that your cluster requires access to and determine whether any of them must bypass the proxy. By default, all cluster system egress traffic is proxied, including calls to the cloud provider API for the cloud that hosts your cluster. The system-wide proxy affects system components only, not user workloads. If necessary, add sites to the spec.noProxy parameter of the Proxy object to bypass the proxy.

Enabling the cluster-wide proxy

The Proxy object is used to manage the cluster-wide egress proxy. When a cluster is installed or upgraded without the proxy configured, a Proxy object is still generated but it will have a nil spec. For example:

apiVersion: config.openshift.io/v1
kind: Proxy
metadata:
  name: cluster
spec:
  trustedCA:
    name: ""
status:

A cluster administrator can configure the proxy for OKD by modifying this cluster Proxy object.

Only the Proxy object named cluster is supported, and no additional proxies can be created.

Enabling the cluster-wide proxy causes the Machine Config Operator (MCO) to trigger node reboot.
Prerequisites

  • Cluster administrator permissions

  • OKD oc CLI tool installed

Procedure

  1. Create a config map that contains any additional CA certificates required for proxying HTTPS connections.

    You can skip this step if the proxy’s identity certificate is signed by an authority from the RHCOS trust bundle.

    1. Create a file called user-ca-bundle.yaml with the following contents, and provide the values of your PEM-encoded certificates:

      apiVersion: v1
data:
  ca-bundle.crt: | (1)
    <MY_PEM_ENCODED_CERTS> (2)
kind: ConfigMap
metadata:
  name: user-ca-bundle (3)
  namespace: openshift-config (4)
      1 This data key must be named ca-bundle.crt.
      2 One or more PEM-encoded X.509 certificates used to sign the proxy’s identity certificate.
      3 The config map name that will be referenced from the Proxy object.
      4 The config map must be in the openshift-config namespace.

    2. Create the config map from this file:

      $ oc create -f user-ca-bundle.yaml

  2. Use the oc edit command to modify the Proxy object:

    $ oc edit proxy/cluster

  3. Configure the necessary fields for the proxy:

    apiVersion: config.openshift.io/v1
kind: Proxy
metadata:
  name: cluster
spec:
  httpProxy: http://<username>:<pswd>@<ip>:<port> (1)
  httpsProxy: https://<username>:<pswd>@<ip>:<port> (2)
  noProxy: example.com (3)
  readinessEndpoints:
  - http://www.google.com (4)
  - https://www.google.com
  trustedCA:
    name: user-ca-bundle (5)
    1 A proxy URL to use for creating HTTP connections outside the cluster. The URL scheme must be http.
    2 A proxy URL to use for creating HTTPS connections outside the cluster. The URL scheme must be either http or https. Specify a URL for the proxy that supports the URL scheme. For example, most proxies will report an error if they are configured to use https but they only support http. This failure message may not propagate to the logs and can appear to be a network connection failure instead. If using a proxy that listens for https connections from the cluster, you may need to configure the cluster to accept the CAs and certificates that the proxy uses.
    3 A comma-separated list of destination domain names, domains, IP addresses (or other network CIDRs), and port numbers to exclude proxying.

    Port numbers are only supported when configuring IPv6 addresses. Port numbers are not supported when configuring IPv4 addresses.

    Preface a domain with . to match subdomains only. For example, .y.com matches x.y.com, but not y.com. Use * to bypass proxy for all destinations.

    If your noproxy field needs to include a domain address, you must explicitly specify that FQDN, or prefix-matched subdomain, in the noproxy field. You cannot use the IP address or CIDR range that encapsulates the domain. This is because the cluster does not wait for DNS to return the IP address before assigning the route connection, and checks explicitly against the request being made. For example, if you have a CIDR block value, such as 10.0.0.0/24, for the noproxy field and attempt to access https://10.0.0.11, it will match successfully. However, attempting to access https://exampleserver.externaldomain.com, whose A record entry is 10.0.0.11, will fail. An additional value of .externaldomain.com for your noproxy field is necessary.

    If you scale up workers that are not included in the network defined by the networking.machineNetwork[].cidr field from the installation configuration, you must add them to this list to prevent connection issues.

    This field is ignored if neither the httpProxy or httpsProxy fields are set.
    4 One or more URLs external to the cluster to use to perform a readiness check before writing the httpProxy and httpsProxy values to status.
    5 A reference to the config map in the openshift-config namespace that contains additional CA certificates required for proxying HTTPS connections. Note that the config map must already exist before referencing it here. This field is required unless the proxy’s identity certificate is signed by an authority from the RHCOS trust bundle.

  4. Save the file to apply the changes.

Removing the cluster-wide proxy

The cluster Proxy object cannot be deleted. To remove the proxy from a cluster, remove all spec fields from the Proxy object.

Prerequisites

  • Cluster administrator permissions

  • OKD oc CLI tool installed

Procedure

  1. Use the oc edit command to modify the proxy:

    $ oc edit proxy/cluster

  2. Remove all spec fields from the Proxy object. For example:

    apiVersion: config.openshift.io/v1
kind: Proxy
metadata:
  name: cluster
spec: {}

  3. Save the file to apply the changes.

Verifying the cluster-wide proxy configuration

After the cluster-wide proxy configuration is deployed, you can verify that it is working as expected. Follow these steps to check the logs and validate the implementation.

Prerequisites

  • You have cluster administrator permissions.

  • You have the OKD oc CLI tool installed.

Procedure

  1. Check the proxy configuration status using the oc command:

    $ oc get proxy/cluster -o yaml

  2. Verify the proxy fields in the output to ensure they match your configuration. Specifically, check the spec.httpProxy, spec.httpsProxy, spec.noProxy, and spec.trustedCA fields.

  3. Inspect the status of the Proxy object:

    $ oc get proxy/cluster -o jsonpath='{.status}'
    Example output

    {
status:
    httpProxy: http://user:xxx@xxxx:3128
    httpsProxy: http://user:xxx@xxxx:3128
    noProxy: .cluster.local,.svc,10.0.0.0/16,10.128.0.0/14,127.0.0.1,169.254.169.254,172.30.0.0/16,localhost,test.no-proxy.com
}

  4. Check the logs of the Machine Config Operator (MCO) to ensure that the configuration changes were applied successfully:

    $ oc logs -n openshift-machine-config-operator $(oc get pods -n openshift-machine-config-operator -l k8s-app=machine-config-operator -o name)

  5. Look for messages that indicate the proxy settings were applied and the nodes were rebooted if necessary.

  6. Verify that system components are using the proxy by checking the logs of a component that makes external requests, such as the Cluster Version Operator (CVO):

    $ oc logs -n openshift-cluster-version $(oc get pods -n openshift-cluster-version -l k8s-app=machine-config-operator -o name)

  7. Look for log entries that show that external requests have been routed through the proxy.

Additional resources