Configuring Typha

Operator

Typha configuration cannot be modified when Calico Enterprise is installed via the operator.

Manifest

Configuration for Typha is read from one of two possible locations, in order, as follows.

1. Environment variables, prefixed with TYPHA_.

2. The Typha configuration file. The path to this file defaults to /etc/calico/typha.cfg but can be overridden using the -c or --config-file options on the command line.

The value of any configuration parameter is the value read from the first location containing a value. For example, if an environment variable contains a value, it takes precedence.

If not set in any of these locations, most configuration parameters have defaults, and it should be rare to have to explicitly set them.

The full list of parameters which can be set is as follows.

General configuration

Configuration parameter	Environment variable	Description	Schema
DatastoreType	TYPHA_DATASTORETYPE	The datastore that Typha should read endpoints and policy information from. [Default: etcdv3]	etcdv3, kubernetes
HealthEnabled	TYPHA_HEALTHENABLED	When enabled, exposes Typha health information via an http endpoint.	boolean
HealthPort	TYPHA_HEALTHPORT	The port that Typha will serve health information over. [Default: 9098]	int
HealthHost	TYPHA_HEALTHHOST	The address that Typha will bind its health endpoint to. [Default: localhost]	string
LogFilePath	TYPHA_LOGFILEPATH	The full path to the Typha log. Set to none to disable file logging. [Default: /var/log/calico/typha.log]	string
LogSeverityFile	TYPHA_LOGSEVERITYFILE	The log severity above which logs are sent to the log file. [Default: Info]	Debug, Info, Warning, Error, Fatal
LogSeverityScreen	TYPHA_LOGSEVERITYSCREEN	The log severity above which logs are sent to the stdout. [Default: Info]	Debug, Info, Warning, Error, Fatal
LogSeveritySys	TYPHA_LOGSEVERITYSYS	The log severity above which logs are sent to the syslog. Set to "" for no logging to syslog. [Default: Info]	Debug, Info, Warning, Error, Fatal
PrometheusGoMetricsEnabled	TYPHA_PROMETHEUSGOMETRICSENABLED	Set to false to disable Go runtime metrics collection, which the Prometheus client does by default. This reduces the number of metrics reported, reducing Prometheus load. [Default: true]	boolean
PrometheusMetricsEnabled	TYPHA_PROMETHEUSMETRICSENABLED	Set to true to enable the Prometheus metrics server in Typha. [Default: false]	boolean
PrometheusMetricsHost	TYPHA_PROMETHEUSMETRICSHOST	TCP network address that the Prometheus metrics server should bind to. [Default: ""]	string
PrometheusMetricsPort	TYPHA_PROMETHEUSMETRICSPORT	TCP port that the Prometheus metrics server should bind to. [Default: 9091]	int
PrometheusProcessMetricsEnabled	TYPHA_PROMETHEUSPROCESSMETRICSENABLED	Set to false to disable process metrics collection, which the Prometheus client does by default. This reduces the number of metrics reported, reducing Prometheus load. [Default: true]	boolean
ServerHandshakeTimeoutSecs	TYPHA_SERVERHANDSHAKETIMEOUTSECS	Maximum time that Typha server should wait for the client TLS handshake to be performed.	int
ShutdownTimeoutSecs	TYPHA_SHUTDOWNTIMEOUTSECS	Maximum time that Typha should take to do a graceful shut down. In Kubernetes, this should match Typha's terminationGracePeriodSeconds.	int
ShutdownConnectionDropIntervalMaxSecs	TYPHA_SHUTDOWNCONNECTIONDROPINTERVALMAXSECS	Maximum time between terminating two connections when doing a graceful shutdown. Prevents very slow shut downs if ShutdownTimeoutSecs is large but Typha only has a small number of clients.	int

note

By default, if the health endpoint is enabled Typha listens on localhost. However, if Typha is used in Kubernetes, the kubelet will do health checks using the pod IP. To work around this discrepancy, the Typha image supports a health-check CLI command that fetches the health endpoint: calico-typha check (readiness|liveness) --port=<port>. If you modify the health port, you will need to add the --port=<port> argument to the liveness and readiness probe commands in the manifest.

Kubernetes API datastore configuration

The Kubernetes API datastore driver reads its configuration from Kubernetes-provided environment variables.

Calico Enterprise specific configuration

Setting	Environment variable	Default	Meaning
PrometheusMetricsCertFile	TYPHA_PROMETHEUSMETRICSCERTFILE	None	Certificate for encrypting Typha Prometheus metrics.
PrometheusMetricsKeyFile	TYPHA_PROMETHEUSMETRICSKEYFILE	None	Private key for encrypting Typha Prometheus metrics.
PrometheusMetricsCAFile	TYPHA_PROMETHEUSMETRICSCAFILE	None	Trusted CA file for clients attempting to read Typha Prometheus metrics.

When the PrometheusMetrics...File parameters are set, Typha's Prometheus port is TLS-secured such that only a validated client can read Prometheus metrics, and the data is encrypted in transit. A valid client must then connect over HTTPS and present a certificate that is signed by one of the trusted CAs in the PrometheusMetricsCAFile setting.

Environment variables

Environment	Description	Schema
USE_POD_CIDR	Use the Kubernetes Node.Spec.PodCIDR field. This field is required when using the Kubernetes API datastore with host-local IPAM. [Default: false]	boolean

Felix-Typha TLS configuration

Configuration parameter	Environment variable	Description	Schema
CAFile	TYPHA_CAFILE	Path to the file containing the root certificate of the CA that issued the Felix client certificate. Configures Typha to trust the CA that signed the Felix client certificate. The file may contain multiple root certificates, causing Typha to trust each of the CAs included. Example: /etc/typha/ca.pem	string
ClientCN	TYPHA_CLIENTCN	If set, the Common Name that Felix's certificate must have. If you have enabled TLS on the communications from Felix to Typha, you must set a value here or in ClientURISAN. You can set values in both, as well, such as to facilitate a migration from using one to the other. If either matches, the communication succeeds. [Default: none]	string
ClientURISAN	TYPHA_CLIENTURISAN	If set, a URI SAN that Felix's certificate must have. We recommend populating this with a SPIFFE string that identifies Felix. All Felix instances should use the same SPIFFE ID. If you have enabled TLS on the communications from Felix to Typha, you must set a value here or in ClientCN. You can set values in both, as well, such as to facilitate a migration from using one to the other. If either matches, the communication succeeds. [Default: none]	string
ServerCertFile	TYPHA_SERVERCERTFILE	Path to the file containing the server certificate issued to Typha. Typha presents this to Felix clients during the TLS handshake. Example: /etc/typha/cert.pem	string
ServerKeyFile	TYPHA_SERVERKEYFILE	Path to the file containing the private key matching the Typha server certificate. Example: /etc/typha/key.pem (optional)	string

For more information on how to use and set these variables, refer to Connections from Node to Typha (Kubernetes).