Getting InfluxDB connection information

Repertoire supports obtaining connection information for InfluxDB databases. Unlike other service discovery calls, the connection information includes credentials and therefore this API requires authentication. Also unlike other service discovery calls, the InfluxDB database may not be in the same Phalanx environment.

Listing available databases

To get a list of InfluxDB databases for which connection information is available, call DiscoveryClient.influxdb_databases:

from rubin.repertoire import DiscoveryClient


discovery = DiscoveryClient()
databases = await discovery.influxdb_databases()

The resulting names are short, human-readable names that can be passed as the database parameter to DiscoveryClient.get_influxdb_connection_info or DiscoveryClient.get_influxdb_credentials as described below.

Getting connection information

To get the connection information for a specific database, use DiscoveryClient.get_influxdb_connection_info:

from rubin.repertoire import DiscoveryClient


discovery = DiscoveryClient()
info = await discovery.get_influxdb_connection_info("some_database")

The resulting object has the following fields:

url

The URL of the InfluxDB service for this database. This is a pydantic.HttpUrl without authentication information. Use the host, port, and path attributes of the URL if you need those components separately (for the constructor of aioinflux.client.InfluxDBClient, for example).

database

The name of the InfluxDB database to use for queries.

schema_registry

The URL (as a string) of the associated Confluent Kafka Schema Registry that contains schema information for this database.

Getting credentials

Getting the connection information with credentials for a specific database requires authentication, since the returned information includes the username and password.

First, obtain a Gafaelfawr token. Inside a service, normally this should be a delegated token received as part of a request and used to act on behalf of the user. See the Gafaelfawr documentation on delegated tokens for more information. In other environments, this may be a user token created through the token UI, or a notebook token created by Nublado.

Then, call DiscoveryClient.get_influxdb_credentials with the name of the database (possibly obtained via DiscoveryClient.influxdb_databases) and that token:

from rubin.repertoire import DiscoveryClient


discovery = DiscoveryClient()
token = "..."  # obtained from somewhere else
info = await discovery.get_influxdb_credentials("some_database", token)

The resulting object has the same fields as that retured by DiscoveryClient.get_influxdb_connection_info, plus two more:

username

Username with which to authenticate.

password

Password with which to authenticate.