Update config of a serving endpoint.

This is a long-running operation, which blocks until Serving Endpoints on Databricks reach NOT_UPDATING state with the timeout of 20 minutes, that you can change via timeout parameter. By default, the state of Databricks Serving Endpoints is reported to console. You can change this behavior by changing the callback parameter.

Usage

update_serving_endpoint_config_and_wait(
  client,
  name,
  auto_capture_config = NULL,
  served_entities = NULL,
  served_models = NULL,
  traffic_config = NULL,
  timeout = 20,
  callback = cli_reporter
)

Arguments

client

Required. Instance of DatabricksClient()

name

Required. The name of the serving endpoint to update.

auto_capture_config

Configuration for Inference Tables which automatically logs requests and responses to Unity Catalog.

served_entities

A list of served entities for the endpoint to serve.

served_models

(Deprecated, use served_entities instead) A list of served models for the endpoint to serve.

traffic_config

The traffic config defining how invocations to the serving endpoint should be routed.

timeout

Time to wait for the operation to complete in minutes.

callback

Function to report the status of the operation. By default, it reports to console.

Details

Updates any combination of the serving endpoint's served entities, the compute configuration of those served entities, and the endpoint's traffic config. An endpoint that already has an update in progress can not be updated until the current update completes or fails.