Current state: Under Discussion
Discussion thread: here
JIRA: TODO
KIP-714 introduced the ClientTelemetryReceiver interface to enable MetricsReporter instances running on brokers to collect client telemetry metrics. Typically a reporter stores metrics it receives in a "metrics registry" and exposes them in a format suitable for the external monitoring system it is built for. This is exactly what JmxReporter does. Kafka clients and brokers explicitly delete their metrics when they are not used anymore. However for client telemetry metrics there isn't an explicit deletion mechanism.
Clients send their telemetry metrics at a regular interval, configured via telemetry subscriptions. Reporters are not aware of this push interval. Without the interval and an explicit deletion mechanism, reporters don't know when to clear client telemetry metrics and stop exposing them. This makes it hard to not expose stale metrics. Having uncertainty in metrics is a major problem as operators rely on them to manage their clusters and investigate issues.
This KIP proposes updating the context associated with each client telemetry push to include the push interval. This will enable reporters to properly manage the life cycle of client telemetry metrics.
package org.apache.kafka.server.telemetry;
import org.apache.kafka.server.authorizer.AuthorizableRequestContext;
/**
* Context provided to {@link ClientTelemetryReceiver} implementations when receiving client metrics.
*/
public interface ClientTelemetryContext extends AuthorizableRequestContext {
/**
* The interval defined via <code>metrics.interval</code> in the client metrics subscription
* @return The interval in milliseconds
*/
int pushIntervalMs();
/**
* The context associated with this request
* @return The AuthorizableRequestContext associated with this request
*/
AuthorizableRequestContext authorizableRequestContext();
} |
A new method in ClientTelemetryReceiver
/**
* Called by the broker when a client reports telemetry metrics. The associated telemetry context
* can be used by the metrics plugin to retrieve additional client information such as client ids,
* endpoints and the push interval.
* <p>
* This method may be called from the request handling thread, and as such should avoid blocking.
*
* @param context the client telemetry context for the corresponding {@code PushTelemetryRequest}
* api call.
* @param payload the encoded telemetry payload as sent by the client.
*/
default void exportMetrics(ClientTelemetryContext context, ClientTelemetryPayload payload) {
exportMetrics(context.authorizableRequestContext(), payload);
} |
Deprecate the existing ClientTelemetryReceiver.exportMetrics() method
/**
* Called by the broker when a client reports telemetry metrics. The associated request context
* can be used by the metrics plugin to retrieve additional client information such as client ids
* or endpoints.
* <p>
* This method may be called from the request handling thread, and as such should avoid blocking.
* <p>
* This method is deprecated, {@link #exportMetrics(ClientTelemetryContext, ClientTelemetryPayload)} should be used
* instead.
*
* @param context the client request context for the corresponding {@code PushTelemetryRequest}
* api call.
* @param payload the encoded telemetry payload as sent by the client.
*/
@Deprecated(since = "4.2.0")
void exportMetrics(AuthorizableRequestContext context, ClientTelemetryPayload payload); |
The new method has a default implementation so it should not break existing classes implementing ClientTelemetryReceiver.
Implementations that want to benefit from the new context need to override both the new and old exportMetrics() methods. The body for the old method can be the following for example, since it won't be called if the new method is also implemented.
public void exportMetrics(AuthorizableRequestContext context, ClientTelemetryPayload payload) {
throw new IllegalStateException("Should not be called");
} |
The plan is to delete the now deprecated exportMetrics() method in the next major version, Kafka 5.0.0.
Adequate unit and integration tests will be added to ensure ClientTelemetryReceiver implementations use the old or new contexts.
N/A