Runtime
Upstream clusters support the following runtime settings:
Active health checking
- health_check.min_interval
Min value for the health checking interval. Default value is 1 ms. The effective health check interval will be no less than 1ms. The health checking interval will be between min_interval and max_interval.
- health_check.max_interval
Max value for the health checking interval. Default value is MAX_INT. The effective health check interval will be no less than 1ms. The health checking interval will be between min_interval and max_interval.
- health_check.verify_cluster
What % of health check requests will be verified against the expected upstream service as the health check filter will write the remote service cluster into the response.
Outlier detection
See the outlier detection architecture overview for more information on outlier detection. The runtime parameters supported by outlier detection are the same as the static configuration parameters, namely:
- outlier_detection.consecutive_5xx
consecutive_5XX setting in outlier detection
- outlier_detection.consecutive_gateway_failure
consecutive_gateway_failure setting in outlier detection
- outlier_detection.consecutive_local_origin_failure
consecutive_local_origin_failure setting in outlier detection
- outlier_detection.interval_ms
interval setting in outlier detection
- outlier_detection.base_ejection_time_ms
base_ejection_time setting in outlier detection
- outlier_detection.max_ejection_percent
max_ejection_percent setting in outlier detection
- outlier_detection.enforcing_consecutive_5xx
enforcing_consecutive_5xx setting in outlier detection
- outlier_detection.enforcing_consecutive_gateway_failure
enforcing_consecutive_gateway_failure setting in outlier detection
- outlier_detection.enforcing_consecutive_local_origin_failure
enforcing_consecutive_local_origin_failure setting in outlier detection
- outlier_detection.enforcing_success_rate
enforcing_success_rate setting in outlier detection
- outlier_detection.enforcing_local_origin_success_rate
enforcing_local_origin_success_rate setting in outlier detection
- outlier_detection.success_rate_minimum_hosts
success_rate_minimum_hosts setting in outlier detection
- outlier_detection.success_rate_request_volume
success_rate_request_volume setting in outlier detection
- outlier_detection.success_rate_stdev_factor
success_rate_stdev_factor setting in outlier detection
- outlier_detection.enforcing_failure_percentage
enforcing_failure_percentage setting in outlier detection
- outlier_detection.enforcing_failure_percentage_local_origin
enforcing_failure_percentage_local_origin setting in outlier detection
- outlier_detection.failure_percentage_request_volume
failure_percentage_request_volume setting in outlier detection
- outlier_detection.failure_percentage_minimum_hosts
failure_percentage_minimum_hosts setting in outlier detection
- outlier_detection.failure_percentage_threshold
failure_percentage_threshold setting in outlier detection
- outlier_detection.max_ejection_time_ms
max_ejection_time setting in outlier detection
- outlier_detection.max_ejection_time_jitter_ms
base_ejection_time setting in outlier detection
Core
- upstream.healthy_panic_threshold
Sets the panic threshold percentage. Defaults to 50%.
- upstream.use_http2
Whether the cluster uses
HTTP/2if configured in HttpProtocolOptions. Set to 0 to disable HTTP/2 even if the feature is configured. Defaults to enabled.- upstream.use_http3
Whether the cluster uses
HTTP/3if configured in HttpProtocolOptions. Set to 0 to disable HTTP/3 even if the feature is configured. Defaults to enabled.
Zone aware load balancing
- upstream.zone_routing.enabled
% of requests that will be routed to the same upstream zone. Defaults to 100% of requests.
- upstream.zone_routing.min_cluster_size
Minimal size of the upstream cluster for which zone aware routing can be attempted. Default value is 6. If the upstream cluster size is smaller than min_cluster_size zone aware routing will not be performed.
Circuit breaking
- circuit_breakers.<cluster_name>.<priority>.max_connections
- circuit_breakers.<cluster_name>.<priority>.max_pending_requests
- circuit_breakers.<cluster_name>.<priority>.max_requests
- circuit_breakers.<cluster_name>.<priority>.max_retries
- circuit_breakers.<cluster_name>.<priority>.retry_budget.budget_percent
- circuit_breakers.<cluster_name>.<priority>.retry_budget.min_retry_concurrency