Adaptive LIFO Queue Policy (proto)

This extension has the qualified name envoy.queue_policy.adaptive_lifo

Note

This extension is functional but has not had substantial production burn time, use only with this caveat.

This extension is intended to be robust against both untrusted downstream and upstream traffic.

Tip

This extension extends and can be used with the following extension category:

• envoy.queue_policy

This extension must be configured with one of the following type URLs:

• type.googleapis.com/envoy.extensions.queue_policy.adaptive_lifo.v3.AdaptiveLifoQueuePolicyConfig

extensions.queue_policy.adaptive_lifo.v3.AdaptiveLifoQueuePolicyConfig

[extensions.queue_policy.adaptive_lifo.v3.AdaptiveLifoQueuePolicyConfig proto]

Configuration for the adaptive LIFO pending requests queue policy. Watermarks are absolute numbers of requests. Each connection pool applies them to its own pending request queue.

Requests are dequeued in FIFO order until the queue size reaches the high watermark. Requests are then dequeued in LIFO order until the queue size reaches or falls below the low watermark.

{
  "high_watermark": ...,
  "low_watermark": ...
}

high_watermark

(uint32) Number of queued requests at which the queue switches from FIFO to LIFO ordering. This must be greater than zero and greater than low_watermark.

low_watermark

(uint32) Number of queued requests at or below which the queue switches from LIFO back to FIFO ordering. This may be zero and must be less than high_watermark.