envoyproxy/envoy

Overload Manager should monitor epoll loop duration and polling delay as resources

Open

#12,190 opened on Jul 20, 2020

View on GitHub
 (2 comments) (0 reactions) (0 assignees)C++ (5,373 forks)batch import
area/overload_managerenhancementhelp wanted

Repository metrics

Stars
 (27,997 stars)
PR merge metrics
 (Avg merge 8d) (378 merged PRs in 30d)

Description

The Overload Manager has a configurable set of resource monitors that can be used to trigger actions when Envoy is overloaded. Monitoring the epoll loop duration and polling delay would allow proxies to take action when they are over-taxed in terms of CPU, possibly with a large number of pending events to handle. This could be done by polling the epoll stat metric, but I'm not sure that's the most efficient option.

Contributor guide

Research direction
Study the Overload Manager implementation in source/common/overload/ and explore how epoll stats are exposed via the stats interface. Look for existing epoll stat metrics and consider how to extend the resource monitor framework.
Tech stack
cpp
Domain
backendnetworking
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
C++Envoy internalsLinux epoll
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30