Document best practices on LB configuration(s).

[timothysc]

This is a...

• Feature Request
• Bug Report

Problem:
Missing documentation on best practices for LB configurations.

Proposed Solution:

• Leverage /healthz endpoint in LB configs settings.

  • http://docs.aws.amazon.com/elasticloadbalancing/latest/classic/elb-healthchecks.html . This is to ensure that requests are properly sent to active api-servers.

• Encourage LBs leverage (source) settings to ensure cache hits.

  • Over the course of several years, the api-server has started to maintain larger and larger caches, encouraging source will leverage those caches more effectively. (We should metrics hit %s honestly)

• Update connection timeouts to their highest possible settings as this can lead to expensive relists.

• Internal and External LBs.

  • Traffic isolation, cache hits, and external LB's don't need such a high TTL.

Page to Update:
TODO

[jamiehannaford]

@timothysc I can work on this (feel free to assign). I'm a bit unclear on #2, what type of apiserver caching are you referring to? Do you expect the LB to handle content caching or the apiserver?

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

Prevent issues from auto-closing with an /lifecycle frozen comment.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or @fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or @fejta.
/lifecycle rotten
/remove-lifecycle stale

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close