Commit graph

4 commits

Author SHA1 Message Date
Ishaan Jaff
8f155327f6 [Fix] Router cooldown logic - use % thresholds instead of allowed fails to cooldown deployments (#5698)
* move cooldown logic to it's own helper

* add new track deployment metrics folder

* increment success, fails for deployment in current minute

* fix cooldown logic

* fix test_aaarouter_dynamic_cooldown_message_retry_time

* fix test_single_deployment_no_cooldowns_test_prod_mock_completion_calls

* clean up get from deployment test

* fix _async_get_healthy_deployments

* add mock InternalServerError

* test deployment failing 25% requests

* add test_high_traffic_cooldowns_one_bad_deployment

* fix vertex load test

* add test for rate limit error models in cool down

* change default cooldown time

* fix cooldown message time

* fix cooldown on 429 error

* fix doc string for _should_cooldown_deployment

* fix sync cooldown logic router
2024-09-14 18:01:19 -07:00
Krrish Dholakia
7243e6c8ce fix(cooldown_cache.py): fix linting errors 2024-08-27 07:40:28 -07:00
Krrish Dholakia
4918082cd7 fix(cooldown_cache.py): fix linting errors 2024-08-24 17:11:32 -07:00
Krrish Dholakia
c795e9feeb fix(router.py): enable dynamic retry after in exception string
Updates cooldown logic to cooldown individual models

 Closes https://github.com/BerriAI/litellm/issues/1339
2024-08-24 16:59:30 -07:00