devin-ai-integration[bot] 9540f19e38 feat(proxy): add fail_closed_rate_limit_enforcement to reject requests with 503 while Redis rate limit counters are unreachable (#43251) ... * feat(proxy): add fail_closed_rate_limit_enforcement to reject requests with 503 while Redis rate limit counters are unreachable * fix(proxy): reject fail-closed rate limit checks before logging the in-memory fallback and pin the boot warning in the lifespan * fix(proxy): coerce the fail-closed flag, fail closed on read-only checks, and refund partial cluster increments * fix(proxy): window-guard rate limit refunds and catch the fail-closed rejection by type * fix(proxy): read the compaction rate-limit gate's limiter from the proxy hook registry * fix(proxy): count the pending request in read-only rate-limit checks and keep the compaction gate off the caller's parallel slot The compaction polyfill's summary-model gate, once it ran against the real v3 limiter, showed two behaviors nobody had chosen. The read-only check compared the stored counter with the same `>` the increment path uses, but a read-only check decides a request that has not been counted yet, so a summary model exactly at its rpm limit still went out. The read-only path now adds the pending increment of 1 before comparing; the increment path is unchanged. The gate also passed the key's max_parallel_requests gauge through, and the read-only gauge count includes the caller's own in-flight slot, so a key with max_parallel_requests: 1 never compacted. The gate now drops that gauge from its descriptors, since the summary call runs inside a request the limiter already admitted. --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>		2026-09-26 12:11:26 -07:00
..
litellm_skills	test: use local model cost map in import-isolation subprocess	2026-08-30 12:47:15 -07:00
test_async_post_call_streaming_iterator_hook.py	test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)	2026-08-22 09:25:58 -07:00
test_autorouter_baseline_cache.py	fix(proxy): estimate auto-router baseline costs from durable cache history	2026-09-19 12:44:47 -07:00
test_batch_enqueued_tokens.py	test(batch): prove an open Redis breaker keeps enqueued-token reservations quiet	2026-09-10 14:26:11 -07:00
test_batch_file_validation.py	Merge remote-tracking branch 'origin/litellm_internal_staging' into litellm_batch_enqueued_token_limit	2026-08-19 19:31:46 -07:00
test_batch_rate_limiter.py	fix(rate_limiter): render the 429 reset time in UTC as labelled	2026-09-19 00:40:51 +00:00
test_dynamic_rate_limiter.py	fix(dynamic_rate_limiter): inject clock so active-project window is stable within a request (#32299)	2026-07-06 18:12:47 -07:00
test_dynamic_rate_limiter_v3.py	fix(rate_limiter): skip non-Latin-1 x-litellm-priority header on /v1/messages (#40636)	2026-09-10 17:48:46 -07:00
test_image_generation_guardrails.py	test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)	2026-08-22 09:25:58 -07:00
test_key_management_event_hooks.py	fix(proxy): write key deleted audit logs for cascade and alias key deletions (#42446)	2026-09-22 11:52:08 -07:00
test_max_budget_per_session_limiter.py	fix(redis): quiet every per-request Redis fallback while the breaker is open	2026-09-10 14:18:53 -07:00
test_max_iterations_limiter.py	Agents - add max budget + tpm/rpm limiting per agent AND per agent session (#22849)	2026-03-07 19:12:42 -08:00
test_model_max_budget_limiter.py	fix(proxy): claim a finished batch's per-model budget charge atomically	2026-09-19 17:41:19 -07:00
test_parallel_request_limiter.py	fix(spend): attribute CLI session spend to the per-user cli-session alias instead of the hashed session token (#40541)	2026-09-24 18:21:47 -07:00
test_parallel_request_limiter_v3.py	feat(proxy): add fail_closed_rate_limit_enforcement to reject requests with 503 while Redis rate limit counters are unreachable (#43251)	2026-09-26 12:11:26 -07:00
test_post_call_failure_hook_integration.py	test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)	2026-08-22 09:25:58 -07:00
test_post_call_response_headers_hook.py	test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)	2026-08-22 09:25:58 -07:00
test_post_call_streaming_hook_integration.py	test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)	2026-08-22 09:25:58 -07:00
test_post_call_success_hook_integration.py	test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)	2026-08-22 09:25:58 -07:00
test_prompt_cache_observer.py	feat(proxy): predict prompt-cache costs across deployments	2026-09-12 14:02:45 -07:00
test_prompt_injection_detection.py	merge: origin/main into litellm_prompt_injection_llm_api_check_dispatch	2026-09-18 07:43:34 +00:00
test_proxy_hooks_init.py	[Fix] Proxy: Address Greptile feedback on hook-cycle PR	2026-05-04 20:13:31 -07:00
test_proxy_rate_limit_provider_field.py	fix(proxy): remove duplicate user budget hook that 429'd zero-cost models	2026-09-16 15:42:15 -07:00
test_proxy_track_cost_callback.py	fix(proxy): fail parked DB lookups at a deadline and flip readiness while they stall (#42654)	2026-09-24 10:09:49 -05:00
test_rate_limiter_toctou.py	test: drop the cwd-relative sys.path.insert calls from the test suite (#37802)	2026-08-22 09:25:58 -07:00
test_send_invite_email.py	feat(proxy)!: default audit logs on for enterprise licenses (#37518)	2026-08-19 18:49:21 -07:00
test_sensitive_data_routing.py	fix(redis): quiet every per-request Redis fallback while the breaker is open	2026-09-10 14:18:53 -07:00
test_tpm_concurrent.py	test: move tests/test_litellm core utils, routing, responses, caching and rust_bridge into tests/unit (#43199)	2026-09-25 17:10:13 -07:00
test_user_management_event_hooks.py	test(proxy): cover the user-created audit hook's database read-back	2026-07-31 19:48:30 +00:00