mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-13 03:59:12 +00:00
* feat(model-catalog): add Rust registry validation (#43136) * doc * feat(model-catalog): add Rust registry validation * test(model-catalog): ignore integration tests * docs(model-catalog): fix validation note grammar --------- Co-authored-by: Yujong Lee <yujong@berri.ai> (cherry picked from commitc289d5d6fb) * test(rust): reorganize core crate tests and split cache and OCR suites (#43177) * test(rust): group cache tests under cache/ and fold test_ocr.py into ocr/ The two failure cases in test_ocr.py duplicated the upstream-500 and timeout rows of PUBLIC_FAILURES, so only the file-input encoding case moves to ocr/test_requests.py Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test(rust): split the response cache suite into one file per backend test_response_cache.py grew to 2400 lines. Each backend now has its own file, shared fixtures live in cache/conftest.py and shared helpers in support/cache.py. The helpers alias the private native test handles once, dropping the per-call reportPrivateUsage hits Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * split tokenizer test * test(core): consolidate route integration tests under tests/ with rstest and wiremock Moves the public-API OCR route tests out of src/ocr/route.rs and document.rs into tests/ocr/, split per provider plus lifecycle, machine, and document tests, merging the duplicated pairs. Messages, audio transcription, and chat completions share one wiremock-based upstream and recording secret source in tests/support, and gain table-driven cases for auth, routing, upstream errors, streaming, and declines. Tests of litellm-llms items move to that crate. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test(messages): keep the stream relay test independent of the stream head contents The stream head carries no headers on main, so the relay test asserts the open-then-deliver order and the relayed body instead of header hand-off. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit10413796c6) * feat(rust): hand upstream response headers to the native Messages stream (#43178) * ci: drop the ocr_testing job now that tests/ocr_tests is gone Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test(ocr): restore the live OCR matrix and the ocr_testing job The public litellm.ocr / aocr / Router interface is unchanged by the Rust migration, so the live provider matrix still applies. Drops the stale VCR skip list for the deleted test_rust_bridge.py. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * test(messages): show streamed upstream headers never reach the native stream The Python handler puts the upstream response headers on the stream's _hidden_params before the first chunk so the proxy can forward them as llm_provider-* headers. The native route drops them, and this test fails on the Rust path while passing on Python. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * feat(messages): hand upstream response headers to the native stream before its first chunk The Messages route fills MessagesStreamHead from the upstream response and yields it on Open. The Python driver converts it through the protocol host and hands it to Stream and SyncStream as their _hidden_params, so a streamed native call carries additional_headers the same way the Python handler does and the proxy can forward them as llm_provider-* headers. The relay contract lives in the core crate test, the hand-off in the host-python driver test, and the header projection in the route host test, so the recording-server test that showed the gap is dropped. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * wip --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> (cherry picked from commit081f73f021) * feat(testkit): agent clients for Claude Code, Codex and opencode (#43181) * feat(testkit): install and configure Claude Code, Codex and opencode against a gateway Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * refactor(testkit): derive targets from target-lexicon and split agents behind a trait Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * refactor(testkit): group sources into agent and install folders Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * feat(testkit): split agents into install, configure and drive with semver-aware launch Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * style(testkit): drop a needless borrow Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> (cherry picked from commit9c10e0f985) * ci: run migrated unit selections on every event in legacy GHA shards (#43182) * ci: run the unit_selection.sh shard files on every event instead of only fork pull requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci: rename fork-flag to unit-flag now that it applies on every event --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit694783ebbe) * test: move tests/test_litellm root and small trees into tests/unit (#43186) * ci: run the unit_selection.sh shard files on every event instead of only fork pull requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci: rename fork-flag to unit-flag now that it applies on every event * test: move tests/test_litellm root and small trees into tests/unit Pure renames, no content changes. Follow-up commits in this PR fix references, merge the three files that already existed in tests/unit, keep live-provider tests in tests/test_litellm and wire CI. * test: carry tests/test_litellm conftest isolation into tests/unit Callback lists, routing fallbacks, cached HTTP clients, logger state, AWS, proxy-URL and keychain env, and session-end client cleanup now reset for unit tests too. The environment isolation owns its MonkeyPatch so a test's own monkeypatch is undone before the model-cost teardown runs. * test: merge, split and prune the moved root and small-tree tests Merge batches/test_batch_utils.py and the chat_completions and messages dispatch tests into the files that already existed in tests/unit. Keep the live Gemini interactions tests, the async image-fetch format test and the OpenAI embedding scorer test in tests/test_litellm since they need real network or keys. Put test_router.py under tests/unit/test_router so the existing package no longer shadows it. Delete eight tests the audit found superseded by stronger ones kept in this move. * ci: run the moved root and small-tree tests under their legacy flags Add the misc and responses-caching-types flags to unit_selection.sh and CircleCI, extend enterprise-routing and mcp-integration, and point the legacy GHA shards, Makefile, redis-compat workflow, merge smoke manifest and change classifier at the new paths. * test: make the new tests/unit directories packages tests/unit/test_package_layout.py requires every directory to carry an __init__.py, and without one the moved and retained test_litellm_responses_bridge.py modules collide on import. * test: scope the unit socket block to tests/unit in shared sessions The GHA shards collect the legacy test-path and the unit selection in one pytest session. The unit conftest's loopback-only block leaked into legacy modules that reach the network at import. The legacy conftest now lifts the restriction at collect and setup time, and the unit conftest re-applies it when collecting its own modules. * test: give the shard-script tests their own GITHUB_OUTPUT They only passed where the runner set it. The CircleCI unit job's env allowlist drops it, so the script's redirect failed there. * test: point the router and module-deletion checks at tests/unit router_code_coverage and code_qa_check_tests only searched tests/test_litellm, so the moved router tests no longer counted. The two silent-experiment tests the audit deleted were the only direct callers of those methods; they are replaced with tests that assert the forwarded shadow request and the recursion guard. --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commitf6882246d4) * refactor(framer): replace Framer trait with tokio-util codecs (#43193) * refactor(framer): replace Framer trait with tokio-util codecs Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * refactor(framer): port SSE and AWS event stream framing to codecs Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(rust): drop clone on Copy capabilities in messages request test Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(rust): use field init shorthand in messages request test Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit3c93ea1697) * test: move tests/test_litellm/llms into tests/unit/llms (#43191) * ci: run the unit_selection.sh shard files on every event instead of only fork pull requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci: rename fork-flag to unit-flag now that it applies on every event * test: move tests/test_litellm root and small trees into tests/unit Pure renames, no content changes. Follow-up commits in this PR fix references, merge the three files that already existed in tests/unit, keep live-provider tests in tests/test_litellm and wire CI. * test: carry tests/test_litellm conftest isolation into tests/unit Callback lists, routing fallbacks, cached HTTP clients, logger state, AWS, proxy-URL and keychain env, and session-end client cleanup now reset for unit tests too. The environment isolation owns its MonkeyPatch so a test's own monkeypatch is undone before the model-cost teardown runs. * test: merge, split and prune the moved root and small-tree tests Merge batches/test_batch_utils.py and the chat_completions and messages dispatch tests into the files that already existed in tests/unit. Keep the live Gemini interactions tests, the async image-fetch format test and the OpenAI embedding scorer test in tests/test_litellm since they need real network or keys. Put test_router.py under tests/unit/test_router so the existing package no longer shadows it. Delete eight tests the audit found superseded by stronger ones kept in this move. * ci: run the moved root and small-tree tests under their legacy flags Add the misc and responses-caching-types flags to unit_selection.sh and CircleCI, extend enterprise-routing and mcp-integration, and point the legacy GHA shards, Makefile, redis-compat workflow, merge smoke manifest and change classifier at the new paths. * test: make the new tests/unit directories packages tests/unit/test_package_layout.py requires every directory to carry an __init__.py, and without one the moved and retained test_litellm_responses_bridge.py modules collide on import. * test: scope the unit socket block to tests/unit in shared sessions The GHA shards collect the legacy test-path and the unit selection in one pytest session. The unit conftest's loopback-only block leaked into legacy modules that reach the network at import. The legacy conftest now lifts the restriction at collect and setup time, and the unit conftest re-applies it when collecting its own modules. * test: move tests/test_litellm/llms into tests/unit/llms Rename-only. Moves the provider tests and the fine-tuning fixtures they load, mirroring the old paths. Follow-up commits merge, split and wire them. * test: merge, split and prune the moved llms tests Merges the Databricks chat transformation tests into the existing unit file, keeps the tests that need real keys or the network in tests/test_litellm, deletes the audited tests a stronger unit test already covers, and points imports at tests.unit.llms. * ci: run the moved llms tests under their legacy flags The Vertex AI and All Other Providers shards keep their legacy test-path for the retained files and add the llm-vertex-ai and llm-other-providers unit selections. CircleCI gets matching unit jobs. * test: make the tests/unit/llms directories packages Adds __init__.py to the moved dirs and drops the legacy ones whose directories no longer hold tests. * test: drop script runners and path hacks the llms split left dangling The __main__ runners in the split openai_like files and the Databricks e2e runner called tests that now live in the other half of the split or were deleted. The retained legacy halves also no longer need sys.path edits. * test: give the shard-script tests their own GITHUB_OUTPUT They only passed where the runner set it. The CircleCI unit job's env allowlist drops it, so the script's redirect failed there. * test: point the router and module-deletion checks at tests/unit router_code_coverage and code_qa_check_tests only searched tests/test_litellm, so the moved router tests no longer counted. The two silent-experiment tests the audit deleted were the only direct callers of those methods; they are replaced with tests that assert the forwarded shadow request and the recursion guard. * test: keep the Databricks manual e2e runner and fix the SageMaker Nova run path The Databricks e2e file is a manual script whose main() calls the tests that were pruned, so pruning them broke the documented run. It is back to its main version. The SageMaker Nova docstring now points at the file's real location in tests/local_testing. * test: keep the job's UNIT_FLAG out of the shard-script tests --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit5e6dc89ba1) * test: move tests/test_litellm integrations and secret_managers into tests/unit (#43194) * ci: run the unit_selection.sh shard files on every event instead of only fork pull requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci: rename fork-flag to unit-flag now that it applies on every event * test: move tests/test_litellm root and small trees into tests/unit Pure renames, no content changes. Follow-up commits in this PR fix references, merge the three files that already existed in tests/unit, keep live-provider tests in tests/test_litellm and wire CI. * test: carry tests/test_litellm conftest isolation into tests/unit Callback lists, routing fallbacks, cached HTTP clients, logger state, AWS, proxy-URL and keychain env, and session-end client cleanup now reset for unit tests too. The environment isolation owns its MonkeyPatch so a test's own monkeypatch is undone before the model-cost teardown runs. * test: merge, split and prune the moved root and small-tree tests Merge batches/test_batch_utils.py and the chat_completions and messages dispatch tests into the files that already existed in tests/unit. Keep the live Gemini interactions tests, the async image-fetch format test and the OpenAI embedding scorer test in tests/test_litellm since they need real network or keys. Put test_router.py under tests/unit/test_router so the existing package no longer shadows it. Delete eight tests the audit found superseded by stronger ones kept in this move. * ci: run the moved root and small-tree tests under their legacy flags Add the misc and responses-caching-types flags to unit_selection.sh and CircleCI, extend enterprise-routing and mcp-integration, and point the legacy GHA shards, Makefile, redis-compat workflow, merge smoke manifest and change classifier at the new paths. * test: make the new tests/unit directories packages tests/unit/test_package_layout.py requires every directory to carry an __init__.py, and without one the moved and retained test_litellm_responses_bridge.py modules collide on import. * test: scope the unit socket block to tests/unit in shared sessions The GHA shards collect the legacy test-path and the unit selection in one pytest session. The unit conftest's loopback-only block leaked into legacy modules that reach the network at import. The legacy conftest now lifts the restriction at collect and setup time, and the unit conftest re-applies it when collecting its own modules. * test: move tests/test_litellm/llms into tests/unit/llms Rename-only. Moves the provider tests and the fine-tuning fixtures they load, mirroring the old paths. Follow-up commits merge, split and wire them. * test: merge, split and prune the moved llms tests Merges the Databricks chat transformation tests into the existing unit file, keeps the tests that need real keys or the network in tests/test_litellm, deletes the audited tests a stronger unit test already covers, and points imports at tests.unit.llms. * ci: run the moved llms tests under their legacy flags The Vertex AI and All Other Providers shards keep their legacy test-path for the retained files and add the llm-vertex-ai and llm-other-providers unit selections. CircleCI gets matching unit jobs. * test: make the tests/unit/llms directories packages Adds __init__.py to the moved dirs and drops the legacy ones whose directories no longer hold tests. * test: drop script runners and path hacks the llms split left dangling The __main__ runners in the split openai_like files and the Databricks e2e runner called tests that now live in the other half of the split or were deleted. The retained legacy halves also no longer need sys.path edits. * test: give the shard-script tests their own GITHUB_OUTPUT They only passed where the runner set it. The CircleCI unit job's env allowlist drops it, so the script's redirect failed there. * test: point the router and module-deletion checks at tests/unit router_code_coverage and code_qa_check_tests only searched tests/test_litellm, so the moved router tests no longer counted. The two silent-experiment tests the audit deleted were the only direct callers of those methods; they are replaced with tests that assert the forwarded shadow request and the recursion guard. * test: move tests/test_litellm integrations and secret_managers into tests/unit Rename-only. Mirrors the old paths, including the directory conftests and the prompt and JSON fixtures. Follow-up commits prune and wire them. * test: prune and repoint the moved integrations tests Deletes the 7 audited tests a stronger test in the same tree already covers, imports the TLS sink helpers from their new conftest path, and restores os.environ after each integrations test. Some presets write OTEL_EXPORTER_OTLP_HEADERS straight into os.environ, and without the legacy tree's test ordering that header leaked into the AgentOps tests. * ci: run the moved integrations tests under their legacy flag The integrations GHA shard and a new CircleCI job run the integrations unit selection. secret_managers joins the misc selection. * docs: point integrations and secret_managers references at tests/unit * test: make the moved integrations directories packages * test: keep the Databricks manual e2e runner and fix the SageMaker Nova run path The Databricks e2e file is a manual script whose main() calls the tests that were pruned, so pruning them broke the documented run. It is back to its main version. The SageMaker Nova docstring now points at the file's real location in tests/local_testing. * test: keep the job's UNIT_FLAG out of the shard-script tests --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commitcf491d1df9) * fix(tests): drop the repeated UNIT_FLAG key in test_unit_shard_missing_paths (#43212) Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> (cherry picked from commit25de1ab2b0) * fix(jwt): let x-litellm-team-id select DB membership teams when the token also carries a team claim (#43206) * fix(jwt): let x-litellm-team-id select DB membership teams when the token also carries a team claim Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * docs(jwt): describe header team selection under fallback_to_db_teams Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: yassin <yassin@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit7b4fd47c6e) * test: move tests/test_litellm core utils, routing, responses, caching and rust_bridge into tests/unit (#43199) * ci: run the unit_selection.sh shard files on every event instead of only fork pull requests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * ci: rename fork-flag to unit-flag now that it applies on every event * test: move tests/test_litellm root and small trees into tests/unit Pure renames, no content changes. Follow-up commits in this PR fix references, merge the three files that already existed in tests/unit, keep live-provider tests in tests/test_litellm and wire CI. * test: carry tests/test_litellm conftest isolation into tests/unit Callback lists, routing fallbacks, cached HTTP clients, logger state, AWS, proxy-URL and keychain env, and session-end client cleanup now reset for unit tests too. The environment isolation owns its MonkeyPatch so a test's own monkeypatch is undone before the model-cost teardown runs. * test: merge, split and prune the moved root and small-tree tests Merge batches/test_batch_utils.py and the chat_completions and messages dispatch tests into the files that already existed in tests/unit. Keep the live Gemini interactions tests, the async image-fetch format test and the OpenAI embedding scorer test in tests/test_litellm since they need real network or keys. Put test_router.py under tests/unit/test_router so the existing package no longer shadows it. Delete eight tests the audit found superseded by stronger ones kept in this move. * ci: run the moved root and small-tree tests under their legacy flags Add the misc and responses-caching-types flags to unit_selection.sh and CircleCI, extend enterprise-routing and mcp-integration, and point the legacy GHA shards, Makefile, redis-compat workflow, merge smoke manifest and change classifier at the new paths. * test: make the new tests/unit directories packages tests/unit/test_package_layout.py requires every directory to carry an __init__.py, and without one the moved and retained test_litellm_responses_bridge.py modules collide on import. * test: scope the unit socket block to tests/unit in shared sessions The GHA shards collect the legacy test-path and the unit selection in one pytest session. The unit conftest's loopback-only block leaked into legacy modules that reach the network at import. The legacy conftest now lifts the restriction at collect and setup time, and the unit conftest re-applies it when collecting its own modules. * test: move tests/test_litellm/llms into tests/unit/llms Rename-only. Moves the provider tests and the fine-tuning fixtures they load, mirroring the old paths. Follow-up commits merge, split and wire them. * test: merge, split and prune the moved llms tests Merges the Databricks chat transformation tests into the existing unit file, keeps the tests that need real keys or the network in tests/test_litellm, deletes the audited tests a stronger unit test already covers, and points imports at tests.unit.llms. * ci: run the moved llms tests under their legacy flags The Vertex AI and All Other Providers shards keep their legacy test-path for the retained files and add the llm-vertex-ai and llm-other-providers unit selections. CircleCI gets matching unit jobs. * test: make the tests/unit/llms directories packages Adds __init__.py to the moved dirs and drops the legacy ones whose directories no longer hold tests. * test: drop script runners and path hacks the llms split left dangling The __main__ runners in the split openai_like files and the Databricks e2e runner called tests that now live in the other half of the split or were deleted. The retained legacy halves also no longer need sys.path edits. * test: give the shard-script tests their own GITHUB_OUTPUT They only passed where the runner set it. The CircleCI unit job's env allowlist drops it, so the script's redirect failed there. * test: point the router and module-deletion checks at tests/unit router_code_coverage and code_qa_check_tests only searched tests/test_litellm, so the moved router tests no longer counted. The two silent-experiment tests the audit deleted were the only direct callers of those methods; they are replaced with tests that assert the forwarded shadow request and the recursion guard. * test: move tests/test_litellm integrations and secret_managers into tests/unit Rename-only. Mirrors the old paths, including the directory conftests and the prompt and JSON fixtures. Follow-up commits prune and wire them. * test: prune and repoint the moved integrations tests Deletes the 7 audited tests a stronger test in the same tree already covers, imports the TLS sink helpers from their new conftest path, and restores os.environ after each integrations test. Some presets write OTEL_EXPORTER_OTLP_HEADERS straight into os.environ, and without the legacy tree's test ordering that header leaked into the AgentOps tests. * ci: run the moved integrations tests under their legacy flag The integrations GHA shard and a new CircleCI job run the integrations unit selection. secret_managers joins the misc selection. * docs: point integrations and secret_managers references at tests/unit * test: make the moved integrations directories packages * test: keep the Databricks manual e2e runner and fix the SageMaker Nova run path The Databricks e2e file is a manual script whose main() calls the tests that were pruned, so pruning them broke the documented run. It is back to its main version. The SageMaker Nova docstring now points at the file's real location in tests/local_testing. * test: move tests/test_litellm core utils, routing, responses, caching and rust_bridge into tests/unit Rename-only. Mirrors the old paths, including fixtures, the stubtest config and the native-route wheel script. Two files that collide with existing unit files are merged in a follow-up commit. * test: merge, prune and repoint the moved core, routing, responses, caching and rust_bridge tests Merges the two files that collided with existing unit files, folding the legacy extra case into test_is_chat_completion_cached_dict, and deletes the 9 audited tests a stronger test in the same file already covers. Keeps what needs the network in tests/test_litellm: test_tokenizers pulls a tokenizer from the Hugging Face hub, and the gpt2 and r50k_base tokenizer cases download their BPE files. The unit core_utils conftest points TIKTOKEN_CACHE_DIR at litellm's bundled encodings so the rest never depend on import order to stay offline, and FakeSecretVault moves to a shared module so both trees can build it. * ci: run the moved core, routing, responses, caching and rust_bridge tests under their flags core_utils gets a core-utils flag and CircleCI job, and its GHA shard keeps the legacy path for the retained network tests. router_utils and router_strategy join enterprise-routing, responses joins responses-caching-types (minus responses/mcp, which mcp-integration owns), caching joins caching-local and rust_bridge joins misc. The redis-compat, test-rust, stubtest and merge-smoke paths follow the move. * docs: point the Rust crate references at tests/unit * test: make the moved core, routing and rust_bridge directories packages * test: keep the no-loop DualCache batch_get_cache regression test It runs the sync path outside any event loop, which the inside-loop test cannot, so a change that picks the Redis client by loop state would only show up there. * test: keep the job's UNIT_FLAG out of the shard-script tests * fix(url_utils): block 192.0.0.0/24 on every Python patch release * test: move the new budget limiter tests into tests/unit/router_strategy * test: move the new sentry scrubbing tests into tests/unit/litellm_core_utils * test: move the new zerobus tests into tests/unit/integrations * test: make tests/unit/integrations/zerobus a package * test: load litellm's own tiktoken cache setup once instead of resetting it per test --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commita11a93f44a) * refactor(rust): move credential inheritance and the SDK limits out of the legacy callback crate into a driver preflight (#43259) The legacy callback crate carried two rewrites that have nothing to do with the Logging contract: litellm_credential_name inheritance and the max_budget and num_retries_per_request checks. Any later callback host would need them unchanged, which is the smell the crate's AGENTS.md now names. They are now a Preflight the driver in litellm-host-python runs on the keyword view begin returned, before the host projects from it, supplied by python-bridge and passed through run_legacy_call. The call order is unchanged (setup, deployment hook, credentials, limits) and a rejection still fails the call as a host failure, so the failure callbacks run as before. The preflight rewrites the adapter's own copy in place, so no extra dict copy and no new lifecycle method Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> (cherry picked from commite9491d31b5) * refactor(http): hand out an owned Client and route all providers through the pool (#43245) * refactor(messages): take the provider client from the injected HTTP pool The messages route kept its own process-wide reqwest client, so it ignored ssl_verify, CA bundles, client certs, proxies and every other setting that litellm-http resolves. The machine now takes the HttpClientPool and the call's HttpClientConfig, as OCR does, and the bridge passes its shared pool. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * refactor(http): hand out an owned Client and move chat, audio and OIDC onto the pool HttpClientPool now returns litellm_http::Client, a newtype only crates/http can build, so every provider client carries the resolved TLS, proxy and timeout settings. Chat completions and audio transcription drop their process-wide reqwest clients and take the pool and call config like messages; their 600s ceiling moves to the request. OidcResolver takes its client instead of building one, and the bridge hands it the pooled one. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * refactor(secrets): build Google, Azure and CyberArk manager clients from the pool The native secret managers built bare reqwest clients, so they ignored the host's TLS and proxy settings. load_native_manager now takes the pool and the host config and hands each manager a pooled client. CyberArk's CYBERARK_SSL_VERIFY and CYBERARK_CLIENT_CERT/KEY become an override on the host config instead of a hand-built client. To express a certificate and key in separate files, HttpClientConfig::client_certificate is now a ClientIdentity that is either one PEM or a split pair. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * chore(clippy): only crates/http may build a reqwest client Fence reqwest::Client, ClientBuilder and the TLS builder methods with disallowed-types and disallowed-methods so new code takes a litellm_http::Client from the pool. crates/http is exempt as the one place clients are built, and testkit as a dev-only installer. Tests move to litellm_http::Client::plain_for_test or a pooled client. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(secrets-cyberark): keep verifying certificates when the host disables it Python hands CyberArk its own ssl_verify, which wins over the global setting, so CYBERARK_SSL_VERIFY unset or true still verifies even when the host sets ssl_verify false. The pooled client copied the host's Disabled and would send the API key unverified; fall back to the built-in roots instead. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * fix(python-bridge): treat a missing litellm package as no host HTTP settings Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit1a58162630) * refactor(rust): promote anthropic messages out of experimental_pass_through (#43269) Co-authored-by: Yujong Lee <yujong@berri.ai> (cherry picked from commit2ef3250ec3) * test: stop CI tests from downloading tokenizer files and images (#43257) * test: load the embedding base image from a committed 100x100 PNG instead of downloading it * test: move the volcengine embedding test into tests/unit * test: check gpt2 and r50k_base tokenizer parity against committed tiktoken reference files * test: check hub tokenizer selection against an in-memory Hugging Face hub * test: serve image URLs from respx in the gemini tool-result and format-param tests * ci: drop the emptied legacy core-utils test path * test: cover the cohere and anthropic tokenizer paths in the hub tokenizer test * test: fetch every format-param image through respx and check its bytes reach the request * test: drop the gpt2 and r50k_base parity tests, which no litellm path uses * test: drop comments that restate assertions in the format-param test (cherry picked from commit2530255624) * fix(rust): preserve nested optional import failures (#43265) * fix(rust): preserve nested optional import failures Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(rust): restore Python modules after settings tests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commite1a9378093) * fix(callbacks-legacy-python): traverse and release the retained headers dict (#43274) * fix(callbacks-legacy-python): traverse and release the retained headers dict LegacyLogging keeps the headers dict it hands to pre_call and post_call, but its traverse never reported that edge to the collector and close never dropped it. A cycle a callback builds through that dict could not be collected, and a closed call kept the dict alive until the driver dropped the whole adapter. Visit and clear headers like body, with regression tests for both * refactor(callbacks-legacy-python): move the test support module into its own file --------- Co-authored-by: Yujong Lee <yujong@berri.ai> (cherry picked from commit4eb340abe0) * test(integration): port langfuse callbacks-in-db coverage to the local harness (#43282) * test(integration): port langfuse callbacks-in-db coverage to the local harness Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(integration): assert the langfuse db rows after the exported span Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commite11c3f5815) * test: finish the non-proxy half of tests/test_litellm (#43281) * test: move key-gated tests/test_litellm SDK tests into tests/llm_translation and drop empty folders * test: make token counter and health check unit tests run offline * ci: point unit shards, rust path filter, Makefile and docs at tests/unit * docs: fix stale test_litellm run paths in moved llm_translation tests * fix: correct databricks e2e sys.path depth and contributing example path --------- Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit99655b6f86) * refactor(rust): prepare inference and auth foundations for the gateway (#43287) * refactor(rust): prepare inference and auth foundations * fix(rust): keep textract operations parsing from kebab-case model names Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit7ae721bf79) * feat(rust): add config router and gateway crates (#43289) Co-authored-by: Yujong Lee <yujong@berri.ai> (cherry picked from commitaffb547525) * test(integration): run the Langfuse DB-callback test on its own scratch database (#43288) * test(integration): run the Langfuse DB-callback test on its own scratch database The test from #43282 wrote success_callback=langfuse and the LANGFUSE_* env into the shared integration LiteLLM_Config. The suite's long-running gateway reloads that table and only ever adds callbacks, so it kept exporting to the test's closed Langfuse fake for the rest of the shard even after the rows were restored. The owned proxy now gets a scratch database, which also removes the snapshot/restore code. scratch_database moves into _support/database.py so test_cache_and_quota and this test share one copy, and the stock-config guard now checks the callback settings instead of the raw YAML text. * test(integration): include failure_callback in the stock-config Langfuse guard (cherry picked from commitdd63637322) * refactor(rust): expand logging and test coverage across gateway and Anthropic messages (#43295) * refactor(rust): prepare inference and auth foundations * fix(rust): keep textract operations parsing from kebab-case model names Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * done * refactor(types): derive Anthropic beta string conversions with Strum * fix(anthropic): report missing max_tokens as a missing field * refactor(rust): type Anthropic messages headers and auth after the Python layout Delete anthropic/messages/headers.rs. Its OAuth handling, credential ladder and beta merging move to anthropic/common_utils.rs where Python keeps them (optionally_handle_anthropic_oauth, get_auth_header, _merge_beta_headers), and the feature beta injection becomes update_headers_with_anthropic_beta on the messages config, as in Python. The BaseAnthropicMessagesConfig impl is unchanged apart from the bodies of validate_environment and request_headers Beta values are now the AnthropicBeta enum and BetaSet, which sort, dedupe and comma-join by construction. Request params gain typed speed, tools and context_management through Recognized, so the beta logic matches on enums instead of string-comparing JSON. OauthToken parses the sk-ant-oat token once and the chat config shares that detection instead of its own copy Case-insensitive header helpers move next to has_header in litellm-http. One deliberate divergence: a Bearer-prefixed OAuth key configured through api_key or ANTHROPIC_API_KEY is sent with a single Bearer scheme, where Python would emit "Bearer Bearer" Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * done * fix(rust): repair test compilation and clippy failures resolve auth before building the outbound request in prepare tests, give the host hook tests their own error type, and drop the disallowed reqwest client and err().expect() from core tests Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> (cherry picked from commit90873c46de) * test(proxy_behavior): scope the management proxy fixture to its package so its spend monitor cannot race the spend tests (#43302) Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit115668f43e) * fix(ci): stop stale CI reds, keep unit tests off the host env, retry CyberArk policy conflicts (#43294) * fix(ci): stop five stale or flaky CI reds and retry CyberArk policy-load conflicts The Langfuse redaction unit test exports to a local OTLP capture instead of polling Langfuse Cloud through a recorded lookup. The passthrough worker-kill test only requires spend rows for requests the surviving worker served. The spend-routes sweep treats the intentional /spend/capture_rate 503 as expected. CyberArk retries a 409 policy load in Python, Rust and the e2e Conjur helper instead of reading it as "variable exists". The integration egress guard now matches the script's own cgroup, so it no longer blocks the CircleCI agent, which runs as the same user. * fix(ci): keep the policy-load backoff typed as float * fix(ci): retry CyberArk policy loads without blocking the event loop and tighten the worker-kill and Langfuse tests * fix(secrets): load CyberArk policy one request at a time per manager * test(secrets): pin that non-conflict CyberArk policy failures are not retried * test(unit): run tests/unit with only an allowlisted host environment CircleCI's unit job inherits every project env var, so real provider keys, REDIS_HOST, DATABASE_URL and AWS or Azure credentials reached tests that assume none are set. Locally, litellm's import-time load_dotenv did the same from any .env up the tree. The unit conftest now drops every variable outside a small allowlist and disables dotenv before litellm is imported. * test(e2e): name a failed search and the stuck batch status instead of misattributing them The websearch session test read an empty web_search_tool_result_error block as a successful search, so a failing search tool surfaced as a session billing bug. The batch cancellation timeout now reports the last status the proxy returned. * fix(ci): scrub the host environment per unit test instead of for the whole pytest process GHA shards run tests/unit next to other suites in one process, so the import-time scrub deleted MCP_TEST_PEER_PYTHON before tests/mcp_tests read it and the MCP upstream fell back to the SDK2 interpreter. The two websearch tests that called OpenAI and Perplexity live are removed: tests/unit no longer sees their keys. * fix(ci): scrub only the host variables present before litellm is imported The per-test scrub also deleted TIKTOKEN_CACHE_DIR, which litellm sets at import to its bundled encodings, so tokenizer paths tried to download them and hit the socket guard. The prisma setup test now passes its own database URL instead of reading one another test leaked into the process environment. * fix(ci): stop the order-dependent unit reds and settle logging tasks on their own queue LoggingWorker marked a task done on whichever queue was current when the callback finished, so a callback that outlived an event-loop change raised "task_done() called too many times" or undercounted the new loop's queue. It now settles the queue the task came from. The rest are test isolation fixes for failures that only appeared when another file ran first on the same xdist worker: a replaced user_api_key_cache, breaker metrics unregistered by prometheus tests, semantic_router's health-check filter on uvicorn.access, logging tasks carried over from bedrock tests, a Router-written model_cost entry, and a stray post captured by the langflow test. The token counter check now asserts bounded chunking instead of wall-clock time. * test(e2e/ui): wait for the logout redirect before visiting a protected page Logout revokes the session server-side before clearing cookies and navigating, so an immediate page.goto either ran with the cookie still set or was aborted by the logout redirect (net::ERR_ABORTED). * test(unit): restore the prometheus metrics config per test and settle logs carried from earlier tests in the a2a cost tests * test(router): pin the router clock in the usage counter tests so a minute rollover cannot empty the read * test(e2e/ui): wait for logout to clear the token cookie instead of for a login redirect * test(integration/mcp): answer the model-info probe another test's proxy sends to the model double (cherry picked from commit14f4c34c61) * test(e2e): assert only litellm-owned batch behavior and move the blank S3 env pin to an integration test (#43321) (cherry picked from commite53e67ede5) * ci: drop main and litellm_* branch filters from the CircleCI litellm-main workflows (#43272) Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commitc822c7fffa) * test(e2e): accept the otel cost write as a linked root trace (#42931) * test(e2e): accept the otel cost write as a linked root trace Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(e2e): window and fail-closed the linked otel trace read-back Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * test(e2e): poll the otel read-back without recursion Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> --------- Co-authored-by: yuneng <yuneng@berri.ai> Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> (cherry picked from commit3ebf6a1fd8) * refactor(rust): exclude gateway and config crates from rc backport * refactor(rust): exclude router crate from rc backport --------- Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: Yujong Lee <yujong@berri.ai> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com> Co-authored-by: yassin <yassin@berri.ai>
1945 lines
74 KiB
Python
1945 lines
74 KiB
Python
import inspect
|
|
import json
|
|
from types import SimpleNamespace
|
|
from unittest.mock import AsyncMock, MagicMock, patch
|
|
|
|
import pytest
|
|
import redis
|
|
import redis.asyncio as async_redis
|
|
from redis.credentials import CredentialProvider
|
|
|
|
import litellm
|
|
from litellm._redis import (
|
|
_AWS_IAM_KWARG_NAMES,
|
|
_coerce_redis_kwargs_types,
|
|
_credential_provider_auth_kwargs,
|
|
_get_redis_client_logic,
|
|
_get_redis_cluster_kwargs,
|
|
_get_redis_env_kwarg_mapping,
|
|
_get_redis_kwargs,
|
|
_get_redis_url_kwargs,
|
|
_pretty_print_redis_config,
|
|
_uses_tls,
|
|
get_redis_async_client,
|
|
get_redis_client,
|
|
get_redis_connection_pool,
|
|
get_redis_url_from_environment,
|
|
)
|
|
from litellm._redis_credential_provider import (
|
|
AzureADCredentialProvider,
|
|
ElastiCacheIAMCredentialProvider,
|
|
GCPIAMCredentialProvider,
|
|
_token_cache,
|
|
)
|
|
from litellm.caching.redis_cache import RedisCache
|
|
from litellm.caching.redis_cluster_cache import RedisClusterCache
|
|
from litellm.constants import REDIS_CLUSTER_HEALTH_CHECK_INTERVAL
|
|
from litellm.proxy._types import CoordinationRedisParams
|
|
|
|
|
|
class _StubCredentialProvider(CredentialProvider):
|
|
def __init__(self, token: str = "stub-token") -> None:
|
|
self._token = token
|
|
|
|
def get_credentials(self):
|
|
return (self._token,)
|
|
|
|
async def get_credentials_async(self):
|
|
return (self._token,)
|
|
|
|
|
|
class _HostileCredentialProvider(CredentialProvider):
|
|
def __init__(self, secret: str) -> None:
|
|
self._payload = secret
|
|
|
|
def get_credentials(self):
|
|
return (self._payload,)
|
|
|
|
async def get_credentials_async(self):
|
|
return (self._payload,)
|
|
|
|
def __repr__(self):
|
|
raise AssertionError("provider repr must never be invoked")
|
|
|
|
def __str__(self):
|
|
raise AssertionError("provider str must never be invoked")
|
|
|
|
def __reduce__(self):
|
|
raise AssertionError("provider must never be serialized")
|
|
|
|
def __getstate__(self):
|
|
raise AssertionError("provider state must never be inspected")
|
|
|
|
|
|
def _gcp_marker_callback() -> MagicMock:
|
|
callback = MagicMock()
|
|
callback._gcp_service_account = "projects/-/serviceAccounts/sa@project.iam.gserviceaccount.com"
|
|
return callback
|
|
|
|
|
|
@pytest.fixture
|
|
def clean_redis_environment(monkeypatch):
|
|
for var in (
|
|
"REDIS_URL",
|
|
"REDIS_CLUSTER_NODES",
|
|
"REDIS_SENTINEL_NODES",
|
|
"AWS_REGION",
|
|
"AWS_DEFAULT_REGION",
|
|
*_get_redis_env_kwarg_mapping(),
|
|
):
|
|
monkeypatch.delenv(var, raising=False)
|
|
|
|
|
|
@pytest.fixture
|
|
def clear_llm_client_cache():
|
|
litellm.in_memory_llm_clients_cache.flush_cache()
|
|
yield
|
|
litellm.in_memory_llm_clients_cache.flush_cache()
|
|
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def clear_gcp_iam_token_cache():
|
|
"""Reset the module-level GCP IAM token cache between tests."""
|
|
_token_cache.clear()
|
|
yield
|
|
_token_cache.clear()
|
|
|
|
|
|
def test_redis_allowlists_include_credential_provider():
|
|
assert "credential_provider" in _get_redis_kwargs()
|
|
assert "credential_provider" in _get_redis_url_kwargs()
|
|
assert "credential_provider" in _get_redis_cluster_kwargs()
|
|
|
|
|
|
def test_credential_provider_is_not_environment_derived():
|
|
mapping = _get_redis_env_kwarg_mapping()
|
|
assert "REDIS_CREDENTIAL_PROVIDER" not in mapping
|
|
assert "credential_provider" not in mapping.values()
|
|
|
|
|
|
_AWS_IAM_SETTINGS = {
|
|
"aws_iam_auth",
|
|
"aws_iam_user_name",
|
|
"aws_iam_cache_name",
|
|
"aws_iam_region",
|
|
"aws_iam_serverless",
|
|
}
|
|
|
|
|
|
def test_aws_iam_settings_are_environment_derived():
|
|
allowed = _get_redis_kwargs()
|
|
mapping = _get_redis_env_kwarg_mapping()
|
|
|
|
assert _AWS_IAM_SETTINGS <= allowed
|
|
assert set(_AWS_IAM_KWARG_NAMES) == _AWS_IAM_SETTINGS
|
|
assert {f for f in CoordinationRedisParams.model_fields if f.startswith("aws_iam_")} == _AWS_IAM_SETTINGS
|
|
assert mapping["REDIS_AWS_IAM_AUTH"] == "aws_iam_auth"
|
|
assert mapping["REDIS_AWS_IAM_USER_NAME"] == "aws_iam_user_name"
|
|
assert mapping["REDIS_AWS_IAM_CACHE_NAME"] == "aws_iam_cache_name"
|
|
assert mapping["REDIS_AWS_IAM_REGION"] == "aws_iam_region"
|
|
assert mapping["REDIS_AWS_IAM_SERVERLESS"] == "aws_iam_serverless"
|
|
|
|
|
|
def test_sync_direct_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
client = get_redis_client(host="redis-host", port=6379, credential_provider=provider)
|
|
|
|
assert client.connection_pool.connection_kwargs["credential_provider"] is provider
|
|
|
|
|
|
def test_sync_direct_provider_supersedes_static_credentials(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
client = get_redis_client(
|
|
host="redis-host",
|
|
port=6379,
|
|
username="redis-user",
|
|
password="redis-password",
|
|
credential_provider=provider,
|
|
)
|
|
connection = client.connection_pool.make_connection()
|
|
|
|
assert connection.credential_provider is provider
|
|
assert connection.username is None
|
|
assert connection.password is None
|
|
|
|
|
|
def test_sync_direct_provider_supersedes_environment_credentials(clean_redis_environment, monkeypatch):
|
|
provider = _StubCredentialProvider()
|
|
monkeypatch.setenv("REDIS_USERNAME", "redis-user")
|
|
monkeypatch.setenv("REDIS_PASSWORD", "redis-password")
|
|
|
|
client = get_redis_client(host="redis-host", port=6379, credential_provider=provider)
|
|
connection = client.connection_pool.make_connection()
|
|
|
|
assert connection.credential_provider is provider
|
|
assert connection.username is None
|
|
assert connection.password is None
|
|
|
|
|
|
def test_sync_url_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
client = get_redis_client(url="redis://redis-host:6379", credential_provider=provider)
|
|
|
|
assert client.connection_pool.connection_kwargs["credential_provider"] is provider
|
|
|
|
|
|
def test_async_direct_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
client = get_redis_async_client(host="redis-host", port=6379, credential_provider=provider)
|
|
|
|
assert client.connection_pool.connection_kwargs["credential_provider"] is provider
|
|
|
|
|
|
def test_async_url_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
client = get_redis_async_client(url="redis://redis-host:6379", credential_provider=provider)
|
|
|
|
assert client.connection_pool.connection_kwargs["credential_provider"] is provider
|
|
|
|
|
|
def test_sync_url_credentials_do_not_replace_explicit_provider(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
client = get_redis_client(
|
|
url="redis://url-user:url-pass@redis-host:6379",
|
|
credential_provider=provider,
|
|
)
|
|
connection = client.connection_pool.make_connection()
|
|
|
|
assert connection.credential_provider is provider
|
|
assert connection.username is None
|
|
assert connection.password is None
|
|
|
|
|
|
def test_async_url_credentials_do_not_replace_explicit_provider(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
client = get_redis_async_client(
|
|
url="redis://url-user:url-pass@redis-host:6379",
|
|
credential_provider=provider,
|
|
)
|
|
connection = client.connection_pool.make_connection()
|
|
|
|
assert connection.credential_provider is provider
|
|
assert connection.username is None
|
|
assert connection.password is None
|
|
|
|
|
|
def test_async_host_port_pool_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
pool = get_redis_connection_pool(host="redis-host", port=6379, credential_provider=provider)
|
|
|
|
assert pool is not None
|
|
assert pool.connection_kwargs["credential_provider"] is provider
|
|
|
|
|
|
def test_async_url_pool_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
pool = get_redis_connection_pool(url="redis://redis-host:6379", credential_provider=provider)
|
|
|
|
assert pool is not None
|
|
assert pool.connection_kwargs["credential_provider"] is provider
|
|
|
|
|
|
def test_async_url_pool_strips_userinfo_for_the_provider(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
pool = get_redis_connection_pool(url="rediss://url-user:url-pass@redis-host:6379/3", credential_provider=provider)
|
|
|
|
connection = pool.make_connection()
|
|
assert connection.credential_provider is provider
|
|
assert connection.username is None
|
|
assert connection.password is None
|
|
assert connection.db == 3
|
|
|
|
|
|
def test_sync_cluster_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
startup_nodes = [{"host": "cluster-node", "port": 6379}]
|
|
|
|
with patch("redis.RedisCluster", autospec=True) as mock_cluster_cls:
|
|
get_redis_client(startup_nodes=startup_nodes, credential_provider=provider, password="redis-secret")
|
|
|
|
cluster_kwargs = mock_cluster_cls.call_args.kwargs
|
|
assert cluster_kwargs["credential_provider"] is provider
|
|
assert "password" not in cluster_kwargs
|
|
assert [(node.host, node.port) for node in cluster_kwargs["startup_nodes"]] == [("cluster-node", 6379)]
|
|
|
|
|
|
def test_sync_cluster_authenticates_with_azure_credentials(clean_redis_environment, monkeypatch):
|
|
monkeypatch.setenv("REDIS_USERNAME", "identity-object-id")
|
|
credential = MagicMock()
|
|
credential.get_token.return_value = SimpleNamespace(token="azure-access-token")
|
|
|
|
with (
|
|
patch("azure.identity.DefaultAzureCredential", return_value=credential),
|
|
patch("redis.RedisCluster", autospec=True) as cluster,
|
|
):
|
|
get_redis_client(
|
|
startup_nodes=[{"host": "cluster-node", "port": 6379}],
|
|
azure_redis_ad_token=True,
|
|
password="stale-password",
|
|
)
|
|
|
|
kwargs = cluster.call_args.kwargs
|
|
provider = kwargs.get("credential_provider")
|
|
assert isinstance(provider, AzureADCredentialProvider)
|
|
assert provider.get_credentials() == ("identity-object-id", "azure-access-token")
|
|
assert "username" not in kwargs
|
|
assert "password" not in kwargs
|
|
assert "redis_connect_func" not in kwargs
|
|
credential.get_token.assert_called_once_with("https://redis.azure.com/.default")
|
|
|
|
|
|
def test_sync_cluster_authenticates_with_gcp_credentials(clean_redis_environment):
|
|
with patch("redis.RedisCluster", autospec=True) as cluster:
|
|
get_redis_client(
|
|
startup_nodes=[{"host": "cluster-node", "port": 6379}],
|
|
redis_connect_func=_gcp_marker_callback(),
|
|
username="stale-user",
|
|
password="stale-password",
|
|
)
|
|
|
|
kwargs = cluster.call_args.kwargs
|
|
assert isinstance(kwargs.get("credential_provider"), GCPIAMCredentialProvider)
|
|
assert "username" not in kwargs
|
|
assert "password" not in kwargs
|
|
assert "redis_connect_func" not in kwargs
|
|
|
|
|
|
def test_async_cluster_preserves_credential_provider_identity(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
startup_nodes = [{"host": "cluster-node", "port": 6379}]
|
|
|
|
client = get_redis_async_client(startup_nodes=startup_nodes, credential_provider=provider)
|
|
|
|
assert client.connection_kwargs["credential_provider"] is provider
|
|
assert client.connection_kwargs["socket_keepalive"] is True
|
|
assert client.connection_kwargs["health_check_interval"] == REDIS_CLUSTER_HEALTH_CHECK_INTERVAL
|
|
|
|
|
|
def test_explicit_provider_skips_automatic_auth_and_callback(clean_redis_environment, monkeypatch):
|
|
provider = _StubCredentialProvider()
|
|
monkeypatch.setenv("REDIS_GCP_SERVICE_ACCOUNT", "service-account@example.com")
|
|
monkeypatch.setenv("REDIS_AZURE_AD_TOKEN", "true")
|
|
|
|
with (
|
|
patch( # test-quality-ok: an auto-auth callback built here is popped again by the provider branch, so the builders are the only place the wasted work is visible
|
|
"litellm._redis.create_gcp_iam_redis_connect_func"
|
|
) as mock_gcp,
|
|
patch( # test-quality-ok: same as above, and reaching this one also builds an Azure credential the caller never asked for
|
|
"litellm._redis.create_azure_ad_redis_connect_func"
|
|
) as mock_azure,
|
|
):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="redis-host",
|
|
port=6379,
|
|
credential_provider=provider,
|
|
redis_connect_func=_gcp_marker_callback(),
|
|
)
|
|
|
|
mock_gcp.assert_not_called()
|
|
mock_azure.assert_not_called()
|
|
assert redis_kwargs["credential_provider"] is provider
|
|
assert "redis_connect_func" not in redis_kwargs
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"overrides",
|
|
[
|
|
{"gcp_ssl_ca_certs": "/tmp/ca.pem"},
|
|
{"gcp_service_account": "sa@example.com", "gcp_ssl_ca_certs": "/tmp/ca.pem"},
|
|
],
|
|
ids=["certs-without-service-account", "both-alongside-a-provider"],
|
|
)
|
|
def test_gcp_kwargs_never_survive_client_logic(clean_redis_environment, overrides):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="redis-host",
|
|
port=6379,
|
|
credential_provider=_StubCredentialProvider() if "gcp_service_account" in overrides else None,
|
|
**overrides,
|
|
)
|
|
|
|
assert "gcp_service_account" not in redis_kwargs
|
|
assert "gcp_ssl_ca_certs" not in redis_kwargs
|
|
|
|
|
|
def test_aws_iam_environment_settings_install_provider(clean_redis_environment, monkeypatch):
|
|
monkeypatch.setenv("REDIS_AWS_IAM_AUTH", "true")
|
|
monkeypatch.setenv("REDIS_AWS_IAM_USER_NAME", "iam-user")
|
|
monkeypatch.setenv("REDIS_AWS_IAM_CACHE_NAME", "cache.example.com")
|
|
monkeypatch.setenv("REDIS_AWS_IAM_REGION", "us-east-1")
|
|
monkeypatch.setenv("REDIS_AWS_IAM_SERVERLESS", "1")
|
|
monkeypatch.setenv("REDIS_SSL", "1")
|
|
|
|
redis_kwargs = _get_redis_client_logic(host="cache.example.com", port=6379)
|
|
|
|
provider = redis_kwargs["credential_provider"]
|
|
assert isinstance(provider, ElastiCacheIAMCredentialProvider)
|
|
assert provider._is_serverless is True
|
|
assert not _AWS_IAM_SETTINGS & redis_kwargs.keys()
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"transport",
|
|
[
|
|
pytest.param({"host": "cache.example.com", "port": 6379}, id="host_without_ssl"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": False}, id="host_ssl_false"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "false"}, id="host_ssl_false_string"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "0"}, id="host_ssl_zero_string"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "no"}, id="host_ssl_no_string"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "off"}, id="host_ssl_off_string"),
|
|
pytest.param({"url": "redis://cache.example.com:6379", "ssl": True}, id="plaintext_url"),
|
|
pytest.param(
|
|
{"startup_nodes": [{"host": "cache.example.com", "port": 6379}]},
|
|
id="cluster_without_ssl",
|
|
),
|
|
pytest.param(
|
|
{
|
|
"url": "rediss://cache.example.com:6379",
|
|
"startup_nodes": [{"host": "cache.example.com", "port": 6379}],
|
|
},
|
|
id="cluster_without_ssl_ignores_url_scheme",
|
|
),
|
|
pytest.param(
|
|
{
|
|
"sentinel_nodes": [("sentinel.example.com", 26379)],
|
|
"service_name": "cache",
|
|
},
|
|
id="sentinel_without_ssl",
|
|
),
|
|
],
|
|
)
|
|
def test_aws_iam_auth_rejects_non_tls_connections(clean_redis_environment, transport):
|
|
with pytest.raises(ValueError, match="requires TLS"):
|
|
_get_redis_client_logic(
|
|
**transport,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"transport",
|
|
[
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": True}, id="host"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "true"}, id="host_ssl_true_string"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "True"}, id="host_ssl_true_capitalized"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "1"}, id="host_ssl_one_string"),
|
|
pytest.param({"host": "cache.example.com", "port": 6379, "ssl": "yes"}, id="host_ssl_yes_string"),
|
|
pytest.param(
|
|
{"startup_nodes": [{"host": "cache.example.com", "port": 6379}], "ssl": "1"},
|
|
id="cluster_ssl_one_string",
|
|
),
|
|
pytest.param({"url": "rediss://cache.example.com:6379"}, id="url"),
|
|
pytest.param(
|
|
{
|
|
"startup_nodes": [{"host": "cache.example.com", "port": 6379}],
|
|
"ssl": True,
|
|
},
|
|
id="cluster",
|
|
),
|
|
pytest.param(
|
|
{
|
|
"sentinel_nodes": [("sentinel.example.com", 26379)],
|
|
"service_name": "cache",
|
|
"ssl": True,
|
|
},
|
|
id="sentinel",
|
|
),
|
|
],
|
|
)
|
|
def test_aws_iam_auth_accepts_tls_connections(clean_redis_environment, transport):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
**transport,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
assert isinstance(redis_kwargs["credential_provider"], ElastiCacheIAMCredentialProvider)
|
|
|
|
|
|
@pytest.mark.parametrize("ssl", ["true", "True", "TRUE", "1", "yes", "YES", "false", "0", "no", "off", "", "maybe"])
|
|
def test_tls_detection_agrees_with_the_ssl_kwarg_coercion(ssl):
|
|
assert _uses_tls({"ssl": ssl}) is _coerce_redis_kwargs_types({"ssl": ssl})["ssl"]
|
|
|
|
|
|
def test_aws_iam_settings_are_removed_for_url_and_static_credentials(clean_redis_environment):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
url="rediss://url-user:url-pass@cache.example.com:6380",
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
username="static-user",
|
|
password="static-password",
|
|
)
|
|
|
|
assert isinstance(redis_kwargs["credential_provider"], ElastiCacheIAMCredentialProvider)
|
|
assert redis_kwargs["url"] == "rediss://cache.example.com:6380"
|
|
assert "username" not in redis_kwargs
|
|
assert "password" not in redis_kwargs
|
|
assert not _AWS_IAM_SETTINGS & redis_kwargs.keys()
|
|
|
|
|
|
@pytest.mark.parametrize("missing", ["aws_iam_user_name", "aws_iam_cache_name", "aws_iam_region"])
|
|
def test_aws_iam_missing_setting_fails_closed(clean_redis_environment, missing):
|
|
settings = {
|
|
"aws_iam_auth": True,
|
|
"aws_iam_user_name": "iam-user",
|
|
"aws_iam_cache_name": "cache.example.com",
|
|
"aws_iam_region": "us-east-1",
|
|
"ssl": True,
|
|
}
|
|
settings[missing] = None
|
|
|
|
with pytest.raises(ValueError, match=missing):
|
|
_get_redis_client_logic(host="cache.example.com", port=6379, **settings)
|
|
|
|
|
|
@pytest.mark.parametrize("region_var", ["AWS_REGION", "AWS_DEFAULT_REGION"])
|
|
def test_aws_iam_region_falls_back_to_environment(clean_redis_environment, monkeypatch, region_var):
|
|
monkeypatch.setenv(region_var, "sa-east-1")
|
|
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
ssl=True,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
)
|
|
|
|
provider = redis_kwargs["credential_provider"]
|
|
assert isinstance(provider, ElastiCacheIAMCredentialProvider)
|
|
assert provider._region == "sa-east-1"
|
|
|
|
|
|
def test_aws_iam_region_prefers_aws_region_over_default_region(clean_redis_environment, monkeypatch):
|
|
monkeypatch.setenv("AWS_REGION", "sa-east-1")
|
|
monkeypatch.setenv("AWS_DEFAULT_REGION", "eu-west-1")
|
|
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
ssl=True,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
)
|
|
|
|
assert redis_kwargs["credential_provider"]._region == "sa-east-1"
|
|
|
|
|
|
def test_aws_iam_region_prefers_explicit_over_environment(clean_redis_environment, monkeypatch):
|
|
monkeypatch.setenv("AWS_REGION", "sa-east-1")
|
|
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
ssl=True,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="explicit-region",
|
|
)
|
|
|
|
assert redis_kwargs["credential_provider"]._region == "explicit-region"
|
|
|
|
|
|
def test_aws_iam_settings_map_to_distinct_provider_fields(clean_redis_environment):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
ssl=True,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user-value",
|
|
aws_iam_cache_name="iam-cache-value",
|
|
aws_iam_region="iam-region-value",
|
|
)
|
|
|
|
provider = redis_kwargs["credential_provider"]
|
|
assert isinstance(provider, ElastiCacheIAMCredentialProvider)
|
|
assert provider._user_name == "iam-user-value"
|
|
assert provider._cache_name == "iam-cache-value"
|
|
assert provider._region == "iam-region-value"
|
|
|
|
|
|
@pytest.mark.parametrize("aws_iam_auth", [None, False, "", "false", "0", "no", "off"])
|
|
def test_aws_iam_auth_disabled_does_not_install_provider(clean_redis_environment, aws_iam_auth):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
aws_iam_auth=aws_iam_auth,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
assert "credential_provider" not in redis_kwargs
|
|
assert not _AWS_IAM_SETTINGS & redis_kwargs.keys()
|
|
|
|
|
|
@pytest.mark.parametrize("aws_iam_auth", [True, "true", "True", "TRUE", "1", "yes"])
|
|
def test_aws_iam_auth_enabled_by_any_truthy_flag(clean_redis_environment, aws_iam_auth):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
ssl=True,
|
|
aws_iam_auth=aws_iam_auth,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache-name",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
assert isinstance(redis_kwargs["credential_provider"], ElastiCacheIAMCredentialProvider)
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"aws_iam_serverless, expected",
|
|
[
|
|
pytest.param(None, False, id="unset"),
|
|
pytest.param(False, False, id="bool_false"),
|
|
pytest.param("false", False, id="string_false"),
|
|
pytest.param("0", False, id="string_zero"),
|
|
pytest.param(True, True, id="bool_true"),
|
|
pytest.param("true", True, id="string_true"),
|
|
pytest.param("1", True, id="string_one"),
|
|
],
|
|
)
|
|
def test_aws_iam_serverless_flag_reaches_the_provider(clean_redis_environment, aws_iam_serverless, expected):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
ssl=True,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache-name",
|
|
aws_iam_region="us-east-1",
|
|
aws_iam_serverless=aws_iam_serverless,
|
|
)
|
|
|
|
provider = redis_kwargs["credential_provider"]
|
|
assert isinstance(provider, ElastiCacheIAMCredentialProvider)
|
|
assert provider._is_serverless is expected
|
|
assert "aws_iam_serverless" not in redis_kwargs
|
|
|
|
|
|
def test_explicit_provider_wins_over_aws_iam(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
credential_provider=provider,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
assert redis_kwargs["credential_provider"] is provider
|
|
assert "aws_iam_auth" not in redis_kwargs
|
|
|
|
|
|
def test_gcp_wins_over_aws_iam(clean_redis_environment):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
gcp_service_account="sa@example.com",
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
assert "credential_provider" not in redis_kwargs
|
|
assert redis_kwargs["redis_connect_func"]._gcp_service_account == "sa@example.com"
|
|
|
|
|
|
def test_azure_wins_over_aws_iam(clean_redis_environment):
|
|
redis_kwargs = _get_redis_client_logic(
|
|
host="cache.example.com",
|
|
port=6379,
|
|
azure_redis_ad_token="true",
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
assert "credential_provider" not in redis_kwargs
|
|
assert redis_kwargs["redis_connect_func"]._azure_redis_ad_token is True
|
|
|
|
|
|
def test_async_cluster_installs_aws_iam_provider(clean_redis_environment):
|
|
startup_nodes = [{"host": "cluster-node", "port": 6379}]
|
|
|
|
client = get_redis_async_client(
|
|
startup_nodes=startup_nodes,
|
|
ssl=True,
|
|
aws_iam_auth=True,
|
|
aws_iam_user_name="iam-user",
|
|
aws_iam_cache_name="cache.example.com",
|
|
aws_iam_region="us-east-1",
|
|
)
|
|
|
|
assert isinstance(client.connection_kwargs["credential_provider"], ElastiCacheIAMCredentialProvider)
|
|
|
|
|
|
def test_provider_keeps_the_rest_of_the_url_intact(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
redis_kwargs = _get_redis_client_logic(
|
|
url="rediss://url-user:url-pass@redis-host:6379/3?protocol=3",
|
|
credential_provider=provider,
|
|
)
|
|
|
|
assert redis_kwargs["url"] == "rediss://redis-host:6379/3?protocol=3"
|
|
|
|
|
|
def test_provider_free_url_is_left_untouched(clean_redis_environment):
|
|
url = "redis://url-user:url-pass@redis-host:6379/3"
|
|
|
|
redis_kwargs = _get_redis_client_logic(url=url)
|
|
|
|
assert redis_kwargs["url"] == url
|
|
|
|
|
|
def test_credential_provider_auth_kwargs_supersedes_credentials_an_explicit_provider_replaces():
|
|
provider = _StubCredentialProvider()
|
|
|
|
auth_kwargs = _credential_provider_auth_kwargs(
|
|
{
|
|
"host": "redis-host",
|
|
"port": 6379,
|
|
"credential_provider": provider,
|
|
"redis_connect_func": _gcp_marker_callback(),
|
|
"username": "url-user",
|
|
"password": "url-pass",
|
|
}
|
|
)
|
|
|
|
assert auth_kwargs["credential_provider"] is provider
|
|
assert auth_kwargs["host"] == "redis-host"
|
|
assert auth_kwargs["port"] == 6379
|
|
assert "redis_connect_func" not in auth_kwargs
|
|
assert "username" not in auth_kwargs
|
|
assert "password" not in auth_kwargs
|
|
|
|
|
|
def test_credential_provider_auth_kwargs_leaves_provider_free_kwargs_alone():
|
|
redis_kwargs = {"host": "redis-host", "port": 6379, "username": "url-user", "password": "url-pass"}
|
|
|
|
assert _credential_provider_auth_kwargs(redis_kwargs) == redis_kwargs
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_redis_cache_test_connection_uses_shared_factory(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
|
|
with (
|
|
patch("redis.Redis", autospec=True),
|
|
patch("redis.asyncio.BlockingConnectionPool", autospec=True),
|
|
patch("redis.asyncio.Redis", autospec=True) as mock_async_redis,
|
|
):
|
|
mock_async_redis.return_value.ping = AsyncMock(return_value=True)
|
|
mock_async_redis.return_value.aclose = AsyncMock()
|
|
cache = RedisCache(host="redis-host", port=6379, credential_provider=provider, password="redis-secret")
|
|
result = await cache.test_connection()
|
|
|
|
client_kwargs = mock_async_redis.call_args.kwargs
|
|
assert result["status"] == "success"
|
|
assert client_kwargs["credential_provider"] is provider
|
|
assert "password" not in client_kwargs
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_redis_cluster_cache_test_connection_uses_shared_factory(clean_redis_environment):
|
|
provider = _StubCredentialProvider()
|
|
recorder = MagicMock()
|
|
|
|
class _StubAsyncCluster:
|
|
def __init__(self, **kwargs):
|
|
recorder(**kwargs)
|
|
|
|
async def ping(self):
|
|
return True
|
|
|
|
async def aclose(self):
|
|
return None
|
|
|
|
with (
|
|
patch("redis.RedisCluster", autospec=True),
|
|
patch("redis.asyncio.cluster.RedisCluster", _StubAsyncCluster),
|
|
):
|
|
cache = RedisClusterCache(startup_nodes=[{"host": "redis-host", "port": 6379}], credential_provider=provider)
|
|
result = await cache.test_connection()
|
|
|
|
cluster_kwargs = recorder.call_args.kwargs
|
|
assert result["status"] == "success"
|
|
assert cluster_kwargs["credential_provider"] is provider
|
|
|
|
|
|
def test_redis_cache_key_does_not_inspect_provider(clear_llm_client_cache):
|
|
provider = _HostileCredentialProvider("synthetic-secret")
|
|
second_provider = _StubCredentialProvider("another-token")
|
|
|
|
with (
|
|
patch("redis.Redis", autospec=True),
|
|
patch("redis.asyncio.BlockingConnectionPool", autospec=True),
|
|
):
|
|
cache = RedisCache(host="redis-host", port=6379, credential_provider=provider)
|
|
second_cache = RedisCache(host="redis-host", port=6379, credential_provider=second_provider)
|
|
|
|
first_key = cache._get_async_client_cache_key()
|
|
assert first_key == cache._get_async_client_cache_key()
|
|
assert first_key != second_cache._get_async_client_cache_key()
|
|
|
|
|
|
def test_pretty_print_never_expands_credential_provider(capsys):
|
|
secret = "aaaa-UNIQUE-SENTINEL-bbbb"
|
|
|
|
with patch( # test-quality-ok: enable the debug-only printer without changing process-wide logger state
|
|
"litellm._redis.verbose_logger.isEnabledFor", return_value=True
|
|
):
|
|
_pretty_print_redis_config(
|
|
redis_kwargs={
|
|
"host": "redis-host",
|
|
"port": 6379,
|
|
"credential_provider": _HostileCredentialProvider(secret),
|
|
}
|
|
)
|
|
|
|
output = capsys.readouterr().out
|
|
assert secret not in output
|
|
assert "UNIQUE" not in output
|
|
assert "_payload" not in output
|
|
assert "credential_provider" in output
|
|
|
|
|
|
def test_redis_cache_key_does_not_serialize_connect_func():
|
|
def connect(connection):
|
|
return None
|
|
|
|
cache = RedisCache.__new__(RedisCache)
|
|
cache.redis_kwargs = {"host": "redis-host", "port": 6379, "redis_connect_func": connect}
|
|
|
|
first_key = cache._get_async_client_cache_key()
|
|
assert first_key == cache._get_async_client_cache_key()
|
|
|
|
|
|
def test_redis_cache_key_keys_opaque_kwargs_by_identity():
|
|
|
|
class _Opaque:
|
|
pass
|
|
|
|
first = RedisCache.__new__(RedisCache)
|
|
first.redis_kwargs = {"host": "redis-host", "retry": _Opaque()}
|
|
second = RedisCache.__new__(RedisCache)
|
|
second.redis_kwargs = {"host": "redis-host", "retry": _Opaque()}
|
|
|
|
assert first._get_async_client_cache_key() == first._get_async_client_cache_key()
|
|
assert first._get_async_client_cache_key() != second._get_async_client_cache_key()
|
|
|
|
|
|
def test_get_redis_url_from_environment_single_url(monkeypatch):
|
|
"""Test when REDIS_URL is directly provided"""
|
|
# Set the environment variable
|
|
monkeypatch.setenv("REDIS_URL", "redis://redis-server:6379/0")
|
|
|
|
# Call the function to get the Redis URL
|
|
redis_url = get_redis_url_from_environment()
|
|
|
|
# Assert that the returned URL matches the expected value
|
|
assert redis_url == "redis://redis-server:6379/0"
|
|
|
|
|
|
def test_get_redis_url_from_environment_host_port(monkeypatch):
|
|
"""Test when REDIS_HOST and REDIS_PORT are provided"""
|
|
# Set the environment variables
|
|
monkeypatch.setenv("REDIS_HOST", "redis-server")
|
|
monkeypatch.setenv("REDIS_PORT", "6379")
|
|
# Ensure authentication variables are not set
|
|
monkeypatch.delenv("REDIS_USERNAME", raising=False)
|
|
monkeypatch.delenv("REDIS_PASSWORD", raising=False)
|
|
monkeypatch.delenv("REDIS_SSL", raising=False)
|
|
|
|
# Call the function to get the Redis URL
|
|
redis_url = get_redis_url_from_environment()
|
|
|
|
# Assert that the returned URL matches the expected value
|
|
assert redis_url == "redis://redis-server:6379"
|
|
|
|
|
|
def test_get_redis_url_from_environment_with_ssl(monkeypatch):
|
|
"""Test when SSL is enabled"""
|
|
# Set the environment variables
|
|
monkeypatch.setenv("REDIS_HOST", "redis-server")
|
|
monkeypatch.setenv("REDIS_PORT", "6379")
|
|
monkeypatch.setenv("REDIS_SSL", "true")
|
|
# Ensure authentication variables are not set
|
|
monkeypatch.delenv("REDIS_USERNAME", raising=False)
|
|
monkeypatch.delenv("REDIS_PASSWORD", raising=False)
|
|
|
|
# Call the function to get the Redis URL
|
|
redis_url = get_redis_url_from_environment()
|
|
|
|
# Assert that the returned URL uses rediss:// protocol
|
|
assert redis_url == "rediss://redis-server:6379"
|
|
|
|
|
|
def test_get_redis_url_from_environment_with_username_password(monkeypatch):
|
|
"""Test when username and password are provided"""
|
|
# Set the environment variables
|
|
monkeypatch.setenv("REDIS_HOST", "redis-server")
|
|
monkeypatch.setenv("REDIS_PORT", "6379")
|
|
monkeypatch.setenv("REDIS_USERNAME", "user")
|
|
monkeypatch.setenv("REDIS_PASSWORD", "password")
|
|
|
|
# Call the function to get the Redis URL
|
|
redis_url = get_redis_url_from_environment()
|
|
|
|
# Assert that the returned URL includes username:password@
|
|
assert redis_url == "redis://user:password@redis-server:6379"
|
|
|
|
|
|
def test_get_redis_url_from_environment_with_password_only(monkeypatch):
|
|
"""Test when only password is provided"""
|
|
# Set the environment variables
|
|
monkeypatch.setenv("REDIS_HOST", "redis-server")
|
|
monkeypatch.setenv("REDIS_PORT", "6379")
|
|
monkeypatch.setenv("REDIS_PASSWORD", "password")
|
|
# Ensure username is not set
|
|
monkeypatch.delenv("REDIS_USERNAME", raising=False)
|
|
monkeypatch.delenv("REDIS_SSL", raising=False)
|
|
|
|
# Call the function to get the Redis URL
|
|
redis_url = get_redis_url_from_environment()
|
|
|
|
# Assert that the returned URL includes :password@
|
|
assert redis_url == "redis://password@redis-server:6379"
|
|
|
|
|
|
def test_get_redis_url_from_environment_with_all_options(monkeypatch):
|
|
"""Test when all options are provided"""
|
|
# Set the environment variables
|
|
monkeypatch.setenv("REDIS_HOST", "redis-server")
|
|
monkeypatch.setenv("REDIS_PORT", "6379")
|
|
monkeypatch.setenv("REDIS_USERNAME", "user")
|
|
monkeypatch.setenv("REDIS_PASSWORD", "password")
|
|
monkeypatch.setenv("REDIS_SSL", "true")
|
|
|
|
# Call the function to get the Redis URL
|
|
redis_url = get_redis_url_from_environment()
|
|
|
|
# Assert that the returned URL includes all components
|
|
assert redis_url == "rediss://user:password@redis-server:6379"
|
|
|
|
|
|
def test_get_redis_url_from_environment_missing_host_port(monkeypatch):
|
|
"""Test error when required variables are missing"""
|
|
# Make sure these environment variables don't exist
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_HOST", raising=False)
|
|
monkeypatch.delenv("REDIS_PORT", raising=False)
|
|
|
|
# Call the function and expect a ValueError
|
|
with pytest.raises(ValueError, match="Either 'REDIS_URL' or both 'REDIS_HOST' and 'REDIS_PORT") as excinfo:
|
|
get_redis_url_from_environment()
|
|
|
|
# Check the error message
|
|
assert "Either 'REDIS_URL' or both 'REDIS_HOST' and 'REDIS_PORT' must be specified" in str(excinfo.value)
|
|
|
|
|
|
def test_get_redis_url_from_environment_missing_port(monkeypatch):
|
|
"""Test error when only REDIS_HOST is provided but REDIS_PORT is missing"""
|
|
# Make sure REDIS_URL doesn't exist and set only REDIS_HOST
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_PORT", raising=False)
|
|
monkeypatch.setenv("REDIS_HOST", "redis-server")
|
|
|
|
# Call the function and expect a ValueError
|
|
with pytest.raises(ValueError, match="Either 'REDIS_URL' or both 'REDIS_HOST' and 'REDIS_PORT") as excinfo:
|
|
get_redis_url_from_environment()
|
|
|
|
# Check the error message
|
|
assert "Either 'REDIS_URL' or both 'REDIS_HOST' and 'REDIS_PORT' must be specified" in str(excinfo.value)
|
|
|
|
|
|
def test_max_connections_in_cluster_kwargs():
|
|
"""Test that max_connections is included in Redis cluster kwargs"""
|
|
kwargs = _get_redis_cluster_kwargs()
|
|
assert "max_connections" in kwargs, "max_connections should be in available Redis cluster kwargs"
|
|
|
|
|
|
def test_socket_timeouts_in_cluster_kwargs():
|
|
"""Test that Redis cluster clients can receive socket timeout configuration"""
|
|
kwargs = _get_redis_cluster_kwargs()
|
|
assert "socket_timeout" in kwargs
|
|
assert "socket_connect_timeout" in kwargs
|
|
|
|
|
|
def test_reconnect_kwargs_in_cluster_kwargs():
|
|
"""Health check and keepalive must survive the cluster kwarg allow-list so
|
|
operators can tune Redis cluster reconnection behavior via config."""
|
|
kwargs = _get_redis_cluster_kwargs()
|
|
assert "health_check_interval" in kwargs
|
|
assert "socket_keepalive" in kwargs
|
|
|
|
|
|
def test_retry_attempts_in_cluster_kwargs():
|
|
"""cluster_error_retry_attempts must survive the cluster kwarg allow-list so
|
|
operators can bound worst-case retry latency on a Redis Cluster: it was being
|
|
silently dropped because the allow-list was built from redis.RedisCluster's
|
|
decorated __init__ without unwrapping it, so getfullargspec saw an empty
|
|
(self, *args, **kwargs) wrapper signature."""
|
|
kwargs = _get_redis_cluster_kwargs()
|
|
assert "cluster_error_retry_attempts" in kwargs
|
|
|
|
|
|
def test_async_only_kwargs_in_cluster_kwargs_when_async_client_requested():
|
|
"""decode_responses is on the async cluster client's constructor and not the sync
|
|
one, on every redis-py the matrix covers. Introspecting the sync class regardless
|
|
of which client is actually built silently drops it for every async cluster caller."""
|
|
sync_kwargs = _get_redis_cluster_kwargs()
|
|
async_kwargs = _get_redis_cluster_kwargs(async_redis.RedisCluster)
|
|
|
|
assert "decode_responses" not in sync_kwargs
|
|
assert "decode_responses" in async_kwargs
|
|
|
|
|
|
@patch( # test-quality-ok: redis-py >= 6 keeps no cluster_error_retry_attempts attribute on the built client, so the constructor call is the only place the value is observable
|
|
"litellm.caching.redis_cluster_node_isolation.get_litellm_async_redis_cluster_class"
|
|
)
|
|
def test_async_cluster_forwards_retry_attempts(mock_get_cluster_class):
|
|
"""Regression: cluster_error_retry_attempts must reach the constructed async
|
|
cluster client. Silently dropping it removes an operator's only lever for
|
|
bounding a stuck node's worst-case retry latency, and the client falls back
|
|
to redis-py's own default (3 retries) instead."""
|
|
mock_cluster_cls = mock_get_cluster_class.return_value
|
|
get_redis_async_client(
|
|
startup_nodes=[{"host": "cluster-node", "port": 6379}],
|
|
cluster_error_retry_attempts=2,
|
|
)
|
|
|
|
call_kwargs = mock_cluster_cls.call_args[1]
|
|
assert call_kwargs["cluster_error_retry_attempts"] == 2
|
|
|
|
|
|
def test_async_cluster_passes_async_only_kwargs():
|
|
"""Regression: decode_responses is an async-cluster-only constructor arg. When
|
|
the allow-list came from the sync class it was filtered out and values came
|
|
back as bytes instead of str."""
|
|
client = get_redis_async_client(
|
|
startup_nodes=[{"host": "cluster-node", "port": 6379}],
|
|
decode_responses=True,
|
|
)
|
|
|
|
assert client.connection_kwargs["decode_responses"] is True
|
|
|
|
|
|
@pytest.mark.parametrize("cluster_client", [redis.RedisCluster, async_redis.RedisCluster], ids=["sync", "async"])
|
|
def test_cluster_kwargs_exclude_variadic_parameters(cluster_client):
|
|
"""*args / **kwargs are signature placeholders, not connection settings, and
|
|
must never land in the allow-list regardless of which cluster client is
|
|
introspected."""
|
|
variadic = {
|
|
name
|
|
for name, param in inspect.signature(cluster_client).parameters.items()
|
|
if param.kind in (param.VAR_POSITIONAL, param.VAR_KEYWORD)
|
|
}
|
|
|
|
leaked = variadic & set(_get_redis_cluster_kwargs(cluster_client))
|
|
assert not leaked, f"variadic params leaked into the allow-list: {leaked}"
|
|
|
|
|
|
@patch("litellm.caching.redis_cluster_node_isolation.get_litellm_async_redis_cluster_class")
|
|
def test_async_cluster_sets_reconnect_defaults(mock_get_cluster_class):
|
|
"""
|
|
The async RedisCluster client must be built with a periodic health check and
|
|
TCP keepalive so a connection silently dropped by a cluster restart (e.g.
|
|
ElastiCache Serverless maintenance) is revalidated and reconnected before
|
|
reuse instead of stalling in re-initialization. Regression for LIT-4083.
|
|
"""
|
|
mock_cluster_cls = mock_get_cluster_class.return_value
|
|
get_redis_async_client(startup_nodes=[{"host": "cluster-node", "port": 6379}])
|
|
|
|
mock_cluster_cls.assert_called_once()
|
|
call_kwargs = mock_cluster_cls.call_args[1]
|
|
assert call_kwargs["health_check_interval"] == REDIS_CLUSTER_HEALTH_CHECK_INTERVAL
|
|
assert call_kwargs["health_check_interval"] > 0
|
|
assert call_kwargs["socket_keepalive"] is True
|
|
|
|
|
|
@patch("litellm.caching.redis_cluster_node_isolation.get_litellm_async_redis_cluster_class")
|
|
def test_async_cluster_reconnect_defaults_are_overridable(mock_get_cluster_class):
|
|
"""An explicit health_check_interval / socket_keepalive from config must win
|
|
over the built-in reconnect defaults."""
|
|
mock_cluster_cls = mock_get_cluster_class.return_value
|
|
get_redis_async_client(
|
|
startup_nodes=[{"host": "cluster-node", "port": 6379}],
|
|
health_check_interval=7,
|
|
socket_keepalive=False,
|
|
)
|
|
|
|
call_kwargs = mock_cluster_cls.call_args[1]
|
|
assert call_kwargs["health_check_interval"] == 7
|
|
assert call_kwargs["socket_keepalive"] is False
|
|
|
|
|
|
def test_get_redis_async_client_with_connection_pool():
|
|
"""Test that connection_pool parameter is properly passed to Redis client"""
|
|
# Create a mock connection pool
|
|
mock_pool = MagicMock(spec=async_redis.BlockingConnectionPool)
|
|
|
|
# Mock the Redis client creation
|
|
with (
|
|
patch("litellm._redis.async_redis.Redis") as mock_redis,
|
|
patch("litellm._redis._get_redis_client_logic") as mock_logic,
|
|
):
|
|
# Configure mock to return basic redis kwargs
|
|
mock_logic.return_value = {"host": "localhost", "port": 6379, "db": 0}
|
|
|
|
# Call get_redis_async_client with connection_pool
|
|
get_redis_async_client(connection_pool=mock_pool)
|
|
|
|
# Verify Redis was called with connection_pool in kwargs
|
|
call_kwargs = mock_redis.call_args[1]
|
|
assert "connection_pool" in call_kwargs, "connection_pool should be passed to Redis client"
|
|
assert call_kwargs["connection_pool"] == mock_pool, "connection_pool should match the provided pool"
|
|
|
|
|
|
def test_get_redis_async_client_without_connection_pool():
|
|
"""Test that Redis client works without connection_pool parameter"""
|
|
with (
|
|
patch("litellm._redis.async_redis.Redis") as mock_redis,
|
|
patch("litellm._redis._get_redis_client_logic") as mock_logic,
|
|
):
|
|
# Configure mock to return basic redis kwargs
|
|
mock_logic.return_value = {"host": "localhost", "port": 6379, "db": 0}
|
|
|
|
# Call get_redis_async_client without connection_pool
|
|
get_redis_async_client()
|
|
|
|
# Verify Redis was called without connection_pool in kwargs
|
|
call_kwargs = mock_redis.call_args[1]
|
|
assert "connection_pool" not in call_kwargs, "connection_pool should not be in kwargs when not provided"
|
|
|
|
|
|
def test_gcp_iam_credential_provider_get_credentials():
|
|
"""GCPIAMCredentialProvider.get_credentials() returns a token tuple."""
|
|
service_account = "projects/-/serviceAccounts/test@project.iam.gserviceaccount.com"
|
|
|
|
with patch(
|
|
"litellm._redis_credential_provider._generate_gcp_iam_access_token",
|
|
return_value="tok-1",
|
|
) as mock_gen:
|
|
provider = GCPIAMCredentialProvider(service_account)
|
|
creds = provider.get_credentials()
|
|
|
|
assert creds == ("tok-1",)
|
|
mock_gen.assert_called_once_with(service_account)
|
|
|
|
|
|
def test_gcp_iam_credential_provider_caches_token():
|
|
"""
|
|
Repeated calls to get_credentials() reuse the cached token and only call
|
|
_generate_gcp_iam_access_token once, avoiding redundant blocking I/O.
|
|
"""
|
|
service_account = "projects/-/serviceAccounts/test@project.iam.gserviceaccount.com"
|
|
|
|
with patch(
|
|
"litellm._redis_credential_provider._generate_gcp_iam_access_token",
|
|
return_value="tok-cached",
|
|
) as mock_gen:
|
|
provider = GCPIAMCredentialProvider(service_account)
|
|
results = [provider.get_credentials() for _ in range(5)]
|
|
|
|
assert all(r == ("tok-cached",) for r in results)
|
|
# Token must be fetched exactly once regardless of how many connections are established
|
|
mock_gen.assert_called_once_with(service_account)
|
|
|
|
|
|
def test_gcp_iam_credential_provider_refreshes_on_expiry():
|
|
"""
|
|
get_credentials() fetches a new token after the cached one expires,
|
|
ensuring connections always authenticate with a valid token.
|
|
"""
|
|
import time
|
|
|
|
import litellm._redis_credential_provider as cred_module
|
|
|
|
service_account = "projects/-/serviceAccounts/test@project.iam.gserviceaccount.com"
|
|
|
|
with patch(
|
|
"litellm._redis_credential_provider._generate_gcp_iam_access_token",
|
|
side_effect=["tok-1", "tok-2"],
|
|
) as mock_gen:
|
|
provider = GCPIAMCredentialProvider(service_account)
|
|
|
|
# First call — populates cache
|
|
assert provider.get_credentials() == ("tok-1",)
|
|
|
|
# Artificially expire the cached token
|
|
cred_module._token_cache[service_account] = ("tok-1", time.monotonic() - 1)
|
|
|
|
# Second call — cache miss, must refresh
|
|
assert provider.get_credentials() == ("tok-2",)
|
|
|
|
assert mock_gen.call_count == 2
|
|
|
|
|
|
def test_gcp_iam_credential_provider_cache_shared_across_instances():
|
|
"""
|
|
Multiple GCPIAMCredentialProvider instances for the same service account
|
|
share one cached token so concurrent Redis connections don't each trigger
|
|
a blocking IAM round-trip.
|
|
"""
|
|
service_account = "projects/-/serviceAccounts/shared@project.iam.gserviceaccount.com"
|
|
|
|
with patch(
|
|
"litellm._redis_credential_provider._generate_gcp_iam_access_token",
|
|
return_value="tok-shared",
|
|
) as mock_gen:
|
|
p1 = GCPIAMCredentialProvider(service_account)
|
|
p2 = GCPIAMCredentialProvider(service_account)
|
|
|
|
assert p1.get_credentials() == ("tok-shared",)
|
|
assert p2.get_credentials() == ("tok-shared",)
|
|
|
|
# Only one network call despite two provider instances
|
|
mock_gen.assert_called_once()
|
|
|
|
|
|
def test_get_redis_async_client_gcp_cluster_uses_credential_provider():
|
|
"""
|
|
When startup_nodes + gcp_service_account are provided, the async cluster client
|
|
must be constructed with a GCPIAMCredentialProvider — not a static password.
|
|
This ensures that the 1-hour IAM token expiry does not cause auth failures.
|
|
"""
|
|
startup_nodes = [{"host": "redis-node-1", "port": 6379}]
|
|
|
|
mock_connect_func = MagicMock()
|
|
mock_connect_func._gcp_service_account = "projects/-/serviceAccounts/sa@project.iam.gserviceaccount.com"
|
|
|
|
redis_kwargs = {
|
|
"startup_nodes": startup_nodes,
|
|
"redis_connect_func": mock_connect_func,
|
|
}
|
|
|
|
with (
|
|
patch(
|
|
"litellm.caching.redis_cluster_node_isolation.get_litellm_async_redis_cluster_class"
|
|
) as mock_get_cluster_class,
|
|
patch("litellm._redis._get_redis_client_logic", return_value=redis_kwargs),
|
|
):
|
|
mock_cluster = mock_get_cluster_class.return_value
|
|
get_redis_async_client()
|
|
|
|
assert mock_cluster.called
|
|
cluster_call_kwargs = mock_cluster.call_args[1]
|
|
|
|
# Must use credential_provider, not a static password
|
|
assert "credential_provider" in cluster_call_kwargs, (
|
|
"async GCP cluster must use credential_provider for per-connection token refresh"
|
|
)
|
|
assert isinstance(cluster_call_kwargs["credential_provider"], GCPIAMCredentialProvider)
|
|
assert "password" not in cluster_call_kwargs, "async GCP cluster must not use a static password (expires after 1h)"
|
|
|
|
|
|
@patch("litellm._redis.init_redis_cluster")
|
|
def test_sync_client_prefers_cluster_over_url(mock_init_cluster, monkeypatch):
|
|
"""
|
|
Test get_redis_client returns RedisCluster when startup_nodes is present even if
|
|
REDIS_URL is also set.
|
|
"""
|
|
monkeypatch.setenv("REDIS_URL", "redis://fallback-host:6379")
|
|
mock_init_cluster.return_value = MagicMock(spec=redis.RedisCluster)
|
|
|
|
startup_nodes = [{"host": "cluster-node.example.com", "port": 6379}]
|
|
get_redis_client(startup_nodes=startup_nodes)
|
|
|
|
mock_init_cluster.assert_called_once()
|
|
call_kwargs = mock_init_cluster.call_args[0][0]
|
|
assert "startup_nodes" in call_kwargs, "startup_nodes must be forwarded to init_redis_cluster"
|
|
|
|
|
|
@patch("litellm.caching.redis_cluster_node_isolation.get_litellm_async_redis_cluster_class")
|
|
def test_async_client_prefers_cluster_over_url(mock_get_cluster_class, monkeypatch):
|
|
"""
|
|
Test (1) get_redis_async_client returns async RedisCluster when startup_nodes is present
|
|
even if REDIS_URL is also set and (2) startup_nodes is forwarded to RedisCluster.
|
|
"""
|
|
mock_cluster_cls = mock_get_cluster_class.return_value
|
|
monkeypatch.setenv("REDIS_URL", "redis://fallback-host:6379")
|
|
|
|
startup_nodes = [{"host": "cluster-node.example.com", "port": 6379}]
|
|
get_redis_async_client(startup_nodes=startup_nodes)
|
|
|
|
mock_cluster_cls.assert_called_once()
|
|
call_kwargs = mock_cluster_cls.call_args[1]
|
|
assert "startup_nodes" in call_kwargs, "startup_nodes must be forwarded to async RedisCluster"
|
|
assert len(call_kwargs["startup_nodes"]) == 1, "should forward exactly 1 cluster node"
|
|
|
|
|
|
@patch("litellm.caching.redis_cluster_node_isolation.get_litellm_async_redis_cluster_class")
|
|
def test_async_client_prefers_cluster_over_url_via_env_var(mock_get_cluster_class, monkeypatch):
|
|
"""
|
|
Test get_redis_async_client returns async RedisCluster when REDIS_CLUSTER_NODES is set
|
|
even if REDIS_URL is also set.
|
|
"""
|
|
mock_cluster_cls = mock_get_cluster_class.return_value
|
|
monkeypatch.setenv("REDIS_URL", "redis://fallback-host:6379")
|
|
monkeypatch.setenv(
|
|
"REDIS_CLUSTER_NODES",
|
|
json.dumps([{"host": "cluster-node.example.com", "port": 6379}]),
|
|
)
|
|
|
|
get_redis_async_client()
|
|
|
|
mock_cluster_cls.assert_called_once()
|
|
call_kwargs = mock_cluster_cls.call_args[1]
|
|
assert "startup_nodes" in call_kwargs, "startup_nodes must be forwarded to async RedisCluster"
|
|
|
|
|
|
@patch("litellm._redis.init_redis_cluster")
|
|
def test_sync_client_prefers_cluster_over_url_via_env_var(mock_init_cluster, monkeypatch):
|
|
"""
|
|
Test get_redis_client returns RedisCluster when REDIS_CLUSTER_NODES is set even if
|
|
REDIS_URL is also set.
|
|
"""
|
|
monkeypatch.setenv("REDIS_URL", "redis://fallback-host:6379")
|
|
monkeypatch.setenv(
|
|
"REDIS_CLUSTER_NODES",
|
|
json.dumps([{"host": "cluster-node.example.com", "port": 6379}]),
|
|
)
|
|
mock_init_cluster.return_value = MagicMock(spec=redis.RedisCluster)
|
|
|
|
get_redis_client()
|
|
|
|
mock_init_cluster.assert_called_once()
|
|
call_kwargs = mock_init_cluster.call_args[0][0]
|
|
assert "startup_nodes" in call_kwargs, "startup_nodes must be forwarded to init_redis_cluster"
|
|
assert len(call_kwargs["startup_nodes"]) == 1
|
|
|
|
|
|
@patch("litellm._redis.redis.Sentinel")
|
|
def test_sync_sentinel_uses_sentinel_password_and_master_password(mock_sentinel_cls):
|
|
"""Sentinel auth must be passed to the sentinel, not the Redis master client."""
|
|
mock_sentinel = MagicMock()
|
|
mock_sentinel_cls.return_value = mock_sentinel
|
|
|
|
get_redis_client(
|
|
sentinel_nodes=[("sentinel-1", 26379)],
|
|
sentinel_password="sentinel-secret",
|
|
service_name="mymaster",
|
|
password="redis-secret",
|
|
username="redis-user",
|
|
ssl=True,
|
|
ssl_cert_reqs="required",
|
|
ssl_check_hostname=True,
|
|
ssl_ca_certs="/tmp/test-ca.pem",
|
|
max_connections=17,
|
|
socket_timeout=5,
|
|
)
|
|
|
|
mock_sentinel_cls.assert_called_once()
|
|
sentinel_call_kwargs = mock_sentinel_cls.call_args[1]
|
|
assert "password" not in sentinel_call_kwargs
|
|
assert "username" not in sentinel_call_kwargs
|
|
assert "ssl" not in sentinel_call_kwargs
|
|
assert "ssl_cert_reqs" not in sentinel_call_kwargs
|
|
assert "ssl_check_hostname" not in sentinel_call_kwargs
|
|
assert "ssl_ca_certs" not in sentinel_call_kwargs
|
|
assert "max_connections" not in sentinel_call_kwargs
|
|
assert "socket_timeout" not in sentinel_call_kwargs
|
|
assert sentinel_call_kwargs["sentinel_kwargs"] == {
|
|
"password": "sentinel-secret",
|
|
"username": "redis-user",
|
|
"ssl": True,
|
|
"ssl_cert_reqs": "required",
|
|
"ssl_check_hostname": True,
|
|
"ssl_ca_certs": "/tmp/test-ca.pem",
|
|
"max_connections": 17,
|
|
"socket_timeout": 5,
|
|
}
|
|
assert "service_name" not in sentinel_call_kwargs["sentinel_kwargs"]
|
|
assert "sentinel_nodes" not in sentinel_call_kwargs["sentinel_kwargs"]
|
|
assert "sentinel_password" not in sentinel_call_kwargs["sentinel_kwargs"]
|
|
mock_sentinel.master_for.assert_called_once_with(
|
|
"mymaster",
|
|
password="redis-secret",
|
|
username="redis-user",
|
|
ssl=True,
|
|
ssl_cert_reqs="required",
|
|
ssl_check_hostname=True,
|
|
ssl_ca_certs="/tmp/test-ca.pem",
|
|
max_connections=17,
|
|
socket_timeout=5,
|
|
)
|
|
|
|
|
|
@patch("redis.Sentinel")
|
|
def test_sync_sentinel_keeps_provider_off_monitors_and_on_master(mock_sentinel_cls):
|
|
provider = _StubCredentialProvider()
|
|
mock_sentinel = MagicMock()
|
|
mock_sentinel_cls.return_value = mock_sentinel
|
|
|
|
get_redis_client(
|
|
sentinel_nodes=[("sentinel-1", 26379)],
|
|
sentinel_password="sentinel-secret",
|
|
service_name="mymaster",
|
|
password="redis-secret",
|
|
credential_provider=provider,
|
|
)
|
|
|
|
sentinel_kwargs = mock_sentinel_cls.call_args.kwargs["sentinel_kwargs"]
|
|
assert sentinel_kwargs["password"] == "sentinel-secret"
|
|
assert "credential_provider" not in sentinel_kwargs
|
|
assert mock_sentinel.master_for.call_args.kwargs["credential_provider"] is provider
|
|
assert "password" not in mock_sentinel.master_for.call_args.kwargs
|
|
|
|
|
|
@patch("litellm._redis.async_redis.Sentinel")
|
|
def test_async_sentinel_uses_sentinel_password_and_master_password(
|
|
mock_sentinel_cls,
|
|
):
|
|
"""Async sentinel auth must mirror the sync sentinel password routing."""
|
|
mock_sentinel = MagicMock()
|
|
mock_sentinel_cls.return_value = mock_sentinel
|
|
|
|
get_redis_async_client(
|
|
sentinel_nodes=[("sentinel-1", 26379)],
|
|
sentinel_password="sentinel-secret",
|
|
service_name="mymaster",
|
|
password="redis-secret",
|
|
username="redis-user",
|
|
ssl=True,
|
|
ssl_cert_reqs="required",
|
|
ssl_check_hostname=True,
|
|
ssl_ca_certs="/tmp/test-ca.pem",
|
|
max_connections=17,
|
|
socket_timeout=5,
|
|
)
|
|
|
|
mock_sentinel_cls.assert_called_once()
|
|
sentinel_call_kwargs = mock_sentinel_cls.call_args[1]
|
|
assert "password" not in sentinel_call_kwargs
|
|
assert "username" not in sentinel_call_kwargs
|
|
assert "ssl" not in sentinel_call_kwargs
|
|
assert "ssl_cert_reqs" not in sentinel_call_kwargs
|
|
assert "ssl_check_hostname" not in sentinel_call_kwargs
|
|
assert "ssl_ca_certs" not in sentinel_call_kwargs
|
|
assert "max_connections" not in sentinel_call_kwargs
|
|
assert "socket_timeout" not in sentinel_call_kwargs
|
|
assert sentinel_call_kwargs["sentinel_kwargs"] == {
|
|
"password": "sentinel-secret",
|
|
"username": "redis-user",
|
|
"ssl": True,
|
|
"ssl_cert_reqs": "required",
|
|
"ssl_check_hostname": True,
|
|
"ssl_ca_certs": "/tmp/test-ca.pem",
|
|
"max_connections": 17,
|
|
"socket_timeout": 5,
|
|
}
|
|
assert "service_name" not in sentinel_call_kwargs["sentinel_kwargs"]
|
|
assert "sentinel_nodes" not in sentinel_call_kwargs["sentinel_kwargs"]
|
|
assert "sentinel_password" not in sentinel_call_kwargs["sentinel_kwargs"]
|
|
mock_sentinel.master_for.assert_called_once_with(
|
|
"mymaster",
|
|
password="redis-secret",
|
|
username="redis-user",
|
|
ssl=True,
|
|
ssl_cert_reqs="required",
|
|
ssl_check_hostname=True,
|
|
ssl_ca_certs="/tmp/test-ca.pem",
|
|
max_connections=17,
|
|
socket_timeout=5,
|
|
)
|
|
|
|
|
|
@patch("litellm._redis.init_redis_cluster")
|
|
def test_sync_client_preserves_password_for_cluster_when_url_also_set(mock_init_cluster, monkeypatch):
|
|
"""
|
|
Test _get_redis_client_logic does not strip password from redis_kwargs when
|
|
startup_nodes is present even if REDIS_URL is also set.
|
|
"""
|
|
monkeypatch.setenv("REDIS_URL", "redis://fallback-host:6379")
|
|
monkeypatch.setenv("REDIS_PASSWORD", "secret")
|
|
mock_init_cluster.return_value = MagicMock(spec=redis.RedisCluster)
|
|
|
|
startup_nodes = [{"host": "cluster-node.example.com", "port": 6379}]
|
|
get_redis_client(startup_nodes=startup_nodes)
|
|
|
|
mock_init_cluster.assert_called_once()
|
|
call_kwargs = mock_init_cluster.call_args[0][0]
|
|
assert "password" in call_kwargs, "password must not be stripped when routing to cluster"
|
|
assert call_kwargs["password"] == "secret"
|
|
|
|
|
|
def test_connection_pool_returns_none_for_cluster(monkeypatch):
|
|
"""Test get_redis_connection_pool returns None when startup_nodes is present."""
|
|
monkeypatch.setenv("REDIS_URL", "redis://fallback-host:6379")
|
|
startup_nodes = [{"host": "cluster-node.example.com", "port": 6379}]
|
|
result = get_redis_connection_pool(startup_nodes=startup_nodes)
|
|
assert result is None, "connection pool must be None for cluster mode"
|
|
|
|
|
|
@patch("litellm._redis.redis.Redis.from_url")
|
|
def test_sync_client_url_used_when_no_cluster(mock_from_url, monkeypatch):
|
|
"""
|
|
Test get_redis_client default to using URL path when no startup_nodes are provided.
|
|
"""
|
|
monkeypatch.setenv("REDIS_URL", "redis://plain-host:6379")
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
get_redis_client()
|
|
|
|
mock_from_url.assert_called_once()
|
|
|
|
|
|
@patch("litellm._redis.redis.Redis.from_url")
|
|
def test_explicit_host_outranks_environment_redis_url(mock_from_url, monkeypatch):
|
|
"""
|
|
An explicitly configured host must win over REDIS_URL in the environment.
|
|
|
|
Otherwise the url branch strips the caller's host/port and the client
|
|
silently connects to whatever REDIS_URL names, so an explicit config block
|
|
(or a connection test typed into the admin UI) targets the wrong server.
|
|
"""
|
|
monkeypatch.setenv("REDIS_URL", "redis://env-host:6379")
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
client = get_redis_client(host="explicit-host", port=6380)
|
|
|
|
mock_from_url.assert_not_called()
|
|
assert client.connection_pool.connection_kwargs["host"] == "explicit-host"
|
|
assert client.connection_pool.connection_kwargs["port"] == 6380
|
|
|
|
|
|
@patch("litellm._redis.redis.Redis.from_url")
|
|
def test_explicit_url_still_wins_over_environment_host(mock_from_url, monkeypatch):
|
|
"""An explicit url argument keeps taking the from_url path."""
|
|
monkeypatch.setenv("REDIS_HOST", "env-host")
|
|
monkeypatch.setenv("REDIS_PORT", "6379")
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
get_redis_client(url="redis://explicit-host:6380")
|
|
|
|
mock_from_url.assert_called_once()
|
|
assert mock_from_url.call_args.kwargs["url"] == "redis://explicit-host:6380"
|
|
|
|
|
|
@patch("litellm._redis.redis.Redis.from_url")
|
|
def test_environment_redis_url_used_when_caller_names_no_target(mock_from_url, monkeypatch):
|
|
"""With no caller-supplied connection target, REDIS_URL still drives the client."""
|
|
monkeypatch.setenv("REDIS_URL", "redis://env-host:6379")
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
get_redis_client()
|
|
|
|
mock_from_url.assert_called_once()
|
|
|
|
|
|
@pytest.mark.parametrize("falsy_ssl", [False, None, 0, ""])
|
|
def test_connection_pool_falsy_ssl_uses_plain_connection(falsy_ssl, monkeypatch):
|
|
"""
|
|
ssl=False must produce a plain (non-TLS) connection pool.
|
|
|
|
The admin UI's coordination Redis form always sends ssl explicitly, so a
|
|
presence check here turns ssl=False into an SSLConnection; the TLS
|
|
handshake against a plaintext Redis then hangs until the ping timeout and
|
|
every connection test from the UI fails.
|
|
"""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_SSL", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
with patch("litellm._redis.async_redis.BlockingConnectionPool") as mock_pool:
|
|
get_redis_connection_pool(host="plain-redis.example.com", port=6379, ssl=falsy_ssl)
|
|
|
|
call_kwargs = mock_pool.call_args.kwargs
|
|
assert call_kwargs.get("connection_class") is not async_redis.SSLConnection, (
|
|
f"ssl={falsy_ssl!r} must not select SSLConnection"
|
|
)
|
|
assert "ssl" not in call_kwargs, "ssl must never leak into BlockingConnectionPool kwargs"
|
|
|
|
|
|
def test_connection_pool_ssl_true_uses_ssl_connection(monkeypatch):
|
|
"""ssl=True must still opt in to a TLS connection pool."""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_SSL", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
with patch("litellm._redis.async_redis.BlockingConnectionPool") as mock_pool:
|
|
get_redis_connection_pool(host="tls-redis.example.com", port=6380, ssl=True)
|
|
|
|
call_kwargs = mock_pool.call_args.kwargs
|
|
assert call_kwargs.get("connection_class") is async_redis.SSLConnection
|
|
assert "ssl" not in call_kwargs, "ssl must be consumed, not forwarded to the pool"
|
|
|
|
|
|
def test_connection_pool_without_ssl_kwarg_uses_plain_connection(monkeypatch):
|
|
"""Omitting ssl entirely must keep the historical plain-connection default."""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_SSL", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
with patch("litellm._redis.async_redis.BlockingConnectionPool") as mock_pool:
|
|
get_redis_connection_pool(host="plain-redis.example.com", port=6379)
|
|
|
|
call_kwargs = mock_pool.call_args.kwargs
|
|
assert call_kwargs.get("connection_class") is not async_redis.SSLConnection
|
|
assert "ssl" not in call_kwargs
|
|
|
|
|
|
def test_connection_pool_env_redis_ssl_false_uses_plain_connection(monkeypatch):
|
|
"""REDIS_SSL=false from the environment must not select SSLConnection."""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
monkeypatch.setenv("REDIS_SSL", "false")
|
|
|
|
pool = get_redis_connection_pool(host="plain-host", port=6379)
|
|
|
|
assert pool is not None
|
|
assert pool.connection_class is async_redis.Connection
|
|
assert "ssl" not in pool.connection_kwargs
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"redis_config",
|
|
[
|
|
pytest.param({"host": "redis-host", "port": 6379}, id="host_port"),
|
|
pytest.param({"url": "redis://redis-host:6379"}, id="url"),
|
|
],
|
|
)
|
|
def test_connection_pool_keeps_socket_timeout(redis_config, monkeypatch):
|
|
"""The async pool must carry socket_timeout however Redis was configured.
|
|
|
|
The url branch used to rebuild pool kwargs from scratch as {timeout, url,
|
|
max_connections}, dropping socket_timeout. redis-py then leaves both
|
|
socket_timeout and socket_connect_timeout (which falls back to it) unset, so a
|
|
Redis host that drops packets rather than refusing them blocks every caller
|
|
indefinitely instead of failing fast.
|
|
"""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_HOST", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
pool = get_redis_connection_pool(socket_timeout=5.0, **redis_config)
|
|
|
|
assert pool is not None
|
|
assert pool.connection_kwargs.get("socket_timeout") == 5.0
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"redis_config",
|
|
[
|
|
pytest.param({"host": "redis-host", "port": 6379}, id="host_port"),
|
|
pytest.param({"url": "redis://redis-host:6379"}, id="url"),
|
|
],
|
|
)
|
|
def test_sync_client_keeps_socket_timeout(redis_config, monkeypatch):
|
|
"""The sync client is built during RedisCache.__init__ and blocks the caller.
|
|
|
|
Without socket_timeout it stalls for the OS TCP timeout against an unreachable
|
|
host, so merely constructing the cache stops the process.
|
|
"""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_HOST", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
client = get_redis_client(socket_timeout=5.0, **redis_config)
|
|
|
|
assert client.connection_pool.connection_kwargs.get("socket_timeout") == 5.0
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"redis_config",
|
|
[
|
|
pytest.param({"host": "redis-host", "port": 6379}, id="host_port"),
|
|
pytest.param({"url": "redis://redis-host:6379"}, id="url"),
|
|
],
|
|
)
|
|
def test_async_client_keeps_socket_timeout(redis_config, monkeypatch):
|
|
"""Same invariant for the async client built without an injected pool."""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_HOST", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
client = get_redis_async_client(socket_timeout=5.0, **redis_config)
|
|
|
|
assert client.connection_pool.connection_kwargs.get("socket_timeout") == 5.0
|
|
|
|
|
|
def test_url_config_does_not_forward_ssl_kwarg(monkeypatch):
|
|
"""ssl stays consumed rather than forwarded on the url path.
|
|
|
|
TLS is selected by the rediss:// scheme there; handing ssl=True to a redis://
|
|
url yields a plain Connection that rejects the kwarg when it first connects.
|
|
"""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_HOST", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
client = get_redis_client(url="redis://redis-host:6379", ssl=True)
|
|
|
|
assert "ssl" not in client.connection_pool.connection_kwargs
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"client_only_kwarg",
|
|
[
|
|
pytest.param({"single_connection_client": True}, id="single_connection_client"),
|
|
pytest.param({"auto_close_connection_pool": True}, id="auto_close_connection_pool"),
|
|
pytest.param({"ssl_ca_certs": "/tmp/ca.pem"}, id="ssl_ca_certs"),
|
|
pytest.param({"ssl": True}, id="ssl"),
|
|
],
|
|
)
|
|
def test_url_config_drops_kwargs_the_connection_cannot_accept(client_only_kwarg, monkeypatch):
|
|
"""Only kwargs the connection accepts may be forwarded on the url path.
|
|
|
|
from_url hands its kwargs down to the connection class, so client-level settings and
|
|
the SSLConnection-only ssl_* family raise TypeError the first time a connection is
|
|
created. TLS on a url config comes from the rediss:// scheme instead.
|
|
"""
|
|
monkeypatch.delenv("REDIS_URL", raising=False)
|
|
monkeypatch.delenv("REDIS_HOST", raising=False)
|
|
monkeypatch.delenv("REDIS_CLUSTER_NODES", raising=False)
|
|
|
|
pool = get_redis_connection_pool(url="redis://redis-host:6379", socket_timeout=5.0, **client_only_kwarg)
|
|
|
|
assert pool is not None
|
|
pool.make_connection()
|
|
assert pool.connection_kwargs.get("socket_timeout") == 5.0
|
|
|
|
|
|
def test_redis_uses_the_hiredis_response_parser():
|
|
"""The C parser must be the one redis-py actually picks.
|
|
|
|
hiredis is declared in the `proxy` extra purely for speed; nothing imports it, so
|
|
dropping it from pyproject.toml would silently fall back to the pure-Python parser
|
|
with no other symptom. redis-py selects it at import time, so asserting on the
|
|
selection is what catches that.
|
|
"""
|
|
from redis._parsers import _HiredisParser
|
|
from redis.connection import HIREDIS_AVAILABLE, DefaultParser
|
|
|
|
assert HIREDIS_AVAILABLE, "hiredis is not installed; redis-py fell back to the pure-Python parser"
|
|
assert DefaultParser is _HiredisParser, f"redis-py selected {DefaultParser.__name__}, expected _HiredisParser"
|
|
|
|
client = get_redis_client(host="redis-host", port=6379)
|
|
connection = client.connection_pool.make_connection()
|
|
assert isinstance(connection._parser, _HiredisParser)
|
|
|
|
|
|
def test_init_arg_names_sees_through_decorated_inits():
|
|
"""redis-py >= 7.4 wraps AbstractConnection.__init__ with @deprecated_args, whose
|
|
wrapper is declared (self, *args, **kwargs). Introspecting the wrapper directly
|
|
yields no real parameters, which silently emptied the from_url allowlist and
|
|
dropped socket_timeout from url-configured connections. The MRO walk must follow
|
|
__wrapped__ to the true signature.
|
|
"""
|
|
import functools
|
|
|
|
from litellm._redis import _init_arg_names
|
|
|
|
def deprecating(fn):
|
|
@functools.wraps(fn)
|
|
def wrapper(self, *args, **kwargs):
|
|
return fn(self, *args, **kwargs)
|
|
|
|
return wrapper
|
|
|
|
class Base:
|
|
@deprecating
|
|
def __init__(self, socket_timeout=None, socket_connect_timeout=None):
|
|
pass
|
|
|
|
class Concrete(Base):
|
|
def __init__(self, host=None, **kwargs):
|
|
super().__init__(**kwargs)
|
|
|
|
names = _init_arg_names(Concrete)
|
|
assert "socket_timeout" in names
|
|
assert "socket_connect_timeout" in names
|
|
assert "host" in names
|
|
|
|
|
|
def test_url_allowlist_always_carries_socket_timeouts():
|
|
"""The load-bearing invariant behind test_url_config_* against the INSTALLED
|
|
redis-py, whatever its version: if a redis-py release changes how its __init__
|
|
signatures are declared (7.4 did, via @deprecated_args), this is the first
|
|
assertion that goes red.
|
|
"""
|
|
from litellm._redis import _get_redis_url_kwargs
|
|
|
|
allowed = _get_redis_url_kwargs()
|
|
assert "socket_timeout" in allowed
|
|
assert "socket_connect_timeout" in allowed
|
|
|
|
|
|
AZURE_AD_CONNECT_FUNC = {"_azure_credential": object()}
|
|
GCP_IAM_CONNECT_FUNC = {"_gcp_service_account": "projects/-/serviceAccounts/sa@project.iam.gserviceaccount.com"}
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"markers, provider_cls",
|
|
[
|
|
(AZURE_AD_CONNECT_FUNC, AzureADCredentialProvider),
|
|
(GCP_IAM_CONNECT_FUNC, GCPIAMCredentialProvider),
|
|
],
|
|
ids=["azure_ad", "gcp_iam"],
|
|
)
|
|
def test_async_url_client_authenticates_through_credential_provider(markers, provider_cls):
|
|
"""A REDIS_URL config with Azure AD or GCP IAM must still reach the server with a credential.
|
|
|
|
The url branch forwards redis_connect_func straight to the async connection, which runs
|
|
its AUTH exchange with the blocking client API and dies, so the branch has to hand the
|
|
connection a CredentialProvider instead.
|
|
"""
|
|
redis_kwargs = {
|
|
"url": "rediss://redis-host:6380",
|
|
"redis_connect_func": SimpleNamespace(**markers),
|
|
}
|
|
|
|
with patch("litellm._redis._get_redis_client_logic", return_value=redis_kwargs):
|
|
client = get_redis_async_client()
|
|
|
|
connection_kwargs = client.connection_pool.connection_kwargs
|
|
assert isinstance(connection_kwargs.get("credential_provider"), provider_cls)
|
|
assert "redis_connect_func" not in connection_kwargs
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"markers, provider_cls",
|
|
[
|
|
(AZURE_AD_CONNECT_FUNC, AzureADCredentialProvider),
|
|
(GCP_IAM_CONNECT_FUNC, GCPIAMCredentialProvider),
|
|
],
|
|
ids=["azure_ad", "gcp_iam"],
|
|
)
|
|
def test_async_url_connection_pool_authenticates_through_credential_provider(markers, provider_cls):
|
|
"""Same for the pool-based path: every connection the pool hands out needs the provider."""
|
|
redis_kwargs = {
|
|
"url": "rediss://redis-host:6380",
|
|
"redis_connect_func": SimpleNamespace(**markers),
|
|
}
|
|
|
|
with patch("litellm._redis._get_redis_client_logic", return_value=redis_kwargs):
|
|
pool = get_redis_connection_pool()
|
|
|
|
assert isinstance(pool.connection_kwargs.get("credential_provider"), provider_cls)
|
|
assert "redis_connect_func" not in pool.connection_kwargs
|
|
|
|
|
|
def test_async_url_client_drops_username_alongside_credential_provider():
|
|
"""redis-py refuses a connection given both a username and a credential_provider, and
|
|
AzureADCredentialProvider already carries REDIS_USERNAME, so the username must be dropped.
|
|
"""
|
|
redis_kwargs = {
|
|
"url": "rediss://redis-host:6380",
|
|
"username": "redis-user",
|
|
"redis_connect_func": SimpleNamespace(**AZURE_AD_CONNECT_FUNC),
|
|
}
|
|
|
|
with patch("litellm._redis._get_redis_client_logic", return_value=redis_kwargs):
|
|
client = get_redis_async_client()
|
|
|
|
pool = client.connection_pool
|
|
assert "username" not in pool.connection_kwargs
|
|
pool.connection_class(**pool.connection_kwargs)
|
|
|
|
|
|
@pytest.mark.parametrize("build_pool", [False, True], ids=["client", "pool"])
|
|
def test_async_url_keeps_a_coroutine_connect_func(build_pool):
|
|
"""redis-py awaits a coroutine redis_connect_func on an async connection, so one we cannot
|
|
turn into a credential provider has to be left where it is rather than dropped.
|
|
"""
|
|
|
|
async def connect(connection):
|
|
return None
|
|
|
|
redis_kwargs = {
|
|
"url": "rediss://redis-host:6380",
|
|
"redis_connect_func": connect,
|
|
}
|
|
|
|
with patch("litellm._redis._get_redis_client_logic", return_value=redis_kwargs):
|
|
pool = get_redis_connection_pool() if build_pool else get_redis_async_client().connection_pool
|
|
|
|
assert pool.connection_kwargs["redis_connect_func"] is connect
|
|
assert "credential_provider" not in pool.connection_kwargs
|
|
|
|
|
|
def test_async_cluster_drops_a_connect_func_it_cannot_pass_on():
|
|
"""redis-py's async RedisCluster has no redis_connect_func parameter, so a connect func that
|
|
is not translated into a credential provider has to be dropped rather than forwarded.
|
|
"""
|
|
|
|
async def connect(connection):
|
|
return None
|
|
|
|
redis_kwargs = {
|
|
"startup_nodes": [{"host": "cluster-node", "port": 6379}],
|
|
"redis_connect_func": connect,
|
|
}
|
|
|
|
with patch("litellm._redis._get_redis_client_logic", return_value=redis_kwargs):
|
|
client = get_redis_async_client()
|
|
|
|
assert isinstance(client, async_redis.RedisCluster)
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"markers, provider_cls",
|
|
[
|
|
(AZURE_AD_CONNECT_FUNC, AzureADCredentialProvider),
|
|
(GCP_IAM_CONNECT_FUNC, GCPIAMCredentialProvider),
|
|
],
|
|
ids=["azure_ad", "gcp_iam"],
|
|
)
|
|
@pytest.mark.parametrize(
|
|
"sentinel_password",
|
|
[None, "sentinel-secret"],
|
|
ids=["unauthenticated_monitors", "password_protected_monitors"],
|
|
)
|
|
def test_async_sentinel_keeps_the_credential_provider_off_the_monitors(markers, provider_cls, sentinel_password):
|
|
"""The Sentinel monitors are separate servers with their own password, so the data node's token
|
|
never belongs on them: redis-py refuses it next to a Sentinel password, and sends it to an
|
|
unauthenticated monitor as an AUTH the monitor rejects.
|
|
"""
|
|
redis_kwargs = {
|
|
"sentinel_nodes": [("sentinel-1", 26379)],
|
|
"sentinel_password": sentinel_password,
|
|
"service_name": "mymaster",
|
|
"redis_connect_func": SimpleNamespace(**markers),
|
|
}
|
|
|
|
with patch("litellm._redis.async_redis.Sentinel") as mock_sentinel_cls:
|
|
with patch("litellm._redis._get_redis_client_logic", return_value=redis_kwargs):
|
|
get_redis_async_client()
|
|
|
|
sentinel_kwargs = mock_sentinel_cls.call_args[1]["sentinel_kwargs"]
|
|
assert sentinel_kwargs["password"] == sentinel_password
|
|
assert "credential_provider" not in sentinel_kwargs
|
|
|
|
monitor_connection = async_redis.Connection(host="sentinel-1", port=26379, **sentinel_kwargs)
|
|
assert monitor_connection.credential_provider is None
|
|
assert bool(monitor_connection.username or monitor_connection.password) is bool(sentinel_password)
|
|
|
|
master_kwargs = mock_sentinel_cls.return_value.master_for.call_args[1]
|
|
assert isinstance(master_kwargs["credential_provider"], provider_cls)
|
|
assert "password" not in master_kwargs
|