Forwarded client headers on bridged /v1/responses calls were serialized into the
OpenAI JSON body as extra_headers when EXPERIMENTAL_OPENAI_BASE_LLM_HTTP_HANDLER
was set, and OpenAI rejected the request with unknown_parameter. The headers are
already merged into the outgoing HTTP headers, so only set the SDK-style
optional param on the SDK client path
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
get_litellm_params() keys such as model_alias_map, completion_call_id and
custom_prompt_dict were missing from all_litellm_params, so when a bridge
forwarded litellm_params as completion kwargs they were swept into extra_body
and OpenAI rejected the call with "Unknown parameter: 'model_alias_map'".
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Releasing the slot unconditionally from the route raced the limiter's own
success handler on the logging worker: both could read the same stashed
acquisition before either cleared it, and under the integer in-memory
fallback that double-decrements the counter. The route now releases only on
exits without a success callback (pre-call rejection, pre-call cancellation,
and Phase 2 exits without the success stamp), matching the HTTP disconnect
path's ownership rule.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Pre-call limiters reject before a deployment is attached to request_data, so
the failure hook could not resolve api_provider for router aliases and emitted
api_provider="None" on litellm_proxy_failed_requests_metric_total and
litellm_proxy_total_requests_metric_total. Fall back to the provider the
limiter already resolved onto RateLimitError.llm_provider, keeping request
data as the first source and ignoring the proxy placeholder.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
_client_async_logging_helper re-submitted logging_obj.success_handler to the
executor after _dispatch_success_logging had already done so, running the same
success pipeline twice per async request and racing on shared logging state.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
useTeam seeds its cache from the team list, which has no organization_models, so the select briefly rendered unfiltered. Also drop the Prisma include assertion from the backend test.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
team_info resolves prisma_client and get_all_team_memberships from module
state, so there is no collaborator to inject; the two sibling team_info tests
patch the same way.
Claude-Session: https://claude.ai/code/session_011Tn3657NkV6ojLqewL64Kb
The team edit form only offered "All Proxy Models" when the dashboard could
read the parent organization, and /organization/info 403s for anyone who is
not a proxy admin or an admin of that org. A team admin with the internal_user
proxy role therefore saw only "No Default Models", which is the opposite of
what they wanted, and had no way to grant their team everything on the proxy.
/team/info now reports the parent org's model ceiling as organization_models,
which the same authorization already admits, and ModelSelect reads the ceiling
from the team it is editing before falling back to the organization. That also
makes the individual model list respect the org's allow-list instead of
listing every proxy model to a caller whose save would be rejected.
useTeam was typed as Team while returning the /team/info envelope, so its one
other caller unwrapped it behind a cast. It now returns the team itself.
Claude-Session: https://claude.ai/code/session_011Tn3657NkV6ojLqewL64Kb
