Pass-through requests inject the proxy span into upstream headers since #40669, which
replaced an explicit x-pass-traceparent with an unrelated trace and dropped its
x-pass-tracestate. Keep the caller's context when the carrier already names a
different trace, and keep the proxy child span for same-trace or missing headers.
Co-authored-by: yucheng <yucheng@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): name the blocking guardrail in x-litellm-applied-guardrails
When a guardrail hook raises, the common ProxyLogging dispatch (sequential and parallel pre_call, pipeline block, during_call and post_call metrics wrapper, streaming iterator wrapper) now records that guardrail in applied_guardrails before re-raising, and pre_call_hook folds request-declared guardrails in on its raising path. Buffered streams rebuild their response headers after the first chunk so a post_call block reached while buffering carries the blocker too
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* fix(proxy): attribute only the raising layer in stream and pipeline blocks
The streaming wrapper caught every exception crossing its boundary and named its own
callback, so a block by an inner guardrail or a provider stream failure also named every
outer guardrail. The wrapper now runs the hook over an upstream boundary that remembers
the exception it raised, and skips attribution when the same exception passes through
Pipeline blocks converted from SensitiveDataRouteException or ModifyResponseException into
a generic guardrail_pipeline_error now still record the blocking step's guardrail
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
* refactor(proxy): drop explanatory docstrings from the stream attribution helpers
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
---------
Co-authored-by: yucheng <yucheng@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
A deployment whose reasoning_effort is an object is copied through as
reasoning the way the HTTP mapper does it instead of being dropped, and
the relay re-asserts the response.create frame type after merging
extra_body so a type key inside it can never replace it. The lazy
OpenAPI snapshot goes back to main: the earlier regeneration came from a
Python 3.14 interpreter dedenting docstrings, which CI on 3.12 rejects
The httpx.Response built for a Rust upstream failure had no request attached,
so constructing openai.AuthenticationError raised RuntimeError inside the
exception mapper and every bad-key OCR call surfaced as APIConnectionError 500
instead of AuthenticationError 401 (the Python path already returned 401)
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Authorization and pricing read the first model and language query value, but the raw query was forwarded, so Deepgram (which honours the last repeated value) could be sent a model the key was never allowed. Later duplicates of those two keys are now dropped before the upstream URL is built
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
A caller could pick a model with only a pre-recorded registry row, or no row at all, and the session would be billed at the pre-recorded rate or logged at zero cost, so budgets did not apply. The route now closes the WebSocket with 1008 before dialing Deepgram unless deepgram/streaming/<model> (or the -multilingual row for language=multi) is an exact registry hit, and the logging handler applies the same check so a registry change under a live session records the duration with no cost instead of a substitute rate
Regression tests cover the route refusal, an operator-supplied streaming row for another model being accepted, and the handler never substituting the pre-recorded rate
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
A mixed /team/member_add list that names an existing member used to run
add_new_member for them, which created or cloned a budget that the empty
upsert update branch never linked to their membership row. Filter the
requested members against the freshly locked roster first so budgets and
membership rows are only written for members who are actually new
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Replaces the per team advisory lock and Pydantic roster parse in the spend flush
with a single INSERT ... ON CONFLICT statement that checks the stored roster in
SQL, so malformed roster JSON cannot fail the whole flush and large batches no
longer issue two queries per team inside the fixed transaction deadline
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Take the team advisory lock in the spend flush transaction and read the roster
through it, so a delayed flush after /team/member_delete cannot recreate the
deleted LiteLLM_TeamMembership row. TEAM_ADVISORY_LOCK_SQL moves to
team_repository so the spend writer can import it without a circular import
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
add_new_member only wrote a LiteLLM_TeamMembership row when a budget id
resolved, and the spend writer used update_many so a missing row failed
silently. Members without a budget therefore never accrued per-member spend.
The membership row is now always upserted (budget_id NULL when no budget
applies) and the spend write is an upsert so members added before this fix
start accruing on their next request.
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
