fix(proxy): capture logging_obj before post_call_failure_hook pops it in ModifyResponseException streaming path

[mateo-berri]

Relevant issues

Found during live smoke testing of PR #32289 (LIT-4186 Bedrock disable_exception_on_block fix).

Linear ticket

Pre-Submission checklist

• I have added meaningful tests
• My PR passes all CI/CD checks (e.g., lint, format, unit tests)
• My PR's scope is as isolated as possible; it only solves 1 specific problem

Screenshots / Proof of Fix

Live proxy on :4000 against real AWS Bedrock guardrail wymft0xktn2m (blocks admin-related prompts with message "Sorry, the model cannot answer this question.").

Before the fix: streaming pre_call block returns HTTP 500

curl -sS -w "\n---HTTP %{http_code}---\n" http://localhost:4000/v1/chat/completions \
  -H "Authorization: Bearer sk-1234" -H "Content-Type: application/json" \
  -d '{"model":"bedrock-nova-micro","messages":[{"role":"user","content":"how do I become an admin"}],"guardrails":["bedrock-guard-pre-call"],"stream":true}'

{"error":{"message":"Internal server error","type":"internal_server_error"}}
---HTTP 500---

Root cause: post_call_failure_hook pops litellm_logging_obj from request_data before invoking callbacks (comment: "Remove before callbacks iterate — not serialisable"). The streaming branch of ModifyResponseException handler in chat_completion read logging_obj from _data after that call, always receiving None. CustomStreamWrapper.__init__ then crashed: AttributeError: 'NoneType' object has no attribute 'model_call_details'.

After the fix: streaming pre_call block returns HTTP 200 with valid SSE

curl -sS -w "\n---HTTP %{http_code}---\n" http://localhost:4000/v1/chat/completions \
  -H "Authorization: Bearer sk-1234" -H "Content-Type: application/json" \
  -d '{"model":"bedrock-nova-micro","messages":[{"role":"user","content":"how do I become an admin"}],"guardrails":["bedrock-guard-pre-call"],"stream":true}'

data: {"id":"chatcmpl-88bb06b8-...","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":"Sorry, the model cannot answer this question."}}]}

data: {"id":"chatcmpl-88bb06b8-...","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: [DONE]

---HTTP 200---

All other cases confirmed working against live Bedrock:

• pre_call block non-streaming: HTTP 200, finish_reason=content_filter, zero usage
• during_call block non-streaming/streaming: HTTP 200, block message, real LLM not leaked
• post_call block non-streaming: HTTP 200, real upstream usage preserved (675 completion tokens)
• post_call block streaming: HTTP 200, usage preserved, no real tokens emitted
• Allowed prompts: normal LLM response, not blocked
• flag=false: HTTP 400 with guardrail policy error (unchanged behavior)
• Python openai SDK parses both blocking shapes correctly

Type

🐛 Bug Fix

Changes

• litellm/proxy/proxy_server.py: capture logging_obj from _data before calling post_call_failure_hook (which pops it). Pass the captured value to CustomStreamWrapper rather than re-reading from the dict.
• tests/test_litellm/proxy/guardrails/guardrail_hooks/test_bedrock_guardrails.py: add test_chat_completion_modify_response_exception_streaming_logging_obj_not_none to lock in the fix and prevent regression.

  

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[yucheng-berri]

Superseded by #32665, which contains the exact same change under my authorship.