* feat(guardrails): Add deduplication and session tracking
- Implement deduplication logic to prevent duplicate scans (via call_id; add _check_and_mark_scanned) caused by LiteLLM callback system
- Add session tracking using litellm_trace_id as AI Session ID for Prisma AIRS SCM logging
- Extract helper methods _extract_prompt_from_request maintainability
- Use httpxSpecialProvider import (LoggingCallback -> GuardrailCallback)
- Add comprehensive tests for deduplication and session tracking (7 new tests)
- Update documentation with multi-turn conversation tracking examples
* docs: update PANW Prisma AIRS multi-turn conversation example to use industry-standard terminology
- Clearer example for conversation tracking
- Updated terminology from 'AI Session ID' to 'Prisma AIRS AI Session ID' for clarity
* fix: remove unused asyncio import
* fix: correct mypy type ignore comment
* KeyManagementSystem add cyberark
* add CyberArkSecretManager
* add CyberArkSecretManager
* add CyberArkSecretManager
* docs add CyberArkSecretManager
* docs
* refactor to use get_secret_from_manager
* Potential fix for code scanning alert no. 3645: Clear-text logging of sensitive information
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
* Potential fix for code scanning alert no. 3650: Clear-text logging of sensitive information
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
* Potential fix for code scanning alert no. 3649: Clear-text logging of sensitive information
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
* Potential fix for code scanning alert no. 3646: Clear-text logging of sensitive information
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
---------
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
* feat(llm_passthrough_endpoints.py): support milvus passthrough api
* fix(llm_passthrough_endpoints.py): move streaming request value to the top of the function
* docs: document new milvus vector store passthrough flow
* feat: change guardrail_information to list type to support displaying multiple guardrails
* fix: add missing commit and revert auto-format changes in utils.py
---------
Co-authored-by: Krish Dholakia <krrishdholakia@gmail.com>
* Add v1 cut of container api
* fix lint errors
* Add proxy support to container apis & logging support (#16049)
* Add proxy support to container apis
* Add logging support
* Add cost tracking support for containers and documentation
* Add new constant documentation
* Add container cost in model map
* fix failing azure tests
* Update tests based on model map changes
* fix model map tests
* fix model map tests
* Container modeshould be container
* Container tests fix
* Merge branch 'main' into litellm_sameer_oct_staging_2
---------
Co-authored-by: Ishaan Jaffer <ishaanjaffer0324@gmail.com>
* fix model error for apis which don't need model
* fix print statments:
* fix mypy lint errors
---------
Co-authored-by: Ishaan Jaff <ishaanjaffer0324@gmail.com>
* feat(vector_store_endpoints/endpoints.py): add new index_create endpoint
allows admin to create a virtual index, to do permission management for
* feat(key_management_endpoints.py): enable setting allowed_vector_store_indexes on keys
proxy admin can enable dev to create an index on a vector stor
* feat: initial commit adding vector store index passthrough logic to litellm
* feat: add vector store table
* fix(azure_ai/transformation.py): fix headers
* feat: track read/write endpoints by vector store integration
enables permissions by index to work
* fix: azure_ai/vector_stores/search
document the vector store endpoints correctly
ensures permission management works as expected
* fix(proxy/utils.py): improve error message
* docs(azure_ai_vector_stores_passthrough.md): document azure ai passthrough vector store support
* docs(create.md): document azure ai support via passthrough for vector store create
* fix: fix code qa errors
* fix: document new allowed_vector_store_indexes endpoint
* feat(milvus/): initial commit adding milvus vector store support to LiteLLM
allows querying milvus vector store through litellm
* feat(bedrock/vector_stores): support translating openai filters param to aws kb
adds filtering to aws kb
* feat(milvus/): add milvus vector store unified search support
allows calling milvus vector store in through chat completions
* docs(milvus_vector_stores.md): document new milvus vector search integration
* feat(pass_through_endpoints.py): support passing form data through to a passthrough endpoint
Closes LIT-1147
* fix: fix linting errors
* fix(apscheduler): prevent memory leaks from jitter and frequent job intervals
Fixes critical memory leak in APScheduler that causes 35GB+ memory allocations
during proxy startup and operation. The leak was identified through Memray
analysis showing massive allocations in normalize() and _apply_jitter()
functions.
Key changes:
1. Remove jitter parameters from all scheduled jobs - jitter was causing
expensive normalize() calculations leading to memory explosion
2. Configure AsyncIOScheduler with optimized job_defaults:
- misfire_grace_time: 3600s (increased from 120s) to prevent backlog
calculations that trigger memory leaks
- coalesce: true to collapse missed runs
- max_instances: 1 to prevent concurrent job execution
- replace_existing: true to avoid duplicate jobs on restart
3. Increase minimum job intervals:
- PROXY_BATCH_WRITE_AT: 30s (was 10s)
- add_deployment/get_credentials jobs: 30s (was 10s)
4. Use fixed intervals with small random offsets instead of jitter for
job distribution across workers
5. Explicitly configure jobstores and executors to minimize overhead
6. Disable timezone awareness to reduce computation
Memory impact:
- Before: 35GB with 483M allocations during startup
- After: <1GB with normal allocation patterns
Performance notes:
- Minimum job intervals increased from 10s to 30s (configurable via env vars)
- Jobs can still be distributed across workers using random start offsets
- No functional changes to job behavior, only timing and memory optimization
Testing:
- Added comprehensive test suite for scheduler configuration
- Verified no job execution backlog on startup
- Tested duplicate job prevention with replace_existing
Related issue: Memory leak in production proxy servers with APScheduler
\ud83e\udd16 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* docs: update PROXY_BATCH_WRITE_AT default value from 10s to 30s
Update documentation to reflect the new default value for PROXY_BATCH_WRITE_AT
changed in PR #15846. The default was increased from 10 seconds to 30 seconds
to prevent memory leaks in APScheduler.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* refactor: Move APScheduler config to constants.py
Address code review feedback from ishaan-jaff:
- Move scheduler configuration variables (coalesce, misfire_grace_time,
max_instances, replace_existing) to litellm/constants.py
- Update all references in proxy_server.py to use the constants
- Improves maintainability and makes configuration values centralized
Requested-by: @ishaan-jaff
Related: #15846🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
