Files
open-webui/backend/open_webui/retrieval/vector
Classic298andGitHub ea1b58eb6a fix: apply the shared metadata size cap to the last four vector DB backends (#29502)
Chunk metadata inserted into the vector DB carries arbitrary client-
supplied fields (e.g. custom metadata from file uploads), so it needs
the same size cap, type coercion and null-byte sanitization every
other backend already applies through process_metadata(). Four
backends never called it: Qdrant, Qdrant multitenancy, Milvus
multitenancy and Oracle23ai, so an oversized or malformed metadata
blob went into those unfiltered.

Wire process_metadata() in at each backend's single insert/upsert
chokepoint, matching the pattern already used by the other eleven
backends.
2026-09-04 11:42:20 -04:00
..
…
…
2026-08-25 16:27:17 -04:00