Can Vectara report a bad file during upload?
Vectara reported files it could not process during upload. It returned errors for a truncated PDF and a password-protected PDF; the console also showed the truncated file's failure.
1 of 1 test case passed
Every test case in this scenario has a result.
The test case
Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on this test case.
Unprocessable file submitted through the normal upload pathPassEvidence
We submitted a truncated PDF and a password-protected PDF through the upload API.
API response
HTTP 400 { "field_errors": { "body.file": "File could not be parsed. It may be a corrupt archive or a malformed document." }, "request_id": "df999873-18a6-4dbf-a846-eb7f0b14c538" }
API response
HTTP 400 { "field_errors": { "body.file": "File is encrypted. Remove password before indexing." }, "request_id": "5fa477e8-14b0-4746-ad73-059bd6c012d9" }
Transcribed from Proof 3 · Screenshot
Everything the tool returned, in order (3)
- API response
Printed above.
- Upload-status table row
bad_01_truncated.pdf | Error: 400 | Failed to parse PDF: Could not find trailer entry while repairing ()
- API response
The final response, printed above.
| ✓ | Expected: A failure is reported for the affected document.Found: The truncated PDF received HTTP 400 and a console error, with nothing indexed for it. The password-protected PDF also received HTTP 400 with a file-specific reason. |
TC89 — Vectara — R6 Managed RAG v1 Operation: submit a genuinely unprocessable file through the normal upload path, read the ingestion status. File: bad_01_truncated.pdf (truncated PDF, no trailer — genuinely unparseable, not merely unusual) Endpoint: POST /v2/corpora/r6-managed-rag-ss-v1/upload_file ==================================================================================================== API response (4 identical requests, same file — showing message non-determinism) ==================================================================================================== Attempt 1: HTTP 400 {"field_errors": {"body.file": "File could not be parsed. It may be a corrupt archive or a malformed document."}, "request_id": "df999873-18a6-4dbf-a846-eb7f0b14c538"} Attempt 2: HTTP 400 {"field_errors": {"body.file": "Failed to parse PDF: Could not find trailer entry while repairing ()"}, "request_id": "0f2b02ef-014e-4274-acb1-fe9f1df00db9"} Attempt 3: HTTP 400 {"field_errors": {"body.file": "Failed to parse PDF: Could not find trailer entry while repairing ()"}, "request_id": "19a35069-94c5-4dea-830f-c08e396ba04e"} Attempt 4: HTTP 400 {"field_errors": {"body.file": "File could not be parsed. It may be a corrupt archive or a malformed document."}, "request_id": "8f02fe64-33f0-4c83-b1e4-f4beda66cddf"} All 4 attempts: HTTP 400, non-empty field_errors.body.file. Never a silent success, never HTTP 200/201. Wording varies (2 distinct messages observed) but the failure signal itself is 100% consistent. ==================================================================================================== Console UI response (same file, uploaded via the "Upload files" screen) ==================================================================================================== Uploaded 0 files with 1 error File name: bad_01_truncated.pdf Status: Error: 400 Failed to parse PDF: Could not find trailer entry while repairing () The UI surfaces the same failure signal live, per-file, in the upload-status table — not just the API response. Screenshot: tc89-vectara-console-upload-error.png ==================================================================================================== Alternate bad file — password-protected PDF (generated: A01 re-saved with pypdf, user password set, not present in the original fixture set) ==================================================================================================== POST /v2/corpora/r6-managed-rag-ss-v1/upload_file with bad_02_password_protected.pdf: HTTP 400 {"field_errors": {"body.file": "File is encrypted. Remove password before indexing."}, "request_id": "5fa477e8-14b0-4746-ad73-059bd6c012d9"} Even more specific than the truncated-file case: correctly diagnoses the exact problem (encryption) and tells the user exactly what to do about it. Second independent confirmation that Vectara reports per-document ingestion failures rather than claiming success or staying silent.
The requirement was to report a failure for the affected document. Both genuinely unprocessable files received document-specific failures rather than success or silence.
Error wording varied. Across four identical uploads of the truncated PDF, the API alternated between a general parsing error and a specific missing-trailer message. Each returned HTTP 400 with a reason.
Password guidance. For the password-protected PDF, the API identified encryption and said to remove the password before indexing.
Configuration and setup
How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.
- Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
- Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
- Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.
Where this sits in the benchmark
This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.
| Level | Name | Scope |
|---|---|---|
| Benchmark | Managed RAG → | 15 scenarios · 9 tools |
| Capability | Data Ingestion → | |
| Scenario | A bad file in the upload → | |
| Tool | Vectara → |
Global scenario definition → · Global capability definition → · Vectara product page →
History of this result
What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.
Act on this result
Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.
Verify the proof files
These files support this result. Open a file to inspect the original evidence.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.