File Upload Issues

Supported file formats

The platform accepts common scientific, tabular and archive formats. Using the wrong format is the most common cause of upload failures.

Project file uploads

The upload dialog in a project workspace accepts:

Chat attachments

A file attached to a chat is also accepted as PDF, Word, PowerPoint or a raster image (PNG, JPEG, GIF, WebP, BMP, TIFF). Macro-enabled Office files, SVG, HTML and scripts are refused.

General limits

Files over the limit are rejected before the upload starts, so no bandwidth is wasted. CLI uploads have their own limits; see the CLI command reference.

File validation errors

The platform checks each file's type, size and name before it issues the upload URL. Tools parse the file's contents later, when a job reads it.

"File type … is not allowed" or "is not supported"

• The file's type isn't on the accepted list above. Check that the extension matches the content (for example, a .txt file that holds PDB data should be named .pdb).
• Double extensions such as .pdb.bak or .sdf.old aren't recognized. Remove the extra extension.
• "Is blocked for security reasons" means the type is refused outright (macro-enabled Office files, SVG, HTML, scripts, executables).

"Filename contains blocked extension"

• Chat attachments refuse any filename that contains .exe, .dll, .bat, .cmd, .sh, .ps1, .com, .msi, .vbs, .js or .hta anywhere in the name. The check matches text inside the name, not just the ending.
• ⚠︎ The check currently also catches harmless names: results.json contains .js, and run.shape.csv contains .sh. Rename the file (for example results_json.txt) or upload it to the project instead of attaching it to the chat.

"Filename contains invalid characters"

• The name contains .., / or \. Rename the file without them.

"File size exceeds maximum limit"

• The file is over the limit in the table above. Split it, or compress it into a .zip or .gz archive.

"Valid file size is required"

• The file has zero bytes. This usually means the export that created it failed silently. Re-export from the source application.

Invalid PDB format (reported by a tool)

• The file has a .pdb extension but no valid ATOM/HETATM records. Open it in a text editor and check that it contains ATOM lines.
• Some visualization tools export PDB files with non-standard formatting. Re-save through PyMOL, ChimeraX, or another structure editor.

No valid SMILES found (reported by a tool)

• For CSV input, make sure one column holds SMILES and has a clear header such as smiles.
• For .smi files, check that each line holds a valid SMILES string, and remove blank lines between entries.

Invalid molecule at a given row (reported by a tool)

• The SMILES on that row couldn't be parsed by RDKit. Common causes: unbalanced parentheses, invalid ring closures, or unsupported atom types. Fix or remove the row.
• Kekulized SMILES (explicit double bonds) can fail if the aromatic system is inconsistent. Convert to canonical SMILES with RDKit or Open Babel before uploading.

Encoding issues

Character encoding problems cause subtle failures where the file looks valid but its content is corrupted.

UTF-8 requirement:

• Save text-based files as UTF-8. Latin-1, Windows-1252 and other encodings can break parsing on special characters (accented letters in compound names, Greek symbols in sequence annotations).
• To check encoding on macOS/Linux: file -I yourfile.csv. It should say charset=utf-8.
• To convert: iconv -f WINDOWS-1252 -t UTF-8 input.csv > output.csv.
• On Windows, use Notepad++ to check and convert encoding: Encoding menu then Convert to UTF-8.

Line endings (CRLF vs LF):

• Mixed line endings in one file can confuse CSV parsers. If you see extra blank rows or garbled data, normalize line endings.
• On macOS/Linux: dos2unix yourfile.csv or sed -i 's/\r$//' yourfile.csv.
• On Windows, open in Notepad++ and select Edit then EOL Conversion then Unix.

Byte Order Mark (BOM):

• Some Windows applications (notably Excel "Save as CSV") prepend a UTF-8 BOM (\xEF\xBB\xBF), which can make the first column header unreadable.
• If the first column header isn't recognized despite looking correct, remove the BOM: sed -i '1s/^\xEF\xBB\xBF//' yourfile.csv.

Batch upload limits

When uploading several files at once:

• Up to 20 files per upload in the upload dialog. For more, run several uploads.
• Each file is checked on its own. If one file is refused, the others still upload.
• Drag and drop. You can drag files onto the upload area. Dragging a folder isn't supported, so select the files inside it instead.

Upload progress stalling

If the progress bar freezes during upload:

Network interruption:

• Each file goes to S3 in a single HTTPS request. If the connection drops (Wi-Fi switch, VPN reconnect), that file's upload fails and must be restarted.
• On an unstable connection, upload files one at a time rather than in a batch.

Browser tab inactive:

• Some browsers throttle network activity in background tabs. Keep the upload tab in the foreground until the upload completes.

Proxy or firewall interference:

• The browser uploads to an Amazon S3 address, not to ivybiosciences.com. A proxy that allows the site but blocks S3 makes every upload fail. Ask IT to allow HTTPS PUT requests to Amazon S3.
• Corporate proxies may cap request body size. Check that the cap is above 100 MB.
• If uploads always stall at the same percentage, a TLS-inspecting proxy may be timing out. Ask IT to raise the timeout or bypass inspection for S3.

Very large files (>50 MB):

• Files near the 100 MB limit can take several minutes on standard broadband. This is normal.
• The progress bar may pause near 100% while the platform records the finished upload. Wait for the confirmation message.

Storage quota

Uploaded files, job results and generated artifacts (reports, exports, passport PDFs) are stored in your workspace on Amazon S3.

Freeing up space:

• Export results you want to keep, then delete jobs and files you no longer need.
• Remove uploaded files that you no longer need from the project file manager.

If an upload is refused with a quota error, contact support with your plan and the error text.

File preparation best practices

Follow these guidelines to avoid common upload issues before they happen.

Filename conventions:

• Use ASCII letters, digits, hyphens, underscores and dots. Avoid accented characters, CJK characters and emoji.
• Replace spaces with hyphens or underscores: my-protein.pdb instead of my protein.pdb.
• Keep filenames under 100 characters for maximum compatibility.
• Don't use < > : " / \ | ? *.

Data file preparation:

• For CSV files, check for trailing commas at the end of lines. Some spreadsheet programs add them.
• Remove summary rows at the bottom of the file (totals, averages), which can confuse parsers.
• Put column headers on the first row. Files with metadata rows above the headers may not parse correctly.

Molecular file preparation:

• Validate PDB files (for example with the wwPDB validation service) before uploading.
• For SDF files, end each molecule record with $$$$ on its own line.
• Sanitize SMILES strings with RDKit or Open Babel before a batch upload to catch errors early.

When reporting upload issues

If none of the above resolves your problem, include the following in your support ticket:

• Example filenames and file sizes
• The exact error message and the time it occurred
• Whether it was a project upload, a chat attachment or a canvas upload
• Your browser and OS version
• Whether the file uploads successfully from a different browser or network
• A sample file (anonymized if it contains proprietary data) that reproduces the error
• The output of file -I yourfile.ext (macOS/Linux) to confirm encoding