headroomlabs-ai/headroom · error · ValueError

Failed to decompress zstd request body: {exc}

Error message

Failed to decompress zstd request body: {exc}

What it means

Error "Failed to decompress zstd request body: {exc}" thrown in headroomlabs-ai/headroom.

Source

Thrown at headroom/proxy/helpers.py:2411

    """
    encoding = (request.headers.get("content-encoding") or "").lower().strip()
    raw = await request.body()

    if encoding in ("zstd", "zstandard"):
        try:
            import zstandard

            dctx = zstandard.ZstdDecompressor()
            reader = dctx.stream_reader(raw)
            raw = reader.read()
            reader.close()
        except ImportError:
            raise ValueError(
                "Request body is zstd-compressed but the 'zstandard' package is not installed. "
                "Install it with: pip install zstandard"
            ) from None
        except Exception as exc:
            raise ValueError(f"Failed to decompress zstd request body: {exc}") from exc
    elif encoding == "gzip":
        import gzip as _gzip

        try:
            raw = _gzip.decompress(raw)
        except Exception as exc:
            raise ValueError(f"Failed to decompress gzip request body: {exc}") from exc
    elif encoding == "deflate":
        import zlib

        try:
            raw = zlib.decompress(raw)
        except Exception as exc:
            raise ValueError(f"Failed to decompress deflate request body: {exc}") from exc
    elif encoding == "br":
        try:
            import brotli

View on GitHub (pinned to 322425c43b)

Solutions

1. Check the underlying exception {exc}: a corrupt or truncated zstd frame usually means a buggy client or a man-in-the-middle
2. Verify the client is actually sending zstd (not mislabeled content)
3. Retry the request; transient truncation during upload can corrupt the stream
4. If reproducible, capture the raw body and validate it with the `zstd` CLI

When it happens

Trigger: Raised when decompressing a zstd-compressed request body fails, typically due to a corrupt/truncated payload or a client mislabeling the content encoding.

Common situations: See trigger scenarios.

AI-assisted analysis of headroomlabs-ai/headroom@322425c43b (2026-08-15). Data as JSON: /api/errors/102738a888d8fa4a. Report an issue: GitHub.