xai-org/grok-build · error

{}

Error message

{}

What it means

In batch_upload (public), each per-key S3 PutObject send() is mapped with anyhow::anyhow!(e), so any S3 SDK error surfaces with the SDK's Display message verbatim as message="{}". This wraps failures such as access denied, missing bucket, network errors, or oversized payloads for a single path in the batch.

Source

Thrown at crates/codegen/xai-file-utils/src/s3.rs:566

        let results: Vec<prod_mc_cli_chat_proxy_types::BatchUploadResult> = futures::stream::iter(
            files,
        )
        .map(|(path, content, content_type)| async move {
            let size = content.len() as i64;
            let bucket_owned = bucket.to_string();
            let upload_result = if content.len() >= MULTIPART_THRESHOLD {
                multipart_upload_bytes(client, bucket, &path, &content, &content_type).await
            } else {
                client
                    .put_object()
                    .bucket(bucket)
                    .key(&path)
                    .content_type(&content_type)
                    .body(aws_sdk_s3::primitives::ByteStream::from(content))
                    .send()
                    .await
                    .map(|_| ())
                    .map_err(|e| anyhow::anyhow!(e))
            };
            match upload_result {
                Ok(_) => prod_mc_cli_chat_proxy_types::BatchUploadResult {
                    path,
                    bucket: Some(bucket_owned),
                    status: prod_mc_cli_chat_proxy_types::BatchUploadStatus::Ok,
                    size: Some(size),
                    generation: None,
                    error: None,
                },
                Err(e) => {
                    let error_msg = format!("{:#}", e);
                    tracing::warn!(path = %path, error = %error_msg, "S3 batch upload item failed");
                    prod_mc_cli_chat_proxy_types::BatchUploadResult {
                        path,
                        bucket: Some(bucket_owned),
                        status: prod_mc_cli_chat_proxy_types::BatchUploadStatus::Error,
                        size: None,

View on GitHub (pinned to bc7f02eddd)

Solutions

1. Read the inner SDK error text to identify the S3 failure code (e.g. NoSuchBucket, AccessDenied) and fix the corresponding config.
2. Verify bucket name, region, and endpoint configuration passed to batch_upload.
3. Check AWS credentials and IAM permissions include s3:PutObject on the target bucket.
4. Add per-item retry or pre-check content size before issuing the PutObject.

// Pre-flight S3 config before batch_upload
assert!(!bucket.is_empty(), "bucket must be set");
assert!(content.len() <= 5 * 1024 * 1024 * 1024, "single PUT too large");
// credentials/region resolved by aws_config::load_from_env() in main

// Surface per-path batch failures with context
match batch_upload(&client, &bucket, items).await {
    Ok(results) => for r in results {
        if r.status != BatchUploadStatus::Ok {
            eprintln!("upload failed for {}: check S3 config/permissions", r.path);
        }
    },
    Err(e) => eprintln!("S3 upload failed: {e:#}"),
}

When it happens

Trigger: S3 PutObject failing for a given key/bucket: wrong bucket name, missing put_object permission, nonexistent region/endpoint, payload exceeding size limits, or network failure during send().await.

Common situations: Misconfigured AWS credentials or region; bucket not existing in the configured region; IAM policy lacking s3:PutObject; corporate proxy blocking S3 endpoints; uploading a body larger than the allowed single-PUT limit.

Related errors

• grove daemon unreachable
• {} failed: {}
• runtime-socket deny resolution failed: {error}
• OIDC expires_in out of range: {secs}
• OIDC endpoint rejected request ({status}): {body}

AI-assisted analysis of xai-org/grok-build@bc7f02eddd (2026-08-31). Data as JSON: /api/errors/508d757400d3f847. Report an issue: GitHub.