warpdotdev/warp · error · AgentDriverError::ConfigBuildFailed

Failed to write temp file '{prefix}': {e}

Error message

Failed to write temp file '{prefix}': {e}

What it means

Wrapped as AgentDriverError::ConfigBuildFailed when write_all on the just-created NamedTempFile fails. Creation succeeded, so this is a mid-write failure: ENOSPC when the disk fills while writing the prompt content, EIO on failing storage, or the temp file being invalidated (dir cleanup races, sandbox revocation).

Source

Thrown at app/src/ai/agent_sdk/driver/harness/mod.rs:580

///
/// Used by third-party harnesses to stage prompts / system prompts on disk
/// before launching the CLI, avoiding shell-quoting issues with complex input.
pub(super) fn write_temp_file(
    prefix: &str,
    content: &str,
    suffix: &str,
) -> Result<NamedTempFile, AgentDriverError> {
    let mut file = tempfile::Builder::new()
        .prefix(prefix)
        .suffix(suffix)
        .tempfile()
        .map_err(|e| {
            AgentDriverError::ConfigBuildFailed(anyhow::anyhow!(
                "Failed to create temp file '{prefix}': {e}"
            ))
        })?;
    file.write_all(content.as_bytes()).map_err(|e| {
        AgentDriverError::ConfigBuildFailed(anyhow::anyhow!(
            "Failed to write temp file '{prefix}': {e}"
        ))
    })?;
    Ok(file)
}

/// Upload a [`SerializedBlock`] as the JSON block snapshot for a third-party harness conversation.
pub(crate) async fn upload_block_snapshot(
    client: &dyn HarnessSupportClient,
    conversation_id: AIConversationId,
    block: SerializedBlock,
) -> Result<()> {
    log::info!("Uploading block snapshot for CLI agent to conversation {conversation_id}");
    let target = client
        .get_block_snapshot_upload_target(&conversation_id)
        .await
        .with_context(|| {
            format!("Unable to get block upload slot for conversation {conversation_id}")

View on GitHub (pinned to e72fd7aacb)

Solutions

  1. Free temp-directory disk space (df -h $TMPDIR) and retry — ENOSPC is the dominant cause given creation already succeeded.
  2. Move TMPDIR to a volume with ample free space for prompt-sized writes.
  3. If using network-backed /tmp, switch to local storage to avoid EIO/flush failures mid-write.

Example fix

# before
# disk nearly full: tempfile created, write_all fails with ENOSPC
warp agent run …

# after
df -h /tmp && docker system prune -a   # or: export TMPDIR=/data/tmp with free space
warp agent run …
Defensive patterns

Strategy: try-catch

Validate before calling

let needed = content.len() as u64 + 1024;
let avail = fs2::available_space(&std::env::temp_dir())?;
anyhow::ensure!(avail > needed, "temp dir needs ~{} bytes free for prompt staging", needed);

Try / catch

match write_temp_file(prefix, content, suffix) {
    Err(AgentDriverError::ConfigBuildFailed(err)) if err.to_string().contains("Failed to write temp file") => {
        // ENOSPC mid-write: free space or switch TMPDIR, then retry once
        std::env::set_var("TMPDIR", alternate_dir_with_space());
        write_temp_file(prefix, content, suffix)
    }
    rest => rest,
}

Prevention

When it happens

Trigger: Calling write_temp_file with large prompt/system-prompt content on a nearly-full disk — the create succeeds, then write_all hits ENOSPC partway; also flaky network-backed temp mounts (NFS /tmp) or macOS periodic tmp cleaners racing the write.

Common situations: Very large system prompts or context dumps exhausting remaining disk; CI containers with small overlay filesystems; temp dirs on network storage that drops mid-write.

Related errors

AI-assisted analysis of warpdotdev/warp@e72fd7aacb (2026-08-16). Data as JSON: /api/errors/c38a3914b35f3bdd. Report an issue: GitHub.