nikivdev/code · error

'codanna mcp get_index_info' failed: {}

Error message

'codanna mcp get_index_info' failed: {}

What it means

capture_index_stats runs `codanna mcp get_index_info --json` and captures its output. If the command exits non-zero, its trimmed stderr is embedded in this bail! error. The stderr content after the colon is codanna's own diagnostic, so this error is a wrapper around whatever codanna reported.

Source

Thrown at src/indexer.rs:103

            "'codanna index' exited with status {}",
            status.code().unwrap_or(-1)
        );
    }
}

fn capture_index_stats(binary: &Path, project_root: &Path) -> Result<String> {
    println!("Fetching Codanna index metadata...");
    let output = Command::new(binary)
        .arg("mcp")
        .arg("get_index_info")
        .arg("--json")
        .current_dir(project_root)
        .output()
        .with_context(|| "failed to run 'codanna mcp get_index_info --json'")?;

    if !output.status.success() {
        let stderr = String::from_utf8_lossy(&output.stderr);
        bail!("'codanna mcp get_index_info' failed: {}", stderr.trim());
    }

    let json: serde_json::Value = serde_json::from_slice(&output.stdout)
        .with_context(|| "failed to parse JSON from 'codanna mcp get_index_info --json'")?;

    serde_json::to_string_pretty(&json).with_context(|| "failed to serialize Codanna stats payload")
}

fn persist_snapshot(
    project_root: &Path,
    binary: &Path,
    payload: &str,
    override_path: Option<PathBuf>,
) -> Result<PathBuf> {
    let db_path = override_path.unwrap_or_else(default_db_path);
    if let Some(parent) = db_path.parent() {
        fs::create_dir_all(parent)
            .with_context(|| format!("failed to create directory {}", parent.display()))?;

View on GitHub (pinned to a747e741ae)

Solutions

1. Read the stderr text in the error message — it contains codanna's actual diagnostic.
2. Verify your codanna version supports `codanna mcp get_index_info --json` by running it manually; upgrade if the subcommand is unknown.
3. Ensure `codanna init` and `codanna index` succeeded before capturing stats.
4. Run the command in the project root manually to reproduce and debug.

// Verify the subcommand exists on this codanna version
let probe = std::process::Command::new("codanna")
    .args(["mcp", "get_index_info", "--json"])
    .output()?;
if !probe.status.success() {
    eprintln!("get_index_info unsupported/failing: {}",
        String::from_utf8_lossy(&probe.stderr));
}

match run_indexer() {
    Ok(stats) => handle(stats),
    Err(e) if e.to_string().contains("get_index_info' failed") => {
        // stderr after the colon is codanna's own diagnostic
        eprintln!("stats capture failed: {e}; check codanna version supports mcp get_index_info");
    }
    Err(e) => return Err(e),
}

When it happens

Trigger: Running the indexer's stats capture step when `codanna mcp get_index_info --json` fails: unknown subcommand in old codanna versions, uninitialized index, or codanna runtime error.

Common situations: Installed codanna version predates the `mcp get_index_info` subcommand; index missing because `codanna index` never ran or failed; codanna crashes while reading index metadata.

Related errors

• 'codanna init' exited with status {}
• 'codanna index' exited with status {}
• Lin.app is not running
• gen agent list failed
• gen agent list failed: {}

AI-assisted analysis of nikivdev/code@a747e741ae (2026-09-01). Data as JSON: /api/errors/cf8f97aca083f23a. Report an issue: GitHub.