jdx/mise · error

rpm -q failed: {}

Error message

rpm -q failed: {}

What it means

The dnf/rpm system package manager runs `rpm -q` to check installed packages. rpm normally reports not-installed packages on stderr ("package X is not installed", "no package X provided"); the library tolerates those lines but treats any other non-empty stderr as a real query failure and bails with this error containing the stderr text.

Source

Thrown at src/system/packages/dnf.rs:135

        let output = tokio::process::Command::new("rpm")
            .args(&args)
            .stdin(Stdio::null())
            .stdout(Stdio::piped())
            .stderr(Stdio::piped())
            .output()
            .await?;
        // rpm -q exits nonzero when any package is not installed; "package X
        // is not installed" goes to stdout or stderr depending on rpm version
        // and won't match the \t format either way — absent packages parse as
        // Missing. Only fail on rpm errors unrelated to missing packages.
        let stderr = String::from_utf8_lossy(&output.stderr);
        if !output.status.success()
            && !stderr.is_empty()
            && !stderr.lines().all(|l| {
                l.trim().is_empty() || l.contains("is not installed") || l.contains("no packages")
            })
        {
            bail!("rpm -q failed: {}", stderr.trim());
        }
        let stdout = String::from_utf8_lossy(&output.stdout);
        Ok(parse_rpm_query(&stdout, pkgs))
    }

    async fn install(&self, pkgs: &[PackageRequest], opts: &InstallOpts) -> Result<()> {
        let args = install_args(pkgs, opts);
        if opts.dry_run {
            miseprintln!("{}", sudo::argv("dnf", &args).join(" "));
            return Ok(());
        }
        sudo::run("dnf", &args, &[])
    }

    async fn upgrade(&self, pkgs: &[PackageRequest], opts: &InstallOpts) -> Result<()> {
        let args = upgrade_args(pkgs);
        if opts.dry_run {
            miseprintln!("{}", sudo::argv("dnf", &args).join(" "));

View on GitHub (pinned to afd2eddd3a)

Solutions

1. Read the stderr in the message to see the actual rpm failure and fix that underlying issue.
2. Run with appropriate privileges (root/sudo) if the rpm database is unreadable.
3. Rebuild the rpm database (rpm --rebuilddb) if it is corrupt or locked.
4. Verify rpm is correctly installed and the package names passed are valid rpm identifiers.

Example fix

// before: failing due to unreadable rpm db
rpm -q curl  # error: cannot open Packages database in /var/lib/rpm
// after
sudo rpm --rebuilddb && rpm -q curl

// Pre-check rpm availability and db readability before querying
if !Path::new("/usr/bin/rpm").exists() { return Err(anyhow!("rpm not available")); }
let db_ok = std::process::Command::new("rpm").args(["-q", "rpm"]).output()
    .map(|o| o.status.success()).unwrap_or(false);

match dnf_manager.installed(&pkgs).await {
    Err(e) if e.to_string().starts_with("rpm -q failed") => {
        eprintln!("rpm query failed: {e}; falling back to dnf list installed");
        dnf_list_installed(&pkgs).await
    }
    other => other,
}

When it happens

Trigger: `rpm -q` exits non-status-success AND stderr is non-empty AND at least one stderr line is neither blank nor contains "is not installed" / "no packages" — e.g. rpm database errors, permission problems, malformed package arguments, or rpm not functioning.

Common situations: Corrupt or locked RPM database (/var/lib/rpm) on a minimal container; running without root where the rpm db is unreadable; a requested package name with invalid characters that rpm rejects; SELinux or dbenv failures; partially installed rpm/dnf on non-RHEL systems.

Related errors

• flatpak {action} failed: {}
• {program} failed with {status}
• pacman -Qi failed: {}
• pacman -Q failed: {}
• ditto failed copying {} to {}

AI-assisted analysis of jdx/mise@afd2eddd3a (2026-09-09). Data as JSON: /api/errors/7691e51d2c3eca59. Report an issue: GitHub.