podman-container-tools/podman · error

invalid value for XDG_RUNTIME_DIR: %m

Error message

invalid value for XDG_RUNTIME_DIR: %m

What it means

On the shortcut path the preamble formats '$XDG_RUNTIME_DIR/libpod/tmp/ns_handles' into a char path[PATH_MAX] (pkg/rootless/rootless_linux.c:842-847). If the formatted length reaches PATH_MAX, errno is set to ENAMETOOLONG and the message is printed (without a trailing newline) before exit. Since XDG_RUNTIME_DIR was already validated as set and non-empty, this means its value is absurdly long (~4000+ characters).

Source

Thrown at pkg/rootless/rootless_linux.c:845

      char gid_fmt[16];
      size_t len;
      int r;

      cwd = getcwd (NULL, 0);
      if (cwd == NULL)
        {
          fprintf (stderr, "error getting current working directory: %m\n");
          _exit (EXIT_FAILURE);
        }

      uid = geteuid ();
      gid = getegid ();

      len = snprintf (path, PATH_MAX, "%s/libpod/tmp/ns_handles", xdg_runtime_dir);
      if (len >= PATH_MAX)
        {
          errno = ENAMETOOLONG;
          fprintf (stderr, "invalid value for XDG_RUNTIME_DIR: %m");
          exit (EXIT_FAILURE);
        }

      if (set_ns_handles (path) == 0)
        goto joined;

      /* If the handle is stale, give up with the shortcut.  */
      if (errno == ESTALE)
        return;

      /* Fall back to pause.pid if:
         - ENOENT ns_handles file doesn't exist
         - EOPNOTSUPP kernel doesn't support open_by_handle_at
         - ENOSYS syscall not available
         - EPERM (could be seccomp when running in a container)
       */
      if (errno != ENOENT && errno != EOPNOTSUPP && errno != ENOSYS && errno != EPERM)
        {

View on GitHub (pinned to a2409076ef)

Solutions

1. Inspect the value: 'echo ${#XDG_RUNTIME_DIR}' - it should be small (e.g. /run/user/1000)
2. Fix the exporting script/config to the standard value: export XDG_RUNTIME_DIR=/run/user/$(id -u)
3. Log out and back in so systemd sets the variable correctly for the session
4. If a legitimate workflow needs long paths, report upstream - PATH_MAX limits apply

Example fix

# before
export XDG_RUNTIME_DIR="/tmp/$(printf 'a%.0s' {1..4200})"
podman ps   # invalid value for XDG_RUNTIME_DIR: File name too long

# after
export XDG_RUNTIME_DIR=/run/user/$(id -u)
podman ps

# Reject an over-long XDG_RUNTIME_DIR before podman builds paths from it
if [ -n "${XDG_RUNTIME_DIR:-}" ] && [ "${#XDG_RUNTIME_DIR}" -gt 3800 ]; then
  echo "XDG_RUNTIME_DIR is too long (${#XDG_RUNTIME_DIR} chars)" >&2
  exit 1
fi

When it happens

Trigger: A broken login/profile script, test harness, or fuzzing environment exporting an XDG_RUNTIME_DIR whose length plus the ~25-char suffix reaches 4096; any podman command from that environment as a non-root user.

Common situations: Almost never legitimate - normally XDG_RUNTIME_DIR is /run/user/<uid>. Seen with misconfigured CI env injection, corrupted env files, or deliberate stress tests of environment variables.

Related errors

• internal error: path too long
• cannot retrieve cmd line
• error getting current working directory: %m
• cannot set %s namespace
• fork: %m

AI-assisted analysis of podman-container-tools/podman@a2409076ef (2026-08-15). Data as JSON: /api/errors/ff3636d71741ddae. Report an issue: GitHub.