juicedata/juicefs · error

no mount command found for %s

Error message

no mount command found for %s

What it means

`getCmdMount` scans `ps` output for a mount command whose fields include the mount point. If the scan completes, no disambiguation error occurred, but no command line was captured (`cmd == ""`), the function returns this error, which `juicefs debug` surfaces as 'failed to get mount command'.

Source

Thrown at cmd/debug_unix.go:93

				if find {
					newCmd := strings.Join(fields[7:], " ")
					newUid, newPid, newPpid := strings.TrimSpace(fields[0]), strings.TrimSpace(fields[1]), strings.TrimSpace(fields[2])
					if newPid == ppid {
						return uid, pid, cmd, nil
					} else if pid == newPpid {
						return newUid, newPid, newCmd, nil
					} else {
						return "", "", "", fmt.Errorf("find more than one mount process for %s", mp)
					}
				}
				cmd = strings.Join(fields[7:], " ")
				uid, pid, ppid = strings.TrimSpace(fields[0]), strings.TrimSpace(fields[1]), strings.TrimSpace(fields[2])
				find = true
			}
		}
	}
	if cmd == "" {
		return "", "", "", fmt.Errorf("no mount command found for %s", mp)
	}
	return uid, pid, cmd, nil
}

View on GitHub (pinned to c9a67b23e8)

Solutions

1. Confirm the mount exists: `mount | grep <mp>` / `df <mp>`; if stale, unmount and remount.
2. Run `juicefs debug` in the same namespace/container as the mount process (e.g. `docker exec` into the container).
3. Re-run with sudo to see all users' processes.
4. If the mount is healthy, this only blocks the 'Mount Command' section — other debug output may still be usable; check earlier warnings.

Example fix

// before (mounted in container, debug on host)
$ juicefs debug /jfs
no mount command found for /jfs
// after
$ docker exec <ctr> juicefs debug /jfs

if fi, err := os.Stat(mp); err != nil || !isMounted(mp) { // e.g. check /proc/mounts
  return fmt.Errorf("%s is not mounted", mp)
}

null

err := runDebug(mp)
if err != nil && strings.Contains(err.Error(), "no mount command found") {
  // mount not visible in this namespace/user context; adjust context and retry
}

When it happens

Trigger: Running `juicefs debug <mp>` on Unix when no process line matches the mount point — the volume is not actually mounted, the mount was performed in a different PID/UTS namespace (container), or the process list is truncated/hidden from the current user.

Common situations: Debugging a path that isn't mounted (stale FUSE mountpoint); the mount runs inside a container while debug runs on the host or vice versa; running as unprivileged user unable to see another user's process command line.

Understand the failure class

Background: 'Could not be found', 'does not exist', 'not found in database': the resource-not-found family when an ID, slug, key, or URI lookup comes back empty — this error's family across 20 libraries.

Related errors

• failed to get mount command: %v
• find more than one mount process for %s
• failed to execute command `%s`: %v
• failed to execute command `%s`: %v
• failed to stat %s: %s

AI-assisted analysis of juicedata/juicefs@c9a67b23e8 (2026-09-06). Data as JSON: /api/errors/a74c630e300b85b4. Report an issue: GitHub.