Grading Checklist

• Identifies the NFS hard mount as the reason processes are stuck in uninterruptible sleep.
• Explains the difference between hard and soft NFS mounts and their behavior when the server is unreachable.
• Uses mount | grep nfs to check mount options.
• Checks stuck processes with ps aux | grep " D " and /proc/PID/wchan.
• Attempts umount -f /mnt/shared-data for force unmount.
• Knows that umount -l (lazy unmount) detaches the filesystem from the namespace, allowing new processes to proceed.
• Explains that lazy unmount does not kill stuck processes but prevents new ones from hanging.
• Recommends using soft,timeo=30,retrans=3 mount options to prevent future hangs.
• Mentions checking NFS server reachability with showmount -e or rpcinfo -p.
• Discusses the intr mount option (deprecated in modern kernels) and alternatives.
• Suggests using autofs for on-demand NFS mounting to reduce impact of server failures.
• Notes that a reboot may be the only way to clear processes stuck in D state if the server never comes back.