# NFS - Network File System 🖧

In a closed network (where you know every device), NFS is a fine choice. With a good network, throughput it disgustingly fast and at the same time less CPU intensive on the server. It’s very simple to set up and you can toggle readonly on shares you don’t need to be writeable. - NFS vs Samba

NFS & Symlink

Symbolic links only contain a path to another file or directory on the originating system where they’re being shared from. Unless you take care to make the links relative or to duplicate the same directory structures on remote systems as the originator of the share, they simply will not work.

This can be easily fixed: see symlink

Client Configuration 💻

Basics

List shared for a given server

$ showmount -e [server ip]

Mount Shared Directory (Manual)

$ sudo mkdir -p /nfs/backup
$ sudo mount 192.168.0.125:/mnt/Backup /nfs/backup

use permanent mount
use autofs / automount (see below)

Automount (systemd)

The main difference is that autofs, with the right auto scripts, will dynamically list the available shares. So you don’t need to pre-define and hard-code which machines/shares should be made available.

With systemd’s automount, only shares which you have pre-configured will be visible.

Automatic mounts with systemd
- Systemd automount vs autofs

Install

$ sudo apt install nfs-common

Autofs Setup

$ sudo apt-get install autofs

Add the following line at the end of /etc/auto.master

/nfs   /etc/auto.nfs	--ghost,--timeout=30

Create /etc/auto.nfs, and specify nfs mount options:

<server-name>   -fstype=nfs4,soft   <server-ip/host>:/
  
tronaut   -fstype=nfs4,soft   tronaut:/

Reload /etc/init.d/autofs

$ systemctl start autofs

# now test that it works
$ ls /nfs/tronaut/  # pressing tab should show shared available

see also FSCache to minimize network traffic for users accessing data from a file system mounted over the network.

Host Configuration 🖥

Install

$ sudo apt install nfs-kernel-server

sudo micro /etc/exports

directory_to_share    client(share_option1,...,share_optionN)

Then restart service: sudo systemctl restart nfs-kernel-server

Example

Setup user specific folder on NAS

On Server create ZFS dataset / NFS Share

$ zfs create -o mountpoint=/mnt/tronaut-yves storage_pool/tronaut-yves

# add it to NFS exports (/etc/exports)
$ sudo micro /etc/exports
# /mnt/tronaut-yves *(rw,sync,no_subtree_check,no_root_squash)

$ sudo systemctl restart nfs-kernel-server

Options details

** * ** Allow all clients to connect, otherwise you can specify clients (by IP, hostname, subnet or domain).

rw: This option gives the client computer both read and write access to the volume.

sync: This option forces NFS to write changes to disk before replying. This results in a more stable and consistent environment since the reply reflects the actual state of the remote volume. However, it also reduces the speed of file operations.

no_subtree_check: This option prevents subtree checking, which is a process where the host must check whether the file is actually still available in the exported tree for every request. This can cause many problems when a file is renamed while the client has it opened. In almost all cases, it is better to disable subtree checking.

no_root_squash: By default, NFS translates requests from a root user remotely into a non-privileged user on the server. This was intended as security feature to prevent a root account on the client from using the file system of the host as root. no_root_squash disables this behavior for certain shares.

How to properly set permissions for NFS folder?

Refs

Use fs-cache + cachefilesd with NFS

This can improve speed, principal goal is to decrease network traffic.

** It does not give offline access to data** - no onedrive.

Needs FS-Cache needs to be installed locally

$ sudo apt install cachefilesd

And it must be enabled /etc/default/cachefilesd -> RUN=yes and then sudo systemctl enable --now cachefilesd

Now when mounting nfs, use -o fsc to enables FS-Cache support sudo mount -o fsc yourserver:/yourpath /mnt/nfs

Permanent Mount

Avoid? Autofs mount for permanent mount

because if you do the content is unreliable, the mount point having been removed.
do not share mount point between autofs and cachedfs (do not use /nfs if this is the autofs mount point)

On Client

$ mkdir yves ~/Tronaut

# manual mount
$ sudo mount -t nfs4 -o fsc tronaut:/mnt/tronaut-yves /home/yves/Tronaut

or create a systemd mount unit

systemd .mount unit filenames must be the escaped version of the mount path.

$ systemd-escape -p --suffix=mount /home/yves/Tronaut
=> home-yves-Tronaut.mount

in /etc/systemd/system/home-yves-Tronaut.mount

# /etc/systemd/system/home-yves-Tronaut.mount
[Unit]
Description=Mount Tronaut NFS Share for yves
After=network-online.target
Wants=network-online.target

[Mount]
What=tronaut:/mnt/tronaut-yves
Where=/home/yves/Tronaut
Type=nfs4
Options=_netdev,fsc,auto,nofail,x-systemd.automount,x-systemd.device-timeout=10,timeo=14

[Install]
WantedBy=multi-user.target

$ sudo systemctl daemon-reexec
$ sudo systemctl enable home-tronaut-yves-mnt-share.mount
$ sudo systemctl start home-yves-Tronaut.mount

Config

sudo xed /etc/cachefilesd.conf

Default backend is in /var/cache/fscache

There are 6 control in /etc/cachefilesd.conf Configure Cache culling: brun N% (percentage of blocks) & frun N% (percentage of files): this describes the amount of free space and the number of available files. If these values cache rise above the set percentage, then culling is turned off
bcull N% (percentage of blocks) & fcull N% (percentage of files): describes the amount of available space or the number of files and if these values fall below the set limit, then cull is started. bstop N% (percentage of blocks) & fstop N% (percentage of files): here if the amount of available space or the number of available files in the cache falls below either of these limits then the allocation of disk space stops until the limits are raised above the set percentage.

Handling host unavailability

When an NFS share becomes unavailable, it will cause any application trying to access the files to hang (including ls).

systemd automount

hard retry failed requests indefinitely until the server responds — effectively blocking until success. timeo=5 Timeout in tenths of a second (i.e., 0.5 seconds).
retrans=2 Retry a request this many times before giving up.
intr The intr option in NFS mount options allows NFS requests to be interrupted if the server becomes unresponsive. This means that system calls like ls, cat, or even umount on the NFS mount can be interrupted with signals like Ctrl+C or SIGTERM.
x-systemd.device-timeout Configure how long systemd should wait for the mount command to finish before giving up on an entry from /etc/fstab. Specify a time in seconds or explicitly append a unit such as “s”, “min”, “h”, “ms”.

Use systemd’s automounting to mount only when accessed.

Example systemd mount unit

[Mount]
What=nfsserver:/export/path
Where=/mnt/nfs
Type=nfs
Options=_netdev,fsc,auto,nofail,x-systemd.device-timeout=10,soft,timeo=5,retrans=3,intr
TimeoutSec=30

Things to consider ⚠️

ssh keys will be shared ash well
- to be able to connect to other system own key need to be added to ~/.ssh/authorized_keys
firefox will refuse to start thinking it’s already started

It is advise to enable different config between the two system (missing application / different version)

consider using a dotfile manager (like chezmoi or yadm) to manage user-specific configs smartly.

# NFS - Network File System 🖧

NFS & Symlink

Client Configuration 💻

Basics

List shared for a given server

Mount Shared Directory (Manual)

Auto-mounting an NFS share

Automount (systemd)

Install

Autofs Setup

Host Configuration 🖥

Install

Example

Options details

see also

Refs

Use fs-cache + cachefilesd with NFS

Permanent Mount

Config

Handling host unavailability

systemd automount

Sharing Home folder 🚧

see also