Mount an NFS file system

Step 1: Install an NFS client

Before mounting an NFS file system on a Linux ECS instance, you must install an NFS client. The client only needs to be installed once per server.

1. Connect to the ECS instance. For more information, see Connect to an ECS instance.

2. Install an NFS client.

   Operating system	Installation command
   Alibaba Cloud Linux	sudo yum install nfs-utils
   CentOS
   Red Hat
   Ubuntu	Run the following commands in order: sudo apt-get update sudo apt-get install nfs-common
   Debian

3. (Optional) Configure the maximum number of concurrent NFS requests.

   Run the following command to change the maximum number of concurrent client requests to 128. For more information, see How do I modify the maximum number of concurrent NFS requests?.

   if (lsmod | grep sunrpc); then
   (modinfo sunrpc | grep tcp_max_slot_table_entries) && sysctl -w sunrpc.tcp_max_slot_table_entries=128
   (modinfo sunrpc | grep tcp_slot_table_entries) && sysctl -w sunrpc.tcp_slot_table_entries=128
   fi
   (modinfo sunrpc | grep tcp_max_slot_table_entries) && echo "options sunrpc tcp_max_slot_table_entries=128" >> /etc/modprobe.d/sunrpc.conf
   (modinfo sunrpc | grep tcp_slot_table_entries) && echo "options sunrpc tcp_slot_table_entries=128" >> /etc/modprobe.d/sunrpc.conf

Step 2: Mount the NFS file system

On Linux, you can mount a File Storage NAS NFS file system manually or automatically. Manual mounts are temporary and must be redone after a server reboot, whereas automatic mounts are persistent. To ensure the mount persists across reboots, we recommend configuring automatic mounting after a successful manual mount.

Manually mount the NFS file system

Mount the NFS file system on your Linux ECS instance by using the domain name of the mount target.

1. Mount the NFS file system.

   • If you are using a General-purpose NAS file system, run one of the following commands.

     NFSv3 protocol

     sudo mount -t nfs -o vers=3,nolock,proto=tcp,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,noresvport file-system-id.region.nas.aliyuncs.com:/ /mnt

     NFSv4.0 protocol

     sudo mount -t nfs -o vers=4,minorversion=0,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,noresvport file-system-id.region.nas.aliyuncs.com:/ /mnt                                

   • If you are using an Extreme NAS file system, run the following command.

     sudo mount -t nfs -o vers=3,nolock,noacl,proto=tcp,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,noresvport file-system-id.region.extreme.nas.aliyuncs.com:/share /mnt

   The following table describes the mount command options.

   Parameter	Description
   General-purpose NAS: file-system-id.region.nas.aliyuncs.com:/ /mnt Extreme NAS: file-system-id.region.extreme.nas.aliyuncs.com:/share /mnt	The format is <domain name of the mount target>:<NAS file system directory> <local mount path on the current server>. Replace the placeholders with the actual values for your environment. Domain name of the mount target: You can find the domain name of the mount target in the File Storage NAS console. Go to the File System List page, click Manage next to the target file system, and then go to the Mount Targets page. For more information, see Manage mount targets. NAS file system directory: General-purpose NAS: The root directory (/) or any subdirectory, such as /share. If you use a subdirectory, ensure that it already exists in the NAS file system. Extreme NAS: The path must start with /share. For example, /share or /share/subdir. If you specify a subdirectory, make sure that the subdirectory already exists in the NAS file system. Local mount path on the current server: any subdirectory on the Linux ECS instance (for example, /mnt). If you specify a subdirectory, make sure that it exists. Note If the local mount path is not empty, its contents are hidden after a successful mount, and the directory displays the data from the NAS file system. We recommend that you use an empty directory for mounting.
   vers	The protocol version of the file system. vers=3: Mounts the file system by using the NFSv3 protocol. vers=4: Mounts the file system by using the NFSv4 protocol. In this parameter, minorversion specifies the minor version number of the protocol. For example, NAS supports version 4.0 of the NFS v4 protocol. Therefore, when you mount a file system by using the NFS v4 protocol, the minor version number is 0. Note General-purpose NAS file systems support NFSv3 and NFSv4.0. Extreme NAS file systems support only NFSv3 and do not support NFSv4.
   rsize	Defines the maximum size of data blocks that the client can read from the file system. Recommended value: 1048576.
   wsize	Defines the maximum size of data blocks that the client can write to the file system. Recommended value: 1048576.
   hard	Without this option, the client may immediately return an error if File Storage NAS becomes unavailable and will not retry the operation. This can lead to data inconsistency or loss. We recommend that you enable this option.
   timeo	The time in deciseconds (0.1 seconds) that the NFS client waits for a response before it retries a request. Recommended value: 600 (60 seconds).
   retrans	The number of times the NFS client retries a request. Recommended value: 2.
   noresvport	Instructs the NFS client to use a new TCP port when a network connection is re-established. This ensures that the connection is not interrupted after a network failure and recovery. We recommend that you enable this option.

   Note

   • If you must change the timeout parameter (timeo), we recommend that you use a value of 150 or greater. The unit for the timeo parameter is 0.1 seconds, so a value of 150 represents 15 seconds.

   • If you need to change the I/O size parameters (rsize and wsize), we recommend that you use the maximum value (1048576) to avoid performance degradation.

   • These parameters are optional. If used, separate them with commas.

   Important

   • We do not recommend using the soft option because it risks data inconsistency.

   • To avoid performance degradation, do not use mount options other than the recommended defaults.

2. Verify the mount.

   • Run the following command:

     mount -l

   • Sample output:

     Output similar to the following indicates a successful mount.

     mqueue on /dev/mqueue type mqueue (rw,relatime)
     hugetlbfs on /dev/hugepages type hugetlbfs (rw,relatime)
     tmpfs on /run/user/0 type tmpfs (rw,nosuid,nodev,relatime,size=786256k,mode=700)
     29xxxxx.nas.aliyuncs.com:/ on / type nfs4 (rw,relatime,vers=4.0,rsize=1048576,wsize=1048576,namlen=255,hard,noresvport,proto=tcp,timeo=600,retrans=2,sec=sys,clientaddr=192.168.63.232,local_lock=none,addr=192.168.63.228)
     [root@iz0jlxxx ~]#

     After a successful mount, you can also run the df -h command to view the capacity information of the file system.

     [root@iZ0jl0xxx        ~]# df -h
     Filesystem                         Size  Used Avail Use% Mounted on
     29cxxx.nas.aliyuncs.com:/            40G  2.1G   36G   6% /
     devtmpfs                           3.8G     0  3.8G   0% /dev
     tmpfs                              3.8G     0  3.8G   0% /dev/shm
     tmpfs                              3.8G  424K  3.8G   1% /run
     tmpfs                              3.8G     0  3.8G   0% /sys/fs/cgroup
     tmpfs                              768M     0  768M   0% /run/user/0

   If the mount fails, troubleshoot the issue. For more information, see Troubleshoot file system mount failures.

3. After a successful mount, you can read from and write to the NAS file system on the Linux ECS instance.

   You can access and use the NAS file system like a regular directory. Example:

   [root@ixxx ~]# mkdir /mnt/dir1
   [root@ixxx ~]# mkdir /mnt/dir2
   [root@ixxx ~]# touch /mnt/file1
   [root@ixxx ~]# echo 'some file conent' > /mnt/file2
   [root@ixxx ~]# ls /mnt
   dir1  dir2  file1  file2  tmp

(Optional) Automatically mount the NFS file system

You can configure the /etc/fstab file on a Linux ECS instance to automatically mount an NFS file system when the instance restarts.

1. If you are using an Extreme NAS file system, perform the operations in this step.

   If you are using a General-purpose NAS file system, skip this step and go to Step 2.

   vi /etc/systemd/system/sockets.target.wants/rpcbind.socket

   Open the /etc/systemd/system/sockets.target.wants/rpcbind.socket file. You must comment out the IPv6-related rpcbind parameters (BindIPv6Only=ipv6-only, ListenStream=[::]:111, and ListenDatagram=[::]:111) and keep only the IPv4 listening address. Otherwise, the rpcbind service for NFS will fail to start automatically.

   Description=RPCbind Server Activation Socket
   [Socket]
   ListenStream=/var/run/rpcbind.sock
   # RPC netconfig can't handle ipv6/ipv4 dual sockets
   #BindIPv6Only=ipv6-only
   ListenStream=0.0.0.0:111
   ListenDatagram=0.0.0.0:111
   #ListenStream=[::]:111
   #ListenDatagram=[::]:111
   [Install]
   WantedBy=sockets.target

   If you configure automatic mounting on a CentOS 6.x system, you must also perform the following operations.

   1. Run the chkconfig netfs on command to ensure that the netfs service starts automatically on boot.

   2. Open the /etc/netconfig configuration file and comment out the inet6-related content as shown below. To disable the IPv6 protocol, add a # symbol to the beginning of the lines for the udp6 and tcp6 entries.

      #
      udp      tpi_clts      v    inet      udp     -    -
      tcp      tpi_cots_ord  v    inet      tcp     -    -
      #udp6    tpi_clts      v    inet6     udp     -    -
      #tcp6    tpi_cots_ord  v    inet6     tcp     -    -
      rawip    tpi_raw       -    inet      -       -    -
      local    tpi_cots_ord  -    loopback  -       -    -
      unix     tpi_cots_ord  -    loopback  -       -    -

2. Open the /etc/fstab configuration file and add a mount configuration.

   Replace the following variables in the configuration:

   • <file-system-id.region.nas.aliyuncs.com>: The NAS mount target address (for example, General-purpose NAS: 237*******-*****.cn-hangzhou.nas.aliyuncs.com, Extreme NAS: 00a*****-****.cn-hangzhou.extreme.nas.aliyuncs.com).

   • </mnt>: Replace with the local mount path. For example: /mnt.

   General-purpose NAS

   • To mount the file system by using the NFSv3 protocol:

     file-system-id.region.nas.aliyuncs.com:/ /mnt nfs vers=3,nolock,proto=tcp,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,_netdev,noresvport 0 0

   • To mount the file system by using the NFSv4.0 protocol:

     file-system-id.region.nas.aliyuncs.com:/ /mnt nfs vers=4,minorversion=0,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,_netdev,noresvport 0 0

   Extreme NAS

   file-system-id.region.extreme.nas.aliyuncs.com:/share /mnt nfs vers=3,nolock,noacl,proto=tcp,noresvport,_netdev 0 0

   For descriptions of the main parameters in the example, see the Mount Command Parameter Description Table. The other parameters are described as follows.

   Parameter	Description
   _netdev	Prevents the client from attempting to mount the file system before the network is ready.
   0 (first item after noresvport)	A non-zero value specifies that the dump utility should back up the file system. For a NAS file system, this value must be 0.
   0 (second item after noresvport)	Specifies the order for the fsck utility to check file systems at startup. For a NAS file system, this value must be 0 to disable the check.

3. Run the following commands to configure the startup file /etc/rc.local.

   [ ! -f /etc/rc.local ] && echo '#!/bin/bash' > /etc/rc.local; 
   echo "for ((i=1; i<=10; i++)); do if ping -c 1 -W 3 aliyuncs.com; then break; else sleep 1; fi; done" >> /etc/rc.local;
   echo "sleep 3; mount -a -t nfs" >> /etc/rc.local; 
   chmod +x /etc/rc.local

4. Run the reboot command to reboot the ECS instance.

   Important

   Restarting an ECS instance interrupts its services. We recommend performing this operation during off-peak hours to minimize disruption.

5. Verify that the automatic mount configuration is successful.

   Within one minute after the ECS restart, run the df -h command to confirm the mounted NAS file system.

FAQ

• How do I resolve file creation latency in an NFS file system?

• How do I resolve data write latency to an NFS file system?

• How do I resolve "Permission denied" errors when accessing an NFS file system?

• How do I handle data desynchronization when multiple ECS instances mount the same NFS file system?

References

• Mount a file system across VPCs or Alibaba Cloud accounts

• Access a file system from a data center

• Use POSIX ACLs to control access

• Use NFSv4 ACLs to control access

• General-purpose NAS backup

• Recycle bin

• Disaster recovery