cloud engineer
Here’s how to fix a duplicate UUID on a XFS file system.
mount -o rw,nouuid /dev/sda1 /data |
Mounting with nouuid will work, but not after a reboot.
To avoid a duplicate ID, run this command to generate a new UUID.
xfs_admin -U generate /dev/sda3 Clearing log and setting UUID writing all SBs new UUID = xxxxx-xxxx-xxxx-xxxxxxxxx |
Services to check.
chkconfig nfs on service rpcbind start service nfs start mount -a |
If missing, install it.
yum install nfs-utils nfs-utils-lib |
Mount it.
mount -a mount 192.168.1.200:/home /mnt/nfs/home |
Add to /etc/fstab.
10.10.10.10:/home on /mnt/nfs/home type nfs (rw,noatime,nolock,bg,nfsvers=2,intr,tcp,actimeo=1800,addr=10.10.10.10) |
Add a disk volume via cloud console. Then perform the following commands:
# check the new block lsblk # format fdisk /dev/sdc # create file system mkfs.xfs -f /dev/sdc1 # get uuid blkid /dev/sdc1 # vi /etc/fstab UUID="xxxxxxxxxxxxxxxxxxxxx" /mnt/directory xfs defaults 0 0 # mount mount -a |
Here’s the command to mount and unmount AWS EFS. Assuming the EFS IP address is 10.0.0.23.
Mount:
mount -t nfs -o nfsvers=4.1,rsize=1048576,wsize=1048576,hard,timeo=600,retrans=2,noresvport 10.0.0.23:/ /efs |
Umount:
# lazy unmount umount -l /efs # force unmount umount -f /efs |
You may have to force it if it’s busy.
If you want to auto mount the EFS volumes when the instance is created, add the following script to userdata:
#cloud-config repo_update: true repo_upgrade: all packages: - amazon-efs-utils runcmd: - file_system_id_01=fs-12345678 - file_system_id_02=fs-34593405 - efs_directory_01=/mnt/efs - efs_directory_02=/mnt/efs2 - mkdir -p ${efs_directory_01} - mkdir -p ${efs_directory_02} - echo "${file_system_id_01}:/ ${efs_directory_01} efs tls,_netdev" >> /etc/fstab - echo "${file_system_id_02}:/ ${efs_directory_02} efs tls,_netdev" >> /etc/fstab - mount -a -t efs defaults |
Just add more if you have more mounts.