The filesystem section of the openmediavault web interface is where you integrate disk volumes to be part of the server. Drives/filesystems that are not mounted through the web interface are not registered in the backend database, this means you cannot use volumes to create shared folders if they were not mounted properly. This is very important, users that come from an existing debian installation with filesystems already present in their fstab file will see that no volumes will be available for creating shared folders even if they are mounted. For the disks to be properly integrated it is better to delete all fstab lines except rootfs and swap, reboot your server and start mounting the disks through the web interface.
The mount process acts like many other services in openmediavault, first it writes a database entry in
config.xml, this entry contains essential information:
- UUID of the database object <uuid>
- Predictable device path of the filesystem <fsname>
- Target mount directory <dir>
- Filesystem options <opts>
- Filesystem type (EXT3, EXT4, etc.) <type>
You can inspect a mntent entry in
config.xmlit should look like this:
<mntent> <uuid>f767ee54-eb3a-44c5-b159-1840a289c84b</uuid> <fsname>/dev/disk/by-label/VOLUME1</fsname> <dir>/srv/dev-disk-by-label-VOLUME1</dir> <type>ext4</type> <opts>defaults,nofail,user_xattr,usrjquota=aquota.user,grpjquota=aquota.group,jqfmt=vfsv0,acl</opts> <freq>0</freq> <passno>2</passno> <hidden>0</hidden> </mntent>
With the mntent entry in
config.xml, omv-salt deploy run fstab script writes the appropriate line in
/etc/fstab. You can identify entries in
/etc/fstabcreated by the web interface by looking at «openmediavault» tags. It is important to mention to not alter the information in between these tags. If you delete or modify a fstab option (noexec or quota for example) the next time you mount a new disk into the server, omv-salt deploy run fstab will deploy the original value there again. If you need persistent change use environmental variables. Finally the backend will proceed to mount the filesystem. After this the volume is ready for creating shared folders.
- The resize button is used for expanding filesystems. This can occur if you decide to resize a disk partition or you have grown a RAID array by adding one or more disks.
- The delete button actually deletes filesystems, using wipefs -a. This will flush filesystem, raid or partition-table signatures (magic strings). Be careful using this. The button is disabled until the filesystem is actually unmounted.
- Disabled until you have deleted all shared folders associated with that volume. Unmount will remove the entry from
- Supported Filesystems
- openmediavault supports the following filesystems that can be mounted through the web interface:
- Creating multi device filesystems is not supported in the web interface. However you can add devices to your btrfs array in CLI and it will not present any problems.
- No extra features of btrfs are available in the webui like snapshots or subvolumes. Additional subvolumes will have either be mounted outside of the OMV fstab tags or manually add mntent entries to config.xml or use advanced configuration
- Support for zfs is available through ZoL an uses a third party plugin provided by omv-extras. The development of the plugin was done in conjunction with core of openmediavault, so new code was added in the filesystem backend to improve support for zfs. The plugin registers datasets and pools in the internal database so you can create shared folders for zfs volumes. The creation of zvols is automatically recognized by openmediavault so you can format them and mount them in the web interface. The iscsiplugin can also use these zvols block devices to export LUN’s.