Manage filesystems using the CLI
This page describes how to view and manage filesystems using the CLI.
Using the CLI, you can perform the following actions:
Use this command to view information on the filesystems in the WEKA system.
weka fs create
Use the following command line to create a filesystem:
weka fs create <name> <group-name> <total-capacity> [--ssd-capacity <ssd-capacity>] [--thin-provision-min-ssd <thin-provision-min-ssd>] [--thin-provision-max-ssd <thin-provision-max-ssd>] [--max-files <max-files>] [--encrypted] [--obs-name <obs-name>] [--auth-required <auth-required>] [--data-reduction]
Name of the filesystem being created.
Name of the filesystem group to which the new filesystem is to be connected.
Total capacity of the new filesystem. Minimum value: 1GiB.
For tiered filesystems, this is the SSD capacity. If not specified, the filesystem is pinned to SSD. To set a thin provisioned filesystem the
SSD capacity is set to total capacity
For thin-provisioned filesystems, this is the minimum SSD capacity that is ensured to be always available to this filesystem. Must be set when defining a thin-provisioned filesystem. Minimum value: 1GiB.
For thin-proviosined filesystem, this is the maximum SSD capacity the filesystem can consume. The value cannot exceed the
Metadata allocation for this filesystem. Automatically calculated by the system based on the SSD capacity.
Encryption of filesystem
Object store name for tiering. Mandatory for tiered filesystems.
Enable data reduction. The filesystem must be non-tired and thin-provisioned. A license with data reduction is required.
Note: When creating an encrypted filesystem a KMS must be defined.
Note: To define an encrypted filesystem without a KMS, it is possible to use the
--allow-no-kmsparameter in the command. This can be useful when running POCs but should not be used in production since the security chain is compromised when a KMS is not used.
If filesystem keys exist when adding a KMS, they are automatically re-encrypted by the KMS for any future use.
To create a new filesystem, the SSD space for the filesystem must be free and unprovisioned. When using thin-provisioned filesystems, that might not be the case. SSD space can be occupied for the thin-provisioned portion of other filesystems. Even if those are tiered, and data can be released (to object-store) or deleted, the SSD space can still get filled when data keeps being written or rehydrated from the object-store.
To create a new filesystem, in this case, use the
weka fs reserveCLI command. Once enough space is cleared from the SSD (either by releasing to object-store or explicitly deleting data), it is possible to create the new filesystem using the reserved space.
weka fs update
Use the following command line to edit an existing filesystem:
weka fs update <name> [--new-name=<new-name>] [--total-capacity=<total-capacity>] [--ssd-capacity=<ssd-capacity>] [--thin-provision-min-ssd <thin-provision-min-ssd>] [--thin-provision-max-ssd <thin-provision-max-ssd>] [--max-files=<max-files>] [--auth-required=<auth-required>]
Name of the filesystem to edit.
New name for the filesystem
Total capacity of the edited filesystem
SSD capacity of the edited filesystem. Minimum value: 1GiB.
For thin-provisioned filesystems, this is the minimum SSD capacity that is ensured to be always available to this filesystem. Minimum value: 1GiB.
For thin-proviosined filesystem, this is the maximum SSD capacity the filesystem can consume. The value can cannot exceed the
Metadata limit for the filesystem
Determines if mounting the filesystem requires being authenticated to Weka (weka user login). Possible values:
weka fs delete
Use the following command line to delete a filesystem:
weka fs delete <name> [--purge-from-obs]
Name of the filesystem to delete.
For a tiered filesystem, if set, all filesystem data is deleted from the object store bucket.
purge-from-obswill remove all data from the object-store. This includes any backup data or snapshots created from this filesystem (if this filesystem has been downloaded from a snapshot of a different filesystem, it will leave the original snapshot data intact).
- If any of the removed snapshots have been (or are) downloaded and used by a different filesystem, that filesystem will stop functioning correctly, data might be unavailable and errors might occur when accessing the data.
It is possible to either un-tier or migrate such a filesystem to a different object store bucket before deleting the snapshots it has downloaded.