Adding and Removing Storage to a DSX

Storage can be added to a DSX node after the initial installation have completed. The following steps need to be followed:

Add storage [External task]
Refresh the configuration [Product task]
Create logical volume [Product task]
Register the new volume [Product task]

Step 1: Add Storage

This step is done outside of product interface. This example adds storage using the VMware vCenter interface. DSX automatically detects this storage and adds the new block devices at the operating system level. Additional steps are required before the new storage is usable.

It is also possible to use a fiber channel storage area network (FC SAN) or an internet small computer systems interface storage area network (iSCSI SAN) as a method to add new storage. However, multi-pathing may not be supported. It is recommended for VMware environment that the Hypervisor is presenting the storage to the virtual machine in such a way that the Hypervisor is configured to support multi-pathing.

Step 2: Refresh the DataCore vFilO Configuration

On DataCore vFilO, log in as admin and execute node-refresh on the DSX where you added the new storage. The following options are available for node-refresh:

--rescan: Forces the DSX to discover new block devices. Does not affect existing, already discovered devices. This is a required option when adding new storage.
--id: ID of the DSX node. Use the node-list command to identify the correct DSX.

# node-refresh --id e3d46105-f27a-5c23-ada9-37d591fad283 --rescan
        Name:                    dsx1.selab.local
        Type:                    PD
        ID:                      e3d46105-f27a-5c23-ada9-37d591fad283
        HW State:                OK
        Node State:              MANAGED
        Node Mode:               ONLINE
        SW version:              3.0.0-9
        Management IP:           172.27.102.142/22
        Applicable Software:     3.0.0-9
        Vendor:                  VMware, Inc.
        Product name:            VMware Virtual Platform
        …

There is additional output from this command which is not shown here.

Step 3: Create New Logical Volume

Identify the added block device(s) first. In this example, the device added is /dev/sdd.

# drive-list --node-name dsx1.selab.local
        total 4
        Type:                    SSD
        ID:                      1d59f84c-b2ce-465e-a31f-7422c39cf46d
        Name:                    /dev/sda
        State:                   OK
        Type:                    SSD
        ID:                      10a02152-1c15-48cc-969d-7afa77a22bf9
        Name:                    /dev/sdb
        State:                   OK
        Type:                    SSD
        ID:                      bc320a77-4196-4c24-b8f3-02bac6f38a92
        Name:                    /dev/sdc
        State:                   OK
        Type:                    SSD
        ID:                      8695bebc-0d26-47fe-93e1-49938a01bbd6
        Name:                    /dev/sdd
        State:                   OK

Create a logical volume using the new device path with the –device-path option.

# logical-volume-create --node-name dsx1.selab.local --device-path /dev/sdd
        Type:                    LOGICAL_VOLUME
        ID:                      ab4d17e3-249a-421f-adec-943b1e339ae7
        Name:                    /hsvol02
        Exported Path:           /pd_export2
        IPs:                     [172.27.102.142/22]
        Node:                    dsx1.selab.local
        Usage:                   DS
        Reserved:                false
        Filesystem Type:         XFS
        Capacity:                [Total: 75GB]
        Hardware:
        [Type: SSD, ID: 8695bebc-0d26-47fe-93e1-49938a01bbd6, Name: /dev/sdd, State: OK, Capacity: 75GB, Product name:            Virtual disk
        Vendor:                  VMware]

Step 4: Register the New Volume

This step creates a new file system on the logical volume and adds the new volume into the available storage configuration. Depending on how the system objectives are configured, it is possible that data will be placed on the new volume as soon as this step is complete.

The logical-volume-name is found in the output from the previous step.

# volume-add --logical-volume-name /hsvol02 --node-name dsx1.selab.local
        ID:                      080ca7bd-afec-480d-90ac-08e09ba359b8
        Name:                    dsx1.selab.local::/hsvol02
        Internal ID:             16
        IPs:                     [172.27.102.142/22]
        Path:                    /hsvol02
        State:                   OK
        Oper state:              UP
        Node:                    dsx1.selab.local
        Capacity:                [Total: 75GB]
        Admin state:             UP
        Created:                 2018-05-09 17:28:13 UTC
        Modified:                2018-05-09 17:28:13 UTC
        Capabilities:            Max read IOPs:                     10258
        Max write IOPs:                    2731
        Min read latency (usec):           404.21286
        Min write latency (usec):          2150.21584
        Max read bandwidth (KB/s):         620782
        Max write bandwidth (KB/s):        243286
        Max NFS READ request size (B):     1048576
        Max NFS WRITE request size (B):    1048576
        Tolerable read latency (nanosec):  1000000000
        Tolerable write latency (nanosec): 1000000000
        Utilization threshold:             75%
        rootFileHandle:                    01000600ab4d17e3249a421fadec943b1e339ae7
        Availability:                      99%
        Durability:                        99.9%
        Encryption:                        false
        Clone:                             true
        Snapshot:                          false
        Locations:
        [Type: STORAGE_VOLUME, ID: 080ca7bd-afec-480d-90ac-08e09ba359b8, Name: dsx1.selab.local::/pd_export2]
        [Type: NODE, ID: e3d46105-f27a-5c23-ada9-37d591fad283, Name: dsx1.selab.local]
        [Type: VOLUME_GROUP, ID: c38cc380-6eb1-42a8-a8b6-58240446e79f, Name: all]

The new volume will show up in the UI after a short wait as the system refreshes itself.

Removing storage

Removing storage from the environment is a non-disruptive task and can be done at any time. It is important that these steps are followed to ensure continued access to the data.

Step 1. Identify the Volume

Identify the volume in the GUI and decommission the volume by pressing the Decommission icon under the actions column.

Data will non-disruptively be placed on other volumes and the volume will be emptied from data. If there is not enough capacity to re-locate all the data, the decommissioning process will not complete until additional capacity or a cloud/object volume has been added to the system.

Step 2. Delete the Volume

Once the decommissioning task has completed, the GUI will list the status as Decommissioned and the Delete button can be pressed under the actions column.

Step 3. Delete the Logical Volume (CLI only)

Login to the admin CLI and execute logical-volume-delete command. Ensure that the correct device path is specified by referencing information from drive-list or node-list --name <NODE NAME> output.

Note that volumes that are in use cannot be deleted and they have to be decommissioned first.