Overview
VAST Cluster supports the block storage protocol. The block storage service lets you set up and manage paths on the cluster like block storage systems and split up the cluster’s space into block storage volumes under those paths.
VAST Cluster supports connectivity to block storage using Nonvolatile Memory Express over Transport Control Protocol (NVMe/TCP).
Block storage on the VAST platform is configured through subsystems (block-enabled views) that contain one or more volumes (NVMe namespaces). These volumes are mapped to initiator hosts using their NQN and accessed via a VIP pool, supporting multipath I/O for high availability and load balancing, similar to VIP usage for NFS in a multi-tenant environment.
Supported Client Environments
Linux: RHEL 8.8, RHEL 9.4, Ubuntu 24.04 LTS
VMware vSphere ESXi 7.0 U3, ESXi 8.0
Host Connectivity
NVMe/TCP (only)
Multipath Support
VAST supports native NVMe multipathing on Linux:
Enabled by default on RHEL 9+, optional in RHEL 8 (
nvme multipath).Clients should connect to all available VIPs in the pool.
VAST returns a maximum of 16 VIPs per subsystem on discovery
Note: Up to 16 paths (one path per virtual IP). During path discovery, up to 16 virtual IPs are chosen from the list of all possible virtual IPs across all virtual IP pools belonging to the relevant tenant.
BLOCK_MAX_HOST_TO_SUBSYSTEM_VIPS - 16 (min 1, max max-number-of-vips). Limit the number of vips to which a host connects to a subsystem.
If you want to connect to more than 16, you can, but you will need to connect manually to each vip.
e.g.
for i in {1..32}; do sudo nvme connect -a 172.21.192.$i -t tcp -q <host_nqn> -n <subsystem_nqn>; doneConfiguring an NVMe/TCP Client on Linux for VAST Cluster Block Storage
Installing the Client
To configure client block hosts for interacting with the cluster as a remote NVMe device, install the NVMe CLI tool
[root@centos8-var-1 ~]# sudo yum install nvme-cli
/usr/lib/python3.6/site-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.20) or chardet (3.0.4) doesn't match a supported version!
RequestsDependencyWarning)
Last metadata expiration check: 1:12:01 ago on Friday 01 August 2025 03:39:12 AM GMT.
Package nvme-cli-1.14-3.el8.x86_64 is already installed.
Dependencies resolved.
Nothing to do.
Complete!
[root@centos8-var-1 ~]#Verify the NVMe Version
[root@centos8-var-1 ~]# nvme version
nvme version 1.14
[root@centos8-var-1 ~]#Obtaining the Host NQN: Host NQN stands for Host NVMe Qualified Name. It's a unique identifier used by a Linux (or other OS) system when connecting to NVMe over Fabrics (NVMe-oF)
[root@centos8-var-1 ~]# cat /etc/nvme/hostnqn
nqn.2014-08.org.nvmexpress:uuid:ec2ace17-a853-4561-86dc-567f491521ab
[root@centos8-var-1 ~]#
Create a Block Storage Subsystem View
The components of a block storage configuration are:
Subsystems: A subsystem is an NVMe container for volumes or namespaces. It is configured with a unique NVMe identifier, an NQN, which makes it discoverable by hosts as a remote block storage device. A subsystem on the VAST Cluster is configured as a block-enabled view on a dedicated path.
Volumes: Units of capacity that can be provisioned on a subsystem. Each volume has a path relative to the subsystem path. The path can be backed up with snapshots. Snapshots of volumes can also be presented to hosts as volumes.
Hosts: NVMe/TCP hosts must be defined on the cluster with their NQN, a unique identifier created on each host by the NVMe/TCP client host configuration.
Block storage subsystems are provisioned on the cluster as views. To create a subsystem, create a view on an empty directory path and enable Block as the view's access protocol. Views that are block-enabled are configured on dedicated paths that are not accessible to other protocols.
To make volumes available to block hosts, they must be assigned or mapped to the hosts.
Procedure
In the VAST Web UI, select Element Store from the left navigation menu and then select Views.
Click Create View to add a new view.Select Policy Name
Select Block as a Protocol
SubSystem NQN: This would generate a Post specifying the SubSystem Name and create the Block View.
Up to 16 network paths are supported on client connections to block subsystem views. One virtual IP pool with a protocol role is used per path.
NOTE: The following are the only view policy settings that are relevant to block view configuration:
1. Tenant
2. VIP Pools
All security flavor options are compatible with block-enabled views. None of the remaining view policy settings affect block storage.
Right-click on the Block View → View. You will see the SubSystem NQN Generated.
Create a Block Volume
Navigate to the Volumes Tab in the Element Store. Click on Create Volume.
NOTE: The Volumes and Hosts tabs appear only when at least one block-enabled view is configured on the cluster.
You can choose to select Single Volumes or Multiple Volumes at Once. Specify the Volume Name and Capacity. Tags are Optional. Click Create
The Volume will be created
Creating Block Host
Navigate to Element Store → Host and Click Create Host.
Select either Single host to create one host or Multiple hosts to create multiple hosts.
Select Tenant
Enter the Name. The hostname must be unique on the Tenant.
Enter the host's NVMe Qualified Name (NQN), a unique identifier used to identify the host in NVMe operations.
NOTE: You can add the same host to multiple Tenants using its identifier.
Optionally, you can specify the tag
Click Create
The Host is added
Assigning Block Hosts to Volumes
Assigning a block host to a volume enables the host to discover the volume. Assignments are managed by a mapping of hosts to volumes on the same tenant.
Navigate to the Volumes Tab in the Element Store. Right-click the Volume and select Manage Mapping.
Select any hosts that you want to assign to the volume. Deselect any hosts that you want to unmap from the volume. Click Save Mapping.
You can view this on the Activities Page.
Once completed, you can see that the Volume is mapped to the Host.
CLI processes
You can list the Block Hosts as well as Volumes that were created using vcli
vcli: admin> blockhost list
+----+-----------+---------------+----------------------------------------------------------------------+-----------------+-------------+------------------------+---------------------+
| Id | Tenant-id | Name | Nqn | Tags | Tenant-name | Mapped-volumes-preview | Mapped-volume-count |
+----+-----------+---------------+----------------------------------------------------------------------+-----------------+-------------+------------------------+---------------------+
| 1 | 1 | centos8-var-1 | nqn.2014-08.org.nvmexpress:uuid:ec2ace17-a853-4561-86dc-567f491521ab | {'Dept': 'Dev'} | default | Volume1 (id=1) | 1 |
+----+-----------+---------------+----------------------------------------------------------------------+-----------------+-------------+------------------------+---------------------+
vcli: admin> volume list
+----+---------+---------+----------+-----------------+--------------+--------------------------------------+--------------------------------------+--------------+-----------------------------+-------------+----------------------------+-------------------------+
| Id | View-id | Name | Size(GB) | Tags | Namespace-id | Nguid | Uuid | Capacity(GB) | Snapshot-data | Tenant-name | Mapped-block-hosts-preview | Mapped-block-host-count |
+----+---------+---------+----------+-----------------+--------------+--------------------------------------+--------------------------------------+--------------+-----------------------------+-------------+----------------------------+-------------------------+
| 1 | 85 | Volume1 | 1.000 | {'Dept': 'Dev'} | 1 | 1257a64f-ff95-447f-8c87-26acc6fc700b | 1257a64f-ff95-447f-936c-a7acc6fc700b | 0.000 | {'id': None, 'name': 'N/A'} | default | centos8-var-1 (id=1) | 1 |
+----+---------+---------+----------+-----------------+--------------+--------------------------------------+--------------------------------------+--------------+-----------------------------+-------------+----------------------------+-------------------------+
To view the mapping of Volumes to the Block Host using vcli
vcli: admin> blockmapping list
+----+-----------+--------------+-----------------------------+
| Id | Volume id | BlockHost id | Snapshot-data |
+----+-----------+--------------+-----------------------------+
| 1 | 1 | 1 | {'id': None, 'name': 'N/A'} |
+----+-----------+--------------+-----------------------------+
vcli: admin> blockmapping show --id 1
+---------------+-----------------------------+
| Id | 1 |
| Volume id | 1 |
| BlockHost id | 1 |
| Snapshot-data | {'id': None, 'name': 'N/A'} |
+---------------+-----------------------------+
vcli: admin>
Connecting to Mapped Volumes from Host
After volumes have been mapped to the host via VMS, you can connect to them as follows.
Load the necessary kernel modules to enable NVMe over Fabrics.
[root@centos8-var-1 ~]# sudo modprobe nvme
[root@centos8-var-1 ~]# sudo modprobe nvme-fabrics
[root@centos8-var-1 ~]#To have the modules load automatically on reboot:
Create a file /etc/modules-load.d/nvme.conf and add the below entries inside.
nvme.conf
nvme
nvme-fabricsRun below.
The command sudo dracut -f in CentOS (or RHEL-based systems) is used to regenerate the initramfs (initial RAM filesystem) for the current kernel.
[root@centos8-var-1 ~]# sudo dracut -f
[root@centos8-var-1 ~]#Discover available NVMe subsystems over TCP
[root@centos8-var-1 ~]# sudo nvme discover -t tcp -a 172.27.33.1 -s 8009
Discovery Log Number of Records 4, Generation counter 47
=====Discovery Log Entry 0======
trtype: tcp
adrfam: ipv4
subtype: nvme subsystem
treq: not required
portid: 1
trsvcid: 4420
subnqn: nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
traddr: 172.27.33.1
sectype: none
=====Discovery Log Entry 1======
trtype: tcp
adrfam: ipv4
subtype: nvme subsystem
treq: not required
portid: 1
trsvcid: 4420
subnqn: nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
traddr: 172.27.33.2
sectype: none
=====Discovery Log Entry 2======
trtype: tcp
adrfam: ipv4
subtype: nvme subsystem
treq: not required
portid: 1
trsvcid: 4420
subnqn: nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
traddr: 172.27.33.3
sectype: none
=====Discovery Log Entry 3======
trtype: tcp
adrfam: ipv4
subtype: nvme subsystem
treq: not required
portid: 1
trsvcid: 4420
subnqn: nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
traddr: 172.27.33.4
sectype: none
[root@centos8-var-1 ~]#Connect to a subsystem:
Obtain the subsystem NQN from VMS:
In the VAST Web UI, open the Views tab on the Element Store page.
Find the subsystem view.
Right-click the view and select View to see its configuration.
The NQN is displayed in the Subsystem NQN field.
nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
Establish Connection to Subsystem
[root@centos8-var-1 ~]# sudo nvme connect -t tcp -n nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1 -a 172.27.33.2
[root@centos8-var-1 ~]#Run the connect-all command to connect all paths:
[root@centos8-var-1 ~]# sudo nvme connect-all -t tcp -a 172.27.33.1 -s 8009
To List the Subsystem Paths
[root@centos8-var-1 ~]# sudo nvme list-subsys
nvme-subsys0 - NQN=nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
\
+- nvme0 tcp traddr=172.27.33.2 trsvcid=8009 live
+- nvme1 tcp traddr=172.27.33.1 trsvcid=4420 live
+- nvme2 tcp traddr=172.27.33.2 trsvcid=4420 live
+- nvme3 tcp traddr=172.27.33.4 trsvcid=4420 live
+- nvme4 tcp traddr=172.27.33.3 trsvcid=4420 live
[root@centos8-var-1 ~]#
To see a list of connected NVMe volumes, use the nvme list command:
[root@centos8-var-1 ~]# sudo nvme list
Node SN Model Namespace Usage Format FW Rev
--------------------- -------------------- ---------------------------------------- --------- -------------------------- ---------------- --------
/dev/nvme0n1 VastData VastData 1 0.00 B / 1.00 GB 512 B + 0 B 24.05
/dev/nvme1n1 VastData VastData 1 0.00 B / 1.00 GB 512 B + 0 B 24.05
/dev/nvme2n1 VastData VastData 1 0.00 B / 1.00 GB 512 B + 0 B 24.05
/dev/nvme3n1 VastData VastData 1 0.00 B / 1.00 GB 512 B + 0 B 24.05
/dev/nvme4n1 VastData VastData 1 0.00 B / 1.00 GB 512 B + 0 B 24.05
[root@centos8-var-1 ~]#To get the list of block devices
[root@centos8-var-1 ~]# lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 100G 0 disk
├─sda1 8:1 0 600M 0 part /boot/efi
├─sda2 8:2 0 1G 0 part /boot
└─sda3 8:3 0 98.4G 0 part
├─cl-root 253:0 0 60.8G 0 lvm /
├─cl-swap 253:1 0 7.9G 0 lvm [SWAP]
└─cl-home 253:2 0 29.7G 0 lvm /home
sr0 11:0 1 9.3G 0 rom /run/media/vj/CentOS-8-4-2105-x86_64-dvd
nvme0n1 259:0 0 4.7G 0 disk
nvme1n1 259:1 0 4.7G 0 disk
nvme2n1 259:2 0 4.7G 0 disk
nvme3n1 259:3 0 4.7G 0 disk
nvme4n1 259:4 0 4.7G 0 disk
Why do I see Multiple NVMe Devices here when I have created just 1 Volume on VAST?
The reason you see multiple NVMe devices (like /dev/nvme0n1 to /dev/nvme4n1even though you created only one volume on VAST, it is because of multipathing in NVMe over Fabrics (NVMe-oF).
Each of those entries represents a different path to the same volume. They aren't different volumes. There are multiple connections (paths) to the same backend device
For each device, look for the NQN and Serial Number (SN) — they will be the same across all devices.
[root@centos8-var-1 ~]# sudo nvme id-ctrl /dev/nvme1n1 | grep subnqn
subnqn : nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
[root@centos8-var-1 ~]# sudo nvme id-ctrl /dev/nvme2n1 | grep subnqn
subnqn : nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
[root@centos8-var-1 ~]# sudo nvme id-ctrl /dev/nvme3n1 | grep subnqn
subnqn : nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
[root@centos8-var-1 ~]# sudo nvme id-ctrl /dev/nvme4n1 | grep subnqn
subnqn : nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1
[root@centos8-var-1 ~]#Run the command below. This tells us that the NVMe device /dev/nvme0n1 does not currently contain a recognizable filesystem or partition table. It's just raw data
[root@centos8-var-1 ~]# sudo file -s /dev/nvme0n1
/dev/nvme0n1: dataYou can now format it with ext4 or the IFS file system
[root@centos8-var-1 ~]# sudo mkfs.ext4 /dev/nvme0n1
mke2fs 1.45.6 (20-Mar-2020)
Discarding device blocks: done
Creating filesystem with 244140 4k blocks and 61056 inodes
Filesystem UUID: 5da976e8-8a76-49c4-a887-5ddc3c99e5a7
Superblock backups stored on blocks:
32768, 98304, 163840, 229376
Allocating group tables: done
Writing inode tables: done
Creating journal (4096 blocks): done
Writing superblocks and filesystem accounting information: done
[root@centos8-var-1 ~]#
Create a Mount Point
[root@centos8-var-1 ~]# sudo mkdir -p /mnt/nvme
[root@centos8-var-1 ~]#Mount the Volume
[root@centos8-var-1 ~]# sudo mount -o noatime,nodiratime /dev/nvme0n1 /mnt/nvme
[root@centos8-var-1 ~]#[root@centos8-var-1 ~]# df -h /mnt/nvme
Filesystem Size Used Avail Use% Mounted on
/dev/nvme0n1 923M 2.4M 857M 1% /mnt/nvme
[root@centos8-var-1 ~]#Basic NVMe Logs
This shows logs related to NVMe device detection, path creation, timeouts, and multipath activity.
[root@centos8-var-1 ~]# dmesg | grep -i nvme
[3453985.376141] TECH PREVIEW: NVMe/TCP may not be fully supported.
[3453985.390872] nvme nvme0: queue_size 128 > ctrl sqsize 32, clamping down
[3453985.391349] nvme nvme0: new ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery", addr 172.27.33.1:8009
[3453985.529705] nvme nvme0: Removing ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery"
[3454167.516170] nvme nvme0: queue_size 128 > ctrl sqsize 32, clamping down
[3454167.516717] nvme nvme0: creating 2 I/O queues.
[3454167.535001] nvme nvme0: mapped 2/0/0 default/read/poll queues.
[3454167.716701] nvme nvme0: new ctrl: NQN "nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1", addr 172.27.33.2:8009
[3454167.721640] nvme0n1: detected capacity change from 0 to 1000000000
[3454238.117614] nvme nvme1: queue_size 128 > ctrl sqsize 32, clamping down
[3454238.118137] nvme nvme1: new ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery", addr 172.27.33.2:8009
[3454238.119639] nvme nvme1: Removing ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery"
[3454238.323556] nvme nvme1: queue_size 128 > ctrl sqsize 32, clamping down
[3454238.324067] nvme nvme1: creating 2 I/O queues.
[3454238.335719] nvme nvme1: mapped 2/0/0 default/read/poll queues.
[3454238.337852] nvme nvme1: new ctrl: NQN "nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1", addr 172.27.33.1:4420
[3454238.342081] nvme1n1: detected capacity change from 0 to 1000000000
[3454238.346924] nvme nvme2: queue_size 128 > ctrl sqsize 32, clamping down
[3454238.347002] nvme nvme2: creating 2 I/O queues.
[3454238.363025] nvme nvme2: mapped 2/0/0 default/read/poll queues.
[3454238.365607] nvme nvme2: new ctrl: NQN "nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1", addr 172.27.33.2:4420
[3454238.373562] nvme2n1: detected capacity change from 0 to 1000000000
[3454241.446786] nvme nvme3: failed to connect socket: -110
[3454241.456412] nvme nvme3: queue_size 128 > ctrl sqsize 32, clamping down
[3454241.456500] nvme nvme3: creating 2 I/O queues.
[3454241.474347] nvme nvme3: mapped 2/0/0 default/read/poll queues.
[3454241.476391] nvme nvme3: new ctrl: NQN "nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1", addr 172.27.33.4:4420
[3454241.481066] nvme3n1: detected capacity change from 0 to 1000000000
[3454248.644095] nvme nvme4: queue_size 128 > ctrl sqsize 32, clamping down
[3454248.644242] nvme nvme4: new ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery", addr 172.27.33.1:8009
[3454248.645443] nvme nvme4: Removing ctrl: NQN "nqn.2014-08.org.nvmexpress.discovery"
[3454248.655353] nvme nvme4: queue_size 128 > ctrl sqsize 32, clamping down
[3454248.655439] nvme nvme4: creating 2 I/O queues.
[3454248.673276] nvme nvme4: mapped 2/0/0 default/read/poll queues.
[3454248.675387] nvme nvme4: new ctrl: NQN "nqn.2024-08.com.vastdata:5dee4d30-4e49-5d5b-a0cd-6a0325b23984:default:subsystem1", addr 172.27.33.3:4420
[3454248.678838] nvme4n1: detected capacity change from 0 to 1000000000
[3454985.277074] EXT4-fs (nvme0n1): mounted filesystem with ordered data mode. Opts: (null)
[root@centos8-var-1 ~]#
Check for Device Recognition
[root@centos8-var-1 ~]# dmesg | grep -i 'nvme[0-9]n1'
[3454167.721640] nvme0n1: detected capacity change from 0 to 1000000000
[3454238.342081] nvme1n1: detected capacity change from 0 to 1000000000
[3454238.373562] nvme2n1: detected capacity change from 0 to 1000000000
[3454241.481066] nvme3n1: detected capacity change from 0 to 1000000000
[3454248.678838] nvme4n1: detected capacity change from 0 to 1000000000
[3454985.277074] EXT4-fs (nvme0n1): mounted filesystem with ordered data mode. Opts: (null)
[root@centos8-var-1 ~]#Look for Errors/Timeouts
dmesg | grep -iE 'error|timeout|fail|nvme'Watch Logs Live while Connecting
dmesg -w | grep -i nvmeUnmount the Volume
[root@centos8-var-1 ~]# sudo umount /dev/nvme0n1To Expand a Volume on VAST
vcli: admin> volume modify --id 1 --size 5gb
vcli: admin> volume show --id 1
+----------------------------+--------------------------------------+
| Id | 1 |
| View-id | 85 |
| Name | Volume1 |
| Size(GB) | 5.000 |
| Tags | {'Dept': 'Dev'} |
| Namespace-id | 1 |
| Nguid | 1257a64f-ff95-447f-8c87-26acc6fc700b |
| Uuid | 1257a64f-ff95-447f-936c-a7acc6fc700b |
| Capacity(GB) | 0.210 |
| Snapshot-data | {'id': None, 'name': 'N/A'} |
| Tenant-name | default |
| Mapped-block-hosts-preview | centos8-var-1 |
| Mapped-block-host-count | None |
+----------------------------+--------------------------------------+Deleting a Volume
NOTE: Deleting a volume also deletes the volume directory and all data on it.
Volumes must be unmapped from hosts before being deleted.
Right-click the volume and select Remove.
You can choose to force unmap when you delete a single volume.
There is no force option for bulk volume deletion. You must unmap hosts from volumes before bulk deleting them.
Deleting a Snapshot Volume
To delete a snapshot volume, unmap the volume from any mapped hosts. When you unmap hosts from a snapshot volume, the snapshot volume is automatically deleted, but the volume itself, if it still exists, is not deleted.
As we can see below, the Remove Option for Snapshot Volume would always be greyed out.
If you try to Delete a Volume via vcli, it will throw an error
vcli: admin> volume delete --id 2
Are you sure you want to delete the volume? [y/N] y
Command REST Error: {"detail":"Volume is mapped to hosts","code":"bad_request"}
vcli: admin>
Deleting a Host
Hosts must be unmapped from volumes before being deleted. You can choose to force-unmap a host to delete it.
Deleting a Host from the VAST Web UI
Right-click the host and select Remove.
If you want to force the deletion even if the host is mapped to volumes, enable the Force slider.
Click Delete Host.
Data Protection
Create Protection Policy
A protection policy defines:
A schedule of points in time at which to take snapshots
Retention time for snapshots on the local cluster
A remote peer, which may be a replication peer (async replication) or an S3 replication peer (backup to S3). It is not specified whether the policy is intended for local backup only
Procedure
Navigate to Protection Policy.
Click Create Protection Policy
Enter Policy Name, Tenant, Snapshot Prefix, Time
The name of each snapshot will be <prefix>_<timestamp>, where <prefix> is the prefix specified here and <timestamp> is the time the snapshot is created, in the format
yyyy-mm-ddTHH:MM:SS.SSSSSSzzz(Tdenotes time and doesn't represent a value,zzzis the timezone, and the time is accurate to the microsecond). For example, if the prefix is dev, a snapshot taken at 8:15 pm UTC on 20th November 2024 would be named dev_2024-11-20T20:15:06.144783UTC.
If you want to make the protection policy indestructible, enable the Indestructible setting. This setting protects the policy and its snapshots from accidental or malicious deletion. Refer indestructibility-overview
NOTE: After saving the protection policy, you won't be able to delete the policy or disable its indestructibility without performing a procedure for authorized unlocking of the cluster's indestructibility mechanism
vcli: admin> protectionpolicy list
+----+-------------+--------------------------------------+------------------+-----------------+------------+----------------+-------------+--------------------+
| ID | Name | Schedule | Replication Peer | Snapshot Prefix | Clone Type | Indestructible | Tenant name | Remote tenant name |
+----+-------------+--------------------------------------+------------------+-----------------+------------+----------------+-------------+--------------------+
| 4 | BlockBackup | [{'every': '15m', | N/A | Dev | LOCAL | False | default | None |
| | | 'keep_local': '1H', | | | | | | |
| | | 'keep_remote': '0s', | | | | | | |
vcli: admin> protectionpolicy show --id 4
+--------------------+---------------------------------------------------------------------------------------------------+
| ID | 4 |
| Name | BlockBackup |
| Schedule | [{'every': '15m', 'keep_local': '1H', 'keep_remote': '0s', 'start_at': '2025-08-01 07:30:00'}] |
| Replication Peer | N/A |
| Snapshot Prefix | Dev |
| Clone Type | LOCAL |
| Indestructible | False |
| Tenant name | default |
| Remote tenant name | None |
+--------------------+---------------------------------------------------------------------------------------------------+Create Protected Path
Navigate to Data Protection → New Local Protected Path
Select the Tenant, Name, Path to be Protected, and Protection Policy. Click Create
vcli: admin> protectedpath list
+----+---------------+-------+--------+-------------+----------------+-------------------+------------------------+----------+-----------------+--------+----------------------+-------------+--------------------+---------------------+------------------------+--------------------------------+-------------------+----------------------------------------------+
| ID | Name | Role | State | Path | Path On Remote | Peer-cluster-name | Protection-policy-name | Is-local | Bandwidth(MB/s) | Health | Aggregated Usage(GB) | Tenant-name | Remote-tenant-name | Replication-streams | Sync Interval(Seconds) | Lease expiration time(Seconds) | Mode | Sync replication connection timeout(Seconds) |
+----+---------------+-------+--------+-------------+----------------+-------------------+------------------------+----------+-----------------+--------+----------------------+-------------+--------------------+---------------------+------------------------+--------------------------------+-------------------+----------------------------------------------+
| 2 | S3LocalBackup | Local | Active | /abhinavs3 | None | None | S3Backup | True | 0.000 | OK | None | default | None | ['S3LocalBackup'] | None | None | ASYNC_REPLICATION | None |
| 4 | Block | Local | Active | /test-block | None | None | BlockBackup | True | 0.000 | OK | 0.0 | default | None | ['Block'] | None | None | ASYNC_REPLICATION | None |
+----+---------------+-------+--------+-------------+----------------+-------------------+------------------------+----------+-----------------+--------+----------------------+-------------+--------------------+---------------------+------------------------+--------------------------------+-------------------+----------------------------------------------+
vcli: admin> protectedpath show --id 4
+----------------------------------------------+-------------------+
| ID | 4 |
| Name | Block |
| Role | Local |
| State | Active |
| Path | /test-block |
| Path On Remote | None |
| Peer-cluster-name | None |
| Protection-policy-name | BlockBackup |
| Is-local | True |
| Bandwidth(MB/s) | 0.000 |
| Health | OK |
| Aggregated Usage(GB) | 0.0 |
| Tenant-name | default |
| Remote-tenant-name | None |
| Replication-streams | ['Block'] |
| Sync Interval(Seconds) | None |
| Lease expiration time(Seconds) | None |
| Mode | ASYNC_REPLICATION |
| Sync replication connection timeout(Seconds) | None |
+----------------------------------------------+-------------------+If you now navigate to the Snapshots Tab in Data Protection, you will see a Snapshot being triggered as per the Schedule.
The Name of the Snapshot has a prefix followed by a timestamp.
To check the Snapshot List via vcli.
We have 1 Snapshot here for the Protected Block Path
vcli: admin> snapshot list
+----+------------------------------+--------------+-------------+-------+----------------+-----------------------------+----------------------+----------------------+------------------+-----------+-------------------+
| ID | Name | Path | Policy | Type | Indestructible | Created | Expiration-time | Aggregated Usage(GB) | Unique Usage(GB) | Tenant-id | Subsystem-related |
+----+------------------------------+--------------+-------------+-------+----------------+-----------------------------+----------------------+----------------------+------------------+-----------+-------------------+
| 2 | Dev_2025-08-01_06_40_00_UTC | /test-block/ | BlockBackup | local | False | 2025-08-01T06:40:00.006000Z | 2025-08-02T06:40:00Z | 0.0 | 0.0 | 1 | True |
| 1 | test_2025-08-01_06_30_00_UTC | /abhinavs3/ | S3Backup | local | False | 2025-08-01T06:30:00.010000Z | 2025-08-03T06:30:00Z | 0.0 | 0.0 | 1 | False |
+----+------------------------------+--------------+-------------+-------+----------------+-----------------------------+----------------------+----------------------+------------------+-----------+-------------------+
vcli: admin>
vcli: admin> snapshot show --id 2
+----------------------+-----------------------------+
| ID | 2 |
| Name | Dev_2025-08-01_06_40_00_UTC |
| Path | /test-block/ |
| Policy | BlockBackup |
| Type | local |
| Indestructible | False |
| Created | 2025-08-01T06:40:00.006000Z |
| Expiration-time | 2025-08-02T06:40:00Z |
| Aggregated Usage(GB) | 0.0 |
| Unique Usage(GB) | 0.0 |
| Tenant-id | 1 |
| Subsystem-related | True |
+----------------------+-----------------------------+
vcli: admin>
Presenting Volume to Snapshots
You can back up a block volume accessible to hosts by mapping the host to a snapshot of the volume. This can be any snapshot on the volume path or on a path that includes it. It can be a snapshot taken manually or from a local protected path.
When you assign hosts to a snapshot of a volume, VMS generates a globally unique UUID and a namespace ID unique within the subsystem, and assigns them to the mapping, enabling the assigned hosts to access the snapshot as a read-only volume. (These identifiers are destroyed if all hosts are unmapped from the snapshot.)
Snapshots that are mapped to hosts cannot be deleted.
If a volume is deleted, but you have the volume backed up by a snapshot, you can make the data available by mapping the volume to the host from the snapshot. You need to know the name of the deleted volume.
Procedure
In the Volumes tab, right-click the volume and select Manage Mappings.
Select the hosts that you want to map to the snapshot.
Click Available volume snapshots.
All available snapshots of the volume are shown. You can see the ID, Name, Path, and Creation Time of each snapshot.
Select the snapshot from which you want to map the volume to the selected host(s).
Click Save And Go To Mapping.
Click Save Mapping.
The mapping task begins, and you can track it on the Activities page.
Mapping a Volume from a Snapshot
In the Data Protection page, select the Snapshots tab.
Find the snapshot that you want to use. This can be a snapshot of the volume path, the subsystem path, or a path under the subsystem that includes the volume.
Right-click the snapshot and select Map Snapshot Volumes.
The Map Volumes dialog suggests volumes.
In the Map Volumes dialog, select the volume you want to map from the snapshot.
Click Next.
Select the volume path on the left and the host(s) on the right.
Click Save Mapping.
The mapping task begins, and you can track it on the Activities page.
If you navigate back to Volume in Element Store, you will see a snapshot of Volume1 being created.
You can see the Volume1 Snapshot entry via vcli as well. Under Snapshot Data, you would see the Snapshot used to Map to the Volume [ {'id': 2, 'name': 'Dev_2025-08-01_06_40_00_UTC'} ]
vcli: admin> volume list
+----+---------+---------+----------+-----------------+--------------+--------------------------------------+--------------------------------------+--------------+--------------------------------------------------+-------------+----------------------------+-------------------------+
| Id | View-id | Name | Size(GB) | Tags | Namespace-id | Nguid | Uuid | Capacity(GB) | Snapshot-data | Tenant-name | Mapped-block-hosts-preview | Mapped-block-host-count |
+----+---------+---------+----------+-----------------+--------------+--------------------------------------+--------------------------------------+--------------+--------------------------------------------------+-------------+----------------------------+-------------------------+
| 1 | 85 | Volume1 | 5.000 | {'Dept': 'Dev'} | 1 | 1257a64f-ff95-447f-8c87-26acc6fc700b | 1257a64f-ff95-447f-936c-a7acc6fc700b | 0.003 | {'id': None, 'name': 'N/A'} | default | centos8-var-1 (id=1) | 1 |
| 2 | 85 | Volume1 | 5.000 | {'Dept': 'Dev'} | 2 | a004fbab-2d88-486b-8c87-2676c4c38dfa | a004fbab-2d88-486b-a00f-1a76c4c38dfa | 0.003 | {'id': 2, 'name': 'Dev_2025-08-01_06_40_00_UTC'} | default | centos8-var-1 (id=1) | 1 |
+----+---------+---------+----------+-----------------+--------------+--------------------------------------+--------------------------------------+--------------+--------------------------------------------------+-------------+----------------------------+-------------------------+
vcli: admin> volume get_snapshots --id 1
+----+---------+------+----------+------+--------------+-------+------+--------------+---------------+-------------+----------------------------+-------------------------+
| Id | View-id | Name | Size(GB) | Tags | Namespace-id | Nguid | Uuid | Capacity(GB) | Snapshot-data | Tenant-name | Mapped-block-hosts-preview | Mapped-block-host-count |
+----+---------+------+----------+------+--------------+-------+------+--------------+---------------+-------------+----------------------------+-------------------------+
| 2 | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
| 3 | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
+----+---------+------+----------+------+--------------+-------+------+--------------+---------------+-------------+----------------------------+-------------------------+
vcli: admin>
Mapping a Volume from a Snapshot after the Snapshot was Deleted
In the Data Protection page, select the Snapshots tab.
Find the snapshot that you want to use. This can be a snapshot of the volume path, the subsystem path, or a path under the subsystem that includes the volume.
Right-click the snapshot and select Map Snapshot Volumes.
In the Map Volumes dialog, click Add volumes manually.
From the Subsystem dropdown, select the subsystem under which the volume existed before it was deleted.
In the Volume Name field, enter the volume name.
Click Add to table.
Select the newly added row in the table.
Click Next.
Select the volume path on the left and the host(s) on the right.
Click Save Mapping.
NVMe Block Commands
Command | What it Does | When / Why You Use It |
|---|---|---|
| Lists all NVMe devices and namespaces | First command to verify NVMe devices are detected by the OS |
| Lists NVMe devices in JSON format | Useful for automation, scripts, or structured debugging |
| Shows NVMe controller information | Check vendor, firmware, capabilities, queue depth |
| Shows namespace (block device) details | Verify size, LBA format, sector size |
| Displays SMART / health data | Check temperature, wear level, media errors |
| Shows recent NVMe errors | Debug I/O failures, timeouts, or controller errors |
| Retrieves log pages manually | Deep debugging when standard logs aren’t enough |
| Shows firmware slot and status | Verify active firmware and available slots |
| Formats the namespace | Destructive – used for reprovisioning or testing |
| Secure erase of device | Compliance or complete data wipe |
| Resets the NVMe controller | Recover from hung or unresponsive NVMe device |
| Rescans namespaces | When namespaces are added/removed dynamically |
| Shows NVMe subsystems | Debug multipath or NVMe-oF setups |
| Discovers NVMe-oF targets | Used in NVMe over Fabrics (TCP/RDMA) |
| Connects to NVMe-oF target | Attach remote NVMe storage |
| Disconnects NVMe-oF target | Clean up fabric connections |
| Real-time NVMe performance view | Live monitoring of I/O activity |
| Shows nvme-cli version | Ensure compatibility with kernel/features |
Command | Purpose |
|---|---|
| View block device hierarchy |
| Full block metadata (great for CSI debugging) |
| Identify filesystem / UUID |
| Udev attributes and symlinks |
| Inspect I/O scheduler and queue depth |
`dmesg | grep nvme` |
Metrics
Predefined Analytics Metrics
For Official Documentation on Block Support, refer to: Overview of block storage on VAST Cluster