NAS Cluster Configuration

The NAS implementation allows you to configure multiple types of clusters, depending on your configuration and requirement.

The primary type of cluster configuration is Scale-out (see Configure Scale-Out NAS Clusters), which allows load balancing based on DNS distribution. With the DNS load balancing, you can ensure load distribution across all nodes.

Another type of cluster configuration specific to a MDC pair is the NAS Failover (see Configure 2-Node NAS Clusters with NAS Failover (active backup)).

Finally, in a configuration with only a WorkFlow Extender, you can use the Haproxy load balancing to distribute the SMB connection across all nodes using HaProxy (see Configure a Load-Balancing Cluster using haProxy (Xcellis WorkFlow Extender)).

Note: For a list of all the Appliance Controller commands, see the Command Index.

Requirements

To configure a NAS cluster, you must meet the following minimum requirements.

  • Ensure that all NAS cluster nodes are licensed for and running StorNext NAS.
  • Designate the NAS cluster's preferred master node, to which the NAS clients connect to access shares.
  • Assign each of the NAS cluster's nodes an IP address that is within the same network and subnet used by the NAS clients — this network is typically the LAN client network.
  • Enable a hosted StorNext file system to maintain configuration information about the NAS cluster. Each node within the NAS cluster must be able to access this file system through the master node.

Note: To enable NAS cluster features, such as NAS failover or DNS load distribution, you need to take additional steps. See NAS Cluster Configuration.

Important

All NFSv3 clients must mount shares through a VIP from the VIP pool to ensure safe lock recovery during failover. Do not use the Master VIP.

Important

NFSv4 failover is supported with NFSv4 version 4.0. Lock failover does not work with NFSv4 version 4.1

Protocol Considerations

Both SMB and NFSv3 protocols are automatically enabled for NAS clusters. If your environment uses NFSv4, you must manually enable that version of the NFS protocol.

  1. Log in to the Appliance Controller CLI.

  2. Create the NAS Cluster

    Enabling a node makes the node a member of the NAS cluster. However, the node cannot actively take client connections until you join it to the NAS cluster.

    To enable a NAS cluster, you will need to do the following:

    • Enable a master node for the NAS cluster.

      When you enable a master node, the Appliance Controller issues the cluster a unique ID that is set to a uuid. This cluster ID is used to prevent non-master nodes from joining the wrong NAS cluster if the command is being issued from another non-master node. In such a case, the non-master node needs a way to identify the NAS cluster to join.

    • Enable additional nodes for the NAS cluster

     

    Enable NAS Cluster Nodes:

    1. Enter the following to enable the master node:

      nascluster enable master <ip_addr> <snfs_root_where_cluster_info_stored>

      The parameters are:

      <ip_addr>

      The IP address to use for the master node.

      Make sure this IP address is in the same network and subnet that the non-master nodes and clients use.

      <snfs_root_where_cluster_info_stored>

      The root directory for the StorNext file system that contains the NAS cluster configuration information.

      Example:

      > nascluster enable master 10.65.188.89 /stornext/snfs1

      Updating NAS cluster configuration ...

      Verifying NAS cluster configuration for 10.65.188.89 ...

      NAS cluster enable node 10.65.188.89 starting...

      Updating system NAS cluster configuration ...

      NFS: services (v4=yes, v3=yes, HA=no)

      Master node successfully enabled for NAS cluster using 10.65.188.89

    2. At the prompt, enter the following for each additional node to enable for the cluster:

      nascluster enable node <ip_addr>

      The parameter is:

      <ip_addr>

      The IP address to use for the non-master node.

      Make sure this IP address is in the same network and subnet that the master node and clients use.

      Example:

      > nascluster enable node 10.65.188.91

      Updating NAS cluster configuration ...

      Verifying NAS cluster configuration for 10.65.188.91 ...

      Node 10.65.188.91 successfully enabled for NAS cluster

    To configure and manage NAS clusters, you must assign a single virtual IP ()VIP) address to the master node. This NAS cluster VIP is always assigned to the active master node. So when you need to upgrade your NAS cluster or manage NAS settings for the cluster, you can do so by accessing the master node directly, which in turn pushes all updates to the additional cluster nodes.

    NAS Clusters that do not use Scale-Out NAS Clustering or DNS Load Distribution

    For SMB or NFSv4 environments with failover enabled, the NAS cluster VIP also facilitates NAS failover. In the event that the active master node becomes unavailable, the NAS VIP transfers NAS management services to the next available node in the NAS cluster, rendering that node the new active master node.

    NAS clusters that use Scale-Out NAS Clustering and DNS Load Distribution

    For scale-out NAS clusters, a VIP pool and DNS load distribution facilitate NAS failover.

    Before assigning a NAS VIP, review the following information.

    Important

    The VIP set for the NAS cluster is NOT the same as the VIP used for the StorNext network. Make sure the NAS VIP host name used for the NAS cluster is NOT the same host name used for the HA pair.

    • For the Xcellis WFD, you need to configure both nodes of the NAS cluster and the NAS VIP on the 10 GbE network port, which is typically the LAN client network. If you use another network port, the NAS cluster will fail to configure correctly.
    • The NAS VIP must be set within the same network and subnet used by the NAS cluster and NAS clients — this network is typically the LAN client network.
    • Enable global file locking for the StorNext file system. See the snfs_config MAN page, or search for "snfs_config − StorNext File System Configuration File" in the the StorNext MAN Pages Reference Guide for more information about the fileLocks parameter. Enable global file locking before setting the NAS VIP for the NAS cluster.
    • If you configure AD or OpenLDAP to authenticate users accessing your NAS cluster, you must add your cluster host name to the same DNS as your AD or OpenLDAP server. Otherwise, users authenticated through AD or OpenLDAP cannot access the NAS shares through the NAS cluster.

     

    1. Enter:

      nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]

      The parameters are:

      <master_virtual_ipaddr>

      The NAS VIP through which clients can access the NAS cluster. Keep in mind that the cluster's NAS VIP is assigned to the master node.

      Make sure this IP address is in the same network and subnet that the NAS cluster nodes and clients use.

      [<virtual_ipaddr> ...]

      (Optional) If you are creating a VIP pool, you can enter the VIPs to assign to each node within the cluster.

      You can also use the nascluster add virtual ipaddr <virtual_ipaddr> [<virtual_ipaddr> ...] command to create a VIP pool.

      Additional Information

      If you need to change a NAS VIP, re-issue the nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]command from the master node.

    When a node is joined to a NAS cluster, it is ready to take client connections, or to become the master node in the event of a NAS failover. To join nodes to a NAS cluster, you must issue the following command for each node that will be part of the cluster. All join commands can be issued from the master node.

    Cluster Synchronization

    When joining a node to a NAS cluster, the node is synchronized with the master node. This synchronization ensures that all nodes in the cluster are using the same authentication scheme, and that they all have the same shares configured.

    Keep in mind that StorNext NASsynchronizes authentication and share configuration information between all nodes in a NAS cluster. Changes made to a single node will not be synchronized between nodes unless you use the auth config or share commands. After you have created a NAS cluster, you can only execute the auth config or share commands from the master node.

    With NAS failover configured, if you connect to the NAS cluster through the NAS VIP, you will ensure that you are always connected to the master node.

     

    1. Enter:

      > nascluster join <snfs_root_where_cluster_info_stored> [ip_addr]

      The parameters are:

      <snfs_root_where_cluster_info_stored>

      The root directory for the StorNext file system that contains the NAS cluster configuration information.

      If there are multiple StorNext file systems configured for your environment, all nodes within the NAS cluster must use the same StorNext file system.

      [ip_addr]

      The IP address of the node being joined to the NAS cluster.

      • If you are entering this command from the node that is joining the cluster, you do not need to specify an IP address.
      • If you are entering this command from the master node, you need to specify the IP address of the node to join to the cluster.

      Make sure that the IP address assigned to the NAS cluster nodes are in the same network and subnet used by the NAS clients.

      > nascluster join /stornext/snfs1

      NAS Cluster join starting ...

      NAS Cluster ID mycluster found ...

      Preparing to join NAS cluster as node 10.65.188.89

      Applying NAS cluster join settings ...

      Updating system NAS cluster configuration ...

      NFS: services (v4=yes, v3=yes, HA=no)

      Check for master takeover ...

      Publish master configuration ...

      Cluster verification for 10.65.188.89 in-progress ...

      Node state: pnn:0 10.65.188.89 UNHEALTHY (THIS NODE), waiting ...

      Node state: pnn:0 10.65.188.89 OK (THIS NODE)

      Cluster verification of 10.65.188.89 successful ...

      Waiting for NAS Master VIP 10.65.166.184 to be reachable ...

      Join to host.example.com as 1234567-MVIP starting ...

      Using AD Computers OU "Computers" ...

      Verify now joined to host.example.com as 1234567-MVIP ...

      Clearing ID map cache ...

      Restart SMB services to join with host.example.com ...

      Successfully joined NAS cluster

      > nascluster join /stornext/snfs1 10.65.190.230

      NAS Cluster join starting ...

      NAS Cluster ID mycluster found ...

      Proxy join to 10.65.190.230 ...

      [10.65.190.230]: NAS Cluster join to cluster ID mycluster starting ...

      [10.65.190.230]: Preparing for NAS cluster join as ID 10.65.190.230

      [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

      [10.65.190.230]: Synchronization of local configuration with master 10.65.176.238 starting...

      [10.65.190.230]: Applying ads auth config sync settings ...

      [10.65.190.230]: Applying ads configuration settings ...

      [10.65.190.230]: Applying NAS cluster join settings ...

      [10.65.190.230]: Updating system NAS cluster configuration ...

      [10.65.190.230]: NFS: services (v4=yes, v3=yes, HA=no)

      [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

      [10.65.190.230]: Cluster verification for 10.65.190.230 in-progress ...

      [10.65.190.230]: Node state: pnn:1 10.65.190.230 UNHEALTHY (THIS NODE), waiting ...

      [10.65.190.230]: Node state: pnn:1 10.65.190.230 OK (THIS NODE)

      [10.65.190.230]: Cluster verification of 10.65.190.230 successful ...

      [10.65.190.230]: Verify now joined to host.example.com as 1234567-MVIP ...

      [10.65.190.230]: Clearing ID map cache ...

      [10.65.190.230]: Restart SMB services to join with host.example.com...

      Successfully joined NAS cluster

  3. Assign the NAS Cluster Hostname:

    After creating your NAS cluster, you need to assign it a hostname, which is required to enable both scale-out NAS failover and DNS load distribution — NAS clients use the cluster hostname to connect to the NAS cluster, and with DNS load distribution enabled, clients access a share by querying the NAS cluster hostname.

    Note: For a list of all the Appliance Controller commands, see the Command Index.

    Recommended Cluster Hostname Formats for DNS Services

    Enter a cluster hostname that is viable within the DNS environment of the NAS cluster and its clients. The NAS cluster's clients must be able to make DNS queries that resolve the cluster hostname, and the DNS environment must be managed such that the cluster hostname can be delegated as a zone by its parent domain. Use the following formats to configure a hostname depending on your environment's OS.

    Note: As of Appliance Controller 3.0.1, a FQDN is REQUIRED for the DNS zone on the system.

    macOS SMB

    smb://<cluster_hostname>/<share>

    NFS

    <cluster_hostname>:/<export_path_to_share>

    Enter the directory path to the share for the <export_path_to_share> parameter.

    Windows

    \\<cluster_hostname>\<share>

    For more information about DNS load distribution, see Scale-Out Task 4: Enable DNS Load Distribution.

    1. Enter the following to enable the master node:

      > nascluster set hostname <hostname> [force]

      Important

      • If you are not currently using DNS services, you still need to assign a NAS cluster hostname to identify the cluster, such as cluster.quantum.com.
      • The NAS cluster hostname must be the FQDN (fully-qualified domain name), in order to support valid load-balancing via DNS.

      Note: As of Appliance Controller 3.0.1, a FQDN is REQUIRED for the DNS zone on the system.

      CORRECT:

      nascluster set hostname mam-snfs-nas.mbcgroup.local

      NOT CORRECT:

      nascluster set hostname mam-snfs-nas

      The parameters are:

      <hostname>

      The hostname to assign to the NAS cluster.

      Based on whether you are using external or internal DNS services, we recommend the following cluster hostname assignments:

      • If you are using external DNS services, assign that DNS hostname to your NAS cluster.
      • If you are using internal DNS services, assign that DNS hostname to your NAS cluster
      • See the "Important" notes in the blue section above for additional information

      [force]

      • If this option is included, then the command is issued without confirmation.
      • If this option is not included, then you must confirm the command.

      Important

      If DNS services are currently in use, changing the hostname will break resolution for all existing clients.

  4. Add a NAS Cluster VIP Pool

    To enable scale-out NAS clustering, you must configure a NAS cluster VIP pool, which consists — at a minimum — of 1 VIP for each node in your cluster.

    The minimum number of VIPs in the pool should normally match the number of nodes in a cluster. In clusters with more than 2 nodes, you can benefit from having the number of VIPs to balance evenly across all the nodes (in the event of multiple failures).

    (Number of nodes) x (Number of nodes -1) = Number of VIP in the pool

    For example, if you have 4 nodes in your cluster, then your number of VIP is 12; with 12 VIP, you can evenly balance across 4 nodes, 3 nodes, 2 nodes, or 1 node.

    Note: You might not need the extra VIP for DNS load balancing.

    Note: For a list of all the Appliance Controller commands, see the Command Index.

    Scale-Out NAS Cluster = NAS Cluster VIP + VIP Pool

    The VIP pool enables NAS failover and DNS load distribution for scale-out NAS clusters.

    Keep in mind that the NAS cluster VIP assigned to the master node is not part of the VIP pool, but instead it is used to access the NAS cluster for maintenance and upgrades, as well as to support the DNS nameserver.

    Clients access the NAS cluster to mount shares, as follows:

    1. Clients query the NAS cluster's DNS name, which is also the cluster's hostname
    2. The NAS cluster's DNS server returns available VIPs from the VIP pool to the client.
    3. The client selects the VIP to which to connect.
    4. If the selected node fails, NAS failover services transfers the client connection and associated VIP to a different node.
    5. When the node comes back online, NAS failover services transfers one or more available VIPs to that node.

      Example

      Table 1: Scale-Out NAS Cluster Configuration

      Node

      VIPs

      Master Node

      dns.name, vip-1

      Node 2

      vip-2

      Node 3

      vip-3

      1. A client connects to the NAS cluster through its DNS name, dns.name, to mount a share.
      2. The DNS server presents vip-2, vip-3, and vip-1 to the client as available connections.
      3. The client selects vip-2 since it is the first VIP presented in the returned addresses, and it connects directly to node 2.
      4. Node 2 unexpectedly loses services.
      5. NAS failover transfers vip-2 and the client connection to node 3 so that the client can still access the share without interrupted services.
      6. When node 2 comes back online, one or more available VIPs are transferred to the node.

    Note: If you choose not to enable DNS load distribution, you can use the VIP pool to assist with load distribution. However, if you use this configuration, you must associate each VIP in the VIP pool with a hostname.

     

    Before adding a NAS VIP pool, review the following information.

    • You can set the NAS VIP pool to any network/subnet of the cluster nodes. If all the VIP are not on the same network, then DNS load balancing cannot guarantee to provide the optimal IP for a given client.

    • If you configure AD or OpenLDAP to authenticate users accessing your NAS cluster, you must add your cluster host name to the same DNS as your AD or OpenLDAP server. Otherwise, users authenticated through AD or OpenLDAP cannot access the NAS shares through the NAS cluster.
    1. Enter:

      > nascluster add virtual ipaddr <virtual_ipaddr> [<virtual_ipaddr> ...]

      The parameter is:

      <virtual_ipaddr> [<virtual_ipaddr> ...]

      The VIPs to assign to each node within the NAS cluster.

      You must assign at least 1 VIP to each node within the cluster. However, we recommend assigning multiple VIPs for each node within the cluster to get better redistribution results after a NAS failover occurs.

      2. Example:

      > nascluster add virtual ipaddr 10.65.164.76 10.65.164.77

      Output (nascluster show):

      > nascluster show

      NAS Cluster IP: 192.168.5.232/bond0:2, Master: Yes, SNFS Root: /stornext/snfs1, Joined: Yes
ID: dfff3314-16bd-11ea-ba5d-ecf4bbe6cf34
Cluster Hostname: vip-pool.example.com
DNS Enabled: No
Load balancing: Proxy-Disabled
NFS-HA(NFSv3): Enabled
Cluster VIP: 192.168.5.230 (active) nas-vip.private.example.com
Scale out VIP pool:
  VIP: 10.65.164.76[em2] (active, node:2) vip-pool.example.com
  VIP: 10.65.164.77[em2] (active, node:1) vip-pool.example.com
Nodes: 2
  *1: 192.168.5.232 (Joined, MDC) mdc1.private.example.com
   2: 192.168.5.231 (Joined, MDC) mdc2.private.example.com
    2. You can add VIPs from multiple networks, if additional interfaces are available. To add additional VIPs to the pool from other networks, add them the same way you just added the initial VIPs to the VIP pool. Enter:

      3. > nascluster add virtual ipaddr <virtual_ipaddr> [<virtual_ipaddr> ...]

      Example:

      > nascluster add virtual ipaddr 192.168.10.10 192.168.10.11

      Output (nascluster show):

      > nascluster show

      NAS Cluster IP: 192.168.5.232/bond0:2, Master: Yes, SNFS Root: /stornext/snfs1, Joined: Yes
ID: dfff3314-16bd-11ea-ba5d-ecf4bbe6cf34
Cluster Hostname: vip-pool.example.com
DNS Enabled: No
Load balancing: Proxy-Disabled
NFS-HA(NFSv3): Enabled
Cluster VIP: 192.168.5.230 (active) nas-vip.private.example.com
Scale out VIP pool:
  VIP: 10.65.164.76[em2] (active, node:2) vip-pool.example.com
  VIP: 10.65.164.77[em2] (active, node:1) vip-pool.example.com
  VIP: 192.168.10.10[bond1] (active, node:2) pool.private.example.com
  VIP: 192.168.10.11[bond1] (active, node:1) pool.private.example.com
Nodes: 2
  *1: 192.168.5.232 (Joined, MDC) mdc1.private.example.com
   2: 192.168.5.231 (Joined, MDC) mdc2.private.example.com

Before you begin, Log in to the Appliance Controller CLI.

Create the NAS Cluster:

Enabling a node makes the node a member of the NAS cluster. However, the node cannot actively take client connections until you join it to the NAS cluster.

To enable a NAS cluster, you will need to do the following:

  • Enable a master node for the NAS cluster.

    When you enable a master node, the Appliance Controller issues the cluster a unique ID that is set to a uuid. This cluster ID is used to prevent non-master nodes from joining the wrong NAS cluster if the command is being issued from another non-master node. In such a case, the non-master node needs a way to identify the NAS cluster to join.

  • Enable additional nodes for the NAS cluster

 

Enable NAS Cluster Nodes:

  1. Enter the following to enable the master node:

    nascluster enable master <ip_addr> <snfs_root_where_cluster_info_stored>

    The parameters are:

    <ip_addr>

    The IP address to use for the master node.

    Make sure this IP address is in the same network and subnet that the non-master nodes and clients use.

    <snfs_root_where_cluster_info_stored>

    The root directory for the StorNext file system that contains the NAS cluster configuration information.

    Example:

    > nascluster enable master 10.65.188.89 /stornext/snfs1

    Updating NAS cluster configuration ...

    Verifying NAS cluster configuration for 10.65.188.89 ...

    NAS cluster enable node 10.65.188.89 starting...

    Updating system NAS cluster configuration ...

    NFS: services (v4=yes, v3=yes, HA=no)

    Master node successfully enabled for NAS cluster using 10.65.188.89

  2. At the prompt, enter the following for each additional node to enable for the cluster:

    nascluster enable node <ip_addr>

    The parameter is:

    <ip_addr>

    The IP address to use for the non-master node.

    Make sure this IP address is in the same network and subnet that the master node and clients use.

    Example:

    > nascluster enable node 10.65.188.91

    Updating NAS cluster configuration ...

    Verifying NAS cluster configuration for 10.65.188.91 ...

    Node 10.65.188.91 successfully enabled for NAS cluster

To configure and manage NAS clusters, you must assign a single virtual IP ()VIP) address to the master node. This NAS cluster VIP is always assigned to the active master node. So when you need to upgrade your NAS cluster or manage NAS settings for the cluster, you can do so by accessing the master node directly, which in turn pushes all updates to the additional cluster nodes.

NAS Clusters that do not use Scale-Out NAS Clustering or DNS Load Distribution

For SMB or NFSv4 environments with failover enabled, the NAS cluster VIP also facilitates NAS failover. In the event that the active master node becomes unavailable, the NAS VIP transfers NAS management services to the next available node in the NAS cluster, rendering that node the new active master node.

NAS clusters that use Scale-Out NAS Clustering and DNS Load Distribution

For scale-out NAS clusters, a VIP pool and DNS load distribution facilitate NAS failover.

Before assigning a NAS VIP, review the following information.

Important

The VIP set for the NAS cluster is NOT the same as the VIP used for the StorNext network. Make sure the NAS VIP host name used for the NAS cluster is NOT the same host name used for the HA pair.

  • For the Xcellis WFD, you need to configure both nodes of the NAS cluster and the NAS VIP on the 10 GbE network port, which is typically the LAN client network. If you use another network port, the NAS cluster will fail to configure correctly.
  • The NAS VIP must be set within the same network and subnet used by the NAS cluster and NAS clients — this network is typically the LAN client network.
  • Enable global file locking for the StorNext file system. See the snfs_config MAN page, or search for "snfs_config − StorNext File System Configuration File" in the the StorNext MAN Pages Reference Guide for more information about the fileLocks parameter. Enable global file locking before setting the NAS VIP for the NAS cluster.
  • If you configure AD or OpenLDAP to authenticate users accessing your NAS cluster, you must add your cluster host name to the same DNS as your AD or OpenLDAP server. Otherwise, users authenticated through AD or OpenLDAP cannot access the NAS shares through the NAS cluster.

 

  1. Enter:

    nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]

    The parameters are:

    <master_virtual_ipaddr>

    The NAS VIP through which clients can access the NAS cluster. Keep in mind that the cluster's NAS VIP is assigned to the master node.

    Make sure this IP address is in the same network and subnet that the NAS cluster nodes and clients use.

    [<virtual_ipaddr> ...]

    (Optional) If you are creating a VIP pool, you can enter the VIPs to assign to each node within the cluster.

    You can also use the nascluster add virtual ipaddr <virtual_ipaddr> [<virtual_ipaddr> ...] command to create a VIP pool.

    Additional Information

    If you need to change a NAS VIP, re-issue the nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]command from the master node.

When a node is joined to a NAS cluster, it is ready to take client connections, or to become the master node in the event of a NAS failover. To join nodes to a NAS cluster, you must issue the following command for each node that will be part of the cluster. All join commands can be issued from the master node.

Cluster Synchronization

When joining a node to a NAS cluster, the node is synchronized with the master node. This synchronization ensures that all nodes in the cluster are using the same authentication scheme, and that they all have the same shares configured.

Keep in mind that StorNext NASsynchronizes authentication and share configuration information between all nodes in a NAS cluster. Changes made to a single node will not be synchronized between nodes unless you use the auth config or share commands. After you have created a NAS cluster, you can only execute the auth config or share commands from the master node.

With NAS failover configured, if you connect to the NAS cluster through the NAS VIP, you will ensure that you are always connected to the master node.

 

  1. Enter:

    > nascluster join <snfs_root_where_cluster_info_stored> [ip_addr]

    The parameters are:

    <snfs_root_where_cluster_info_stored>

    The root directory for the StorNext file system that contains the NAS cluster configuration information.

    If there are multiple StorNext file systems configured for your environment, all nodes within the NAS cluster must use the same StorNext file system.

    [ip_addr]

    The IP address of the node being joined to the NAS cluster.

    • If you are entering this command from the node that is joining the cluster, you do not need to specify an IP address.
    • If you are entering this command from the master node, you need to specify the IP address of the node to join to the cluster.

    Make sure that the IP address assigned to the NAS cluster nodes are in the same network and subnet used by the NAS clients.

    > nascluster join /stornext/snfs1

    NAS Cluster join starting ...

    NAS Cluster ID mycluster found ...

    Preparing to join NAS cluster as node 10.65.188.89

    Applying NAS cluster join settings ...

    Updating system NAS cluster configuration ...

    NFS: services (v4=yes, v3=yes, HA=no)

    Check for master takeover ...

    Publish master configuration ...

    Cluster verification for 10.65.188.89 in-progress ...

    Node state: pnn:0 10.65.188.89 UNHEALTHY (THIS NODE), waiting ...

    Node state: pnn:0 10.65.188.89 OK (THIS NODE)

    Cluster verification of 10.65.188.89 successful ...

    Waiting for NAS Master VIP 10.65.166.184 to be reachable ...

    Join to host.example.com as 1234567-MVIP starting ...

    Using AD Computers OU "Computers" ...

    Verify now joined to host.example.com as 1234567-MVIP ...

    Clearing ID map cache ...

    Restart SMB services to join with host.example.com ...

    Successfully joined NAS cluster

    > nascluster join /stornext/snfs1 10.65.190.230

    NAS Cluster join starting ...

    NAS Cluster ID mycluster found ...

    Proxy join to 10.65.190.230 ...

    [10.65.190.230]: NAS Cluster join to cluster ID mycluster starting ...

    [10.65.190.230]: Preparing for NAS cluster join as ID 10.65.190.230

    [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

    [10.65.190.230]: Synchronization of local configuration with master 10.65.176.238 starting...

    [10.65.190.230]: Applying ads auth config sync settings ...

    [10.65.190.230]: Applying ads configuration settings ...

    [10.65.190.230]: Applying NAS cluster join settings ...

    [10.65.190.230]: Updating system NAS cluster configuration ...

    [10.65.190.230]: NFS: services (v4=yes, v3=yes, HA=no)

    [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

    [10.65.190.230]: Cluster verification for 10.65.190.230 in-progress ...

    [10.65.190.230]: Node state: pnn:1 10.65.190.230 UNHEALTHY (THIS NODE), waiting ...

    [10.65.190.230]: Node state: pnn:1 10.65.190.230 OK (THIS NODE)

    [10.65.190.230]: Cluster verification of 10.65.190.230 successful ...

    [10.65.190.230]: Verify now joined to host.example.com as 1234567-MVIP ...

    [10.65.190.230]: Clearing ID map cache ...

    [10.65.190.230]: Restart SMB services to join with host.example.com...

    Successfully joined NAS cluster

  1. Log in to the Appliance Controller CLI.
  2. Create the NAS Cluster:

    Enabling a node makes the node a member of the NAS cluster. However, the node cannot actively take client connections until you join it to the NAS cluster.

    To enable a NAS cluster, you will need to do the following:

    • Enable a master node for the NAS cluster.

      When you enable a master node, the Appliance Controller issues the cluster a unique ID that is set to a uuid. This cluster ID is used to prevent non-master nodes from joining the wrong NAS cluster if the command is being issued from another non-master node. In such a case, the non-master node needs a way to identify the NAS cluster to join.

    • Enable additional nodes for the NAS cluster

     

    Enable NAS Cluster Nodes:

    1. Enter the following to enable the master node:

      nascluster enable master <ip_addr> <snfs_root_where_cluster_info_stored>

      The parameters are:

      <ip_addr>

      The IP address to use for the master node.

      Make sure this IP address is in the same network and subnet that the non-master nodes and clients use.

      <snfs_root_where_cluster_info_stored>

      The root directory for the StorNext file system that contains the NAS cluster configuration information.

      Example:

      > nascluster enable master 10.65.188.89 /stornext/snfs1

      Updating NAS cluster configuration ...

      Verifying NAS cluster configuration for 10.65.188.89 ...

      NAS cluster enable node 10.65.188.89 starting...

      Updating system NAS cluster configuration ...

      NFS: services (v4=yes, v3=yes, HA=no)

      Master node successfully enabled for NAS cluster using 10.65.188.89

    2. At the prompt, enter the following for each additional node to enable for the cluster:

      nascluster enable node <ip_addr>

      The parameter is:

      <ip_addr>

      The IP address to use for the non-master node.

      Make sure this IP address is in the same network and subnet that the master node and clients use.

      Example:

      > nascluster enable node 10.65.188.91

      Updating NAS cluster configuration ...

      Verifying NAS cluster configuration for 10.65.188.91 ...

      Node 10.65.188.91 successfully enabled for NAS cluster

    To configure and manage NAS clusters, you must assign a single virtual IP ()VIP) address to the master node. This NAS cluster VIP is always assigned to the active master node. So when you need to upgrade your NAS cluster or manage NAS settings for the cluster, you can do so by accessing the master node directly, which in turn pushes all updates to the additional cluster nodes.

    NAS Clusters that do not use Scale-Out NAS Clustering or DNS Load Distribution

    For SMB or NFSv4 environments with failover enabled, the NAS cluster VIP also facilitates NAS failover. In the event that the active master node becomes unavailable, the NAS VIP transfers NAS management services to the next available node in the NAS cluster, rendering that node the new active master node.

    NAS clusters that use Scale-Out NAS Clustering and DNS Load Distribution

    For scale-out NAS clusters, a VIP pool and DNS load distribution facilitate NAS failover.

    Before assigning a NAS VIP, review the following information.

    Important

    The VIP set for the NAS cluster is NOT the same as the VIP used for the StorNext network. Make sure the NAS VIP host name used for the NAS cluster is NOT the same host name used for the HA pair.

    • For the Xcellis WFD, you need to configure both nodes of the NAS cluster and the NAS VIP on the 10 GbE network port, which is typically the LAN client network. If you use another network port, the NAS cluster will fail to configure correctly.
    • The NAS VIP must be set within the same network and subnet used by the NAS cluster and NAS clients — this network is typically the LAN client network.
    • Enable global file locking for the StorNext file system. See the snfs_config MAN page, or search for "snfs_config − StorNext File System Configuration File" in the the StorNext MAN Pages Reference Guide for more information about the fileLocks parameter. Enable global file locking before setting the NAS VIP for the NAS cluster.
    • If you configure AD or OpenLDAP to authenticate users accessing your NAS cluster, you must add your cluster host name to the same DNS as your AD or OpenLDAP server. Otherwise, users authenticated through AD or OpenLDAP cannot access the NAS shares through the NAS cluster.

     

    1. Enter:

      nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]

      The parameters are:

      <master_virtual_ipaddr>

      The NAS VIP through which clients can access the NAS cluster. Keep in mind that the cluster's NAS VIP is assigned to the master node.

      Make sure this IP address is in the same network and subnet that the NAS cluster nodes and clients use.

      [<virtual_ipaddr> ...]

      (Optional) If you are creating a VIP pool, you can enter the VIPs to assign to each node within the cluster.

      You can also use the nascluster add virtual ipaddr <virtual_ipaddr> [<virtual_ipaddr> ...] command to create a VIP pool.

      Additional Information

      If you need to change a NAS VIP, re-issue the nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]command from the master node.

    When a node is joined to a NAS cluster, it is ready to take client connections, or to become the master node in the event of a NAS failover. To join nodes to a NAS cluster, you must issue the following command for each node that will be part of the cluster. All join commands can be issued from the master node.

    Cluster Synchronization

    When joining a node to a NAS cluster, the node is synchronized with the master node. This synchronization ensures that all nodes in the cluster are using the same authentication scheme, and that they all have the same shares configured.

    Keep in mind that StorNext NASsynchronizes authentication and share configuration information between all nodes in a NAS cluster. Changes made to a single node will not be synchronized between nodes unless you use the auth config or share commands. After you have created a NAS cluster, you can only execute the auth config or share commands from the master node.

    With NAS failover configured, if you connect to the NAS cluster through the NAS VIP, you will ensure that you are always connected to the master node.

     

    1. Enter:

      > nascluster join <snfs_root_where_cluster_info_stored> [ip_addr]

      The parameters are:

      <snfs_root_where_cluster_info_stored>

      The root directory for the StorNext file system that contains the NAS cluster configuration information.

      If there are multiple StorNext file systems configured for your environment, all nodes within the NAS cluster must use the same StorNext file system.

      [ip_addr]

      The IP address of the node being joined to the NAS cluster.

      • If you are entering this command from the node that is joining the cluster, you do not need to specify an IP address.
      • If you are entering this command from the master node, you need to specify the IP address of the node to join to the cluster.

      Make sure that the IP address assigned to the NAS cluster nodes are in the same network and subnet used by the NAS clients.

      > nascluster join /stornext/snfs1

      NAS Cluster join starting ...

      NAS Cluster ID mycluster found ...

      Preparing to join NAS cluster as node 10.65.188.89

      Applying NAS cluster join settings ...

      Updating system NAS cluster configuration ...

      NFS: services (v4=yes, v3=yes, HA=no)

      Check for master takeover ...

      Publish master configuration ...

      Cluster verification for 10.65.188.89 in-progress ...

      Node state: pnn:0 10.65.188.89 UNHEALTHY (THIS NODE), waiting ...

      Node state: pnn:0 10.65.188.89 OK (THIS NODE)

      Cluster verification of 10.65.188.89 successful ...

      Waiting for NAS Master VIP 10.65.166.184 to be reachable ...

      Join to host.example.com as 1234567-MVIP starting ...

      Using AD Computers OU "Computers" ...

      Verify now joined to host.example.com as 1234567-MVIP ...

      Clearing ID map cache ...

      Restart SMB services to join with host.example.com ...

      Successfully joined NAS cluster

      > nascluster join /stornext/snfs1 10.65.190.230

      NAS Cluster join starting ...

      NAS Cluster ID mycluster found ...

      Proxy join to 10.65.190.230 ...

      [10.65.190.230]: NAS Cluster join to cluster ID mycluster starting ...

      [10.65.190.230]: Preparing for NAS cluster join as ID 10.65.190.230

      [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

      [10.65.190.230]: Synchronization of local configuration with master 10.65.176.238 starting...

      [10.65.190.230]: Applying ads auth config sync settings ...

      [10.65.190.230]: Applying ads configuration settings ...

      [10.65.190.230]: Applying NAS cluster join settings ...

      [10.65.190.230]: Updating system NAS cluster configuration ...

      [10.65.190.230]: NFS: services (v4=yes, v3=yes, HA=no)

      [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

      [10.65.190.230]: Cluster verification for 10.65.190.230 in-progress ...

      [10.65.190.230]: Node state: pnn:1 10.65.190.230 UNHEALTHY (THIS NODE), waiting ...

      [10.65.190.230]: Node state: pnn:1 10.65.190.230 OK (THIS NODE)

      [10.65.190.230]: Cluster verification of 10.65.190.230 successful ...

      [10.65.190.230]: Verify now joined to host.example.com as 1234567-MVIP ...

      [10.65.190.230]: Clearing ID map cache ...

      [10.65.190.230]: Restart SMB services to join with host.example.com...

      Successfully joined NAS cluster

  3. Enable NFSv4

    If you are using NAS failover for 2-node or scale-out NAS clusters, NFSv3 is automatically enabled. If you are using NFSv4 failover for a 2-node NAS cluster, you must manually enable NFSv4 services.

    Note: MAC OSX clients do not correctly support NFSv4 mounts (ACLs are disabled by default). For MAC clients, Quantum strongly recommends you DO NOT enable NFSv4, but leave Scale-out NFS as-is, leaving NFSv3 enabled.

    Note: For a list of all the Appliance Controller commands, see the Command Index.

    1. Log in to the Appliance Controller CLI.
    2. Enter:

      3. > nas nfsv4 enable

  4. Enable NFSv4 Failover

    Important

    In an environment that does not use NFSv4 failover for NAS clusters, NFSv3 services are automatically enabled. However, you can use the nascluster set nfs-ha (yes|no) [force] command to enable or disable any NFS failover service used in your environment.

    Note: For a list of all the Appliance Controller commands, see the Command Index.

    NAS clusters with NFSv4 failover is supported only on the following systems:

    • Dual-server node Xcellis Workflow Director
    • Dual-server node Xcellis Foundation (upgraded from single-server node configuration)
    • Artico (R630)

    These systems operate in an active/passive NAS failover configuration.

    When configuring this type of NAS cluster, keep the following in mind:

    • This configuration does not support scale-out NAS clustering. You can upgrade this configuration to a scale-out NAS cluster by disabling NFSv4.
    • By enabling NAS failover for NFSv4, you are also requiring that all clients connecting to the NAS cluster are running NFSv4. This version of NFS is required to support lock recovery after failover.

    For more information about NAS failover and supported NAS cluster configurations, see NAS Cluster Overview.

    Make sure to perform the following steps before configuring NFS shares in a cluster for NAS failover.

    1. Verify that all clients accessing the NFS shares are unmounted.
    2. Verify that all clients connecting to the cluster are running NFSv4.

      If your clients are running on a Windows OS, you will need to download a third-party NFSv4 client for Windows.

    3. In addition to running NFSv4, verify that all clients are running CentOS versions 6.4, 6.5, 6.6, or 6.7. Only these versions have been validated in supporting lock recovery.

      4. Important

      NFSv4 failover is supported with NFSv4 version 4.0. Lock failover does not work with NFSv4 version 4.1

    4. Verify that all nodes within the NAS cluster are running on an Xcellis Workflow Director (CentOS7). Only Xcellis Workflow Director supports the NFSv4.1 lock recovery feature.

    Important

    For dual-server node Xcellis Workflow Director, dual-server node Xcellis Foundation, and Artico (R630) systems, StorNext NAS should be configured on the LAN client network. In addition, the IP addresses reserved for both nodes of the NAS cluster and the NAS VIP must be configured on the LAN client network. If you use another network port, the NAS cluster will fail to configure correctly.

    1. Unmount all clients.

      After configuring the NAS cluster for NFSv4 failover, you must remount the clients through the NAS VIP.

    2. Log in to the Appliance Controller CLI.
    3. Enter:

      nascluster set nfs-ha (yes|no) [force]

      The parameter is:

      <yes|no>

      Enter yes to enable NFS failover on the NAS cluster.

      Enter no to disable NFS failover on the NAS cluster.

      [force]

      If this option is included, then the command is issued without confirmation

      If this option is not included, then you must confirm the command.

      Duplicate Request Cache

      When NAS failover occurs in a NAS cluster of NFS shares, the duplicate request cache is not preserved and the new server cannot recover it. The cache is stored in memory.

  1. Log in to the Appliance Controller CLI.
  2. Create the NAS Cluster:

    Enabling a node makes the node a member of the NAS cluster. However, the node cannot actively take client connections until you join it to the NAS cluster.

    To enable a NAS cluster, you will need to do the following:

    • Enable a master node for the NAS cluster.

      When you enable a master node, the Appliance Controller issues the cluster a unique ID that is set to a uuid. This cluster ID is used to prevent non-master nodes from joining the wrong NAS cluster if the command is being issued from another non-master node. In such a case, the non-master node needs a way to identify the NAS cluster to join.

    • Enable additional nodes for the NAS cluster

     

    Enable NAS Cluster Nodes:

    1. Enter the following to enable the master node:

      nascluster enable master <ip_addr> <snfs_root_where_cluster_info_stored>

      The parameters are:

      <ip_addr>

      The IP address to use for the master node.

      Make sure this IP address is in the same network and subnet that the non-master nodes and clients use.

      <snfs_root_where_cluster_info_stored>

      The root directory for the StorNext file system that contains the NAS cluster configuration information.

      Example:

      > nascluster enable master 10.65.188.89 /stornext/snfs1

      Updating NAS cluster configuration ...

      Verifying NAS cluster configuration for 10.65.188.89 ...

      NAS cluster enable node 10.65.188.89 starting...

      Updating system NAS cluster configuration ...

      NFS: services (v4=yes, v3=yes, HA=no)

      Master node successfully enabled for NAS cluster using 10.65.188.89

    2. At the prompt, enter the following for each additional node to enable for the cluster:

      nascluster enable node <ip_addr>

      The parameter is:

      <ip_addr>

      The IP address to use for the non-master node.

      Make sure this IP address is in the same network and subnet that the master node and clients use.

      Example:

      > nascluster enable node 10.65.188.91

      Updating NAS cluster configuration ...

      Verifying NAS cluster configuration for 10.65.188.91 ...

      Node 10.65.188.91 successfully enabled for NAS cluster

    To configure and manage NAS clusters, you must assign a single virtual IP ()VIP) address to the master node. This NAS cluster VIP is always assigned to the active master node. So when you need to upgrade your NAS cluster or manage NAS settings for the cluster, you can do so by accessing the master node directly, which in turn pushes all updates to the additional cluster nodes.

    NAS Clusters that do not use Scale-Out NAS Clustering or DNS Load Distribution

    For SMB or NFSv4 environments with failover enabled, the NAS cluster VIP also facilitates NAS failover. In the event that the active master node becomes unavailable, the NAS VIP transfers NAS management services to the next available node in the NAS cluster, rendering that node the new active master node.

    NAS clusters that use Scale-Out NAS Clustering and DNS Load Distribution

    For scale-out NAS clusters, a VIP pool and DNS load distribution facilitate NAS failover.

    Before assigning a NAS VIP, review the following information.

    Important

    The VIP set for the NAS cluster is NOT the same as the VIP used for the StorNext network. Make sure the NAS VIP host name used for the NAS cluster is NOT the same host name used for the HA pair.

    • For the Xcellis WFD, you need to configure both nodes of the NAS cluster and the NAS VIP on the 10 GbE network port, which is typically the LAN client network. If you use another network port, the NAS cluster will fail to configure correctly.
    • The NAS VIP must be set within the same network and subnet used by the NAS cluster and NAS clients — this network is typically the LAN client network.
    • Enable global file locking for the StorNext file system. See the snfs_config MAN page, or search for "snfs_config − StorNext File System Configuration File" in the the StorNext MAN Pages Reference Guide for more information about the fileLocks parameter. Enable global file locking before setting the NAS VIP for the NAS cluster.
    • If you configure AD or OpenLDAP to authenticate users accessing your NAS cluster, you must add your cluster host name to the same DNS as your AD or OpenLDAP server. Otherwise, users authenticated through AD or OpenLDAP cannot access the NAS shares through the NAS cluster.

     

    1. Enter:

      nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]

      The parameters are:

      <master_virtual_ipaddr>

      The NAS VIP through which clients can access the NAS cluster. Keep in mind that the cluster's NAS VIP is assigned to the master node.

      Make sure this IP address is in the same network and subnet that the NAS cluster nodes and clients use.

      [<virtual_ipaddr> ...]

      (Optional) If you are creating a VIP pool, you can enter the VIPs to assign to each node within the cluster.

      You can also use the nascluster add virtual ipaddr <virtual_ipaddr> [<virtual_ipaddr> ...] command to create a VIP pool.

      Additional Information

      If you need to change a NAS VIP, re-issue the nascluster set virtual ipaddr <master_virtual_ipaddr> [<virtual_ipaddr> ...]command from the master node.

    When a node is joined to a NAS cluster, it is ready to take client connections, or to become the master node in the event of a NAS failover. To join nodes to a NAS cluster, you must issue the following command for each node that will be part of the cluster. All join commands can be issued from the master node.

    Cluster Synchronization

    When joining a node to a NAS cluster, the node is synchronized with the master node. This synchronization ensures that all nodes in the cluster are using the same authentication scheme, and that they all have the same shares configured.

    Keep in mind that StorNext NASsynchronizes authentication and share configuration information between all nodes in a NAS cluster. Changes made to a single node will not be synchronized between nodes unless you use the auth config or share commands. After you have created a NAS cluster, you can only execute the auth config or share commands from the master node.

    With NAS failover configured, if you connect to the NAS cluster through the NAS VIP, you will ensure that you are always connected to the master node.

     

    1. Enter:

      > nascluster join <snfs_root_where_cluster_info_stored> [ip_addr]

      The parameters are:

      <snfs_root_where_cluster_info_stored>

      The root directory for the StorNext file system that contains the NAS cluster configuration information.

      If there are multiple StorNext file systems configured for your environment, all nodes within the NAS cluster must use the same StorNext file system.

      [ip_addr]

      The IP address of the node being joined to the NAS cluster.

      • If you are entering this command from the node that is joining the cluster, you do not need to specify an IP address.
      • If you are entering this command from the master node, you need to specify the IP address of the node to join to the cluster.

      Make sure that the IP address assigned to the NAS cluster nodes are in the same network and subnet used by the NAS clients.

      > nascluster join /stornext/snfs1

      NAS Cluster join starting ...

      NAS Cluster ID mycluster found ...

      Preparing to join NAS cluster as node 10.65.188.89

      Applying NAS cluster join settings ...

      Updating system NAS cluster configuration ...

      NFS: services (v4=yes, v3=yes, HA=no)

      Check for master takeover ...

      Publish master configuration ...

      Cluster verification for 10.65.188.89 in-progress ...

      Node state: pnn:0 10.65.188.89 UNHEALTHY (THIS NODE), waiting ...

      Node state: pnn:0 10.65.188.89 OK (THIS NODE)

      Cluster verification of 10.65.188.89 successful ...

      Waiting for NAS Master VIP 10.65.166.184 to be reachable ...

      Join to host.example.com as 1234567-MVIP starting ...

      Using AD Computers OU "Computers" ...

      Verify now joined to host.example.com as 1234567-MVIP ...

      Clearing ID map cache ...

      Restart SMB services to join with host.example.com ...

      Successfully joined NAS cluster

      > nascluster join /stornext/snfs1 10.65.190.230

      NAS Cluster join starting ...

      NAS Cluster ID mycluster found ...

      Proxy join to 10.65.190.230 ...

      [10.65.190.230]: NAS Cluster join to cluster ID mycluster starting ...

      [10.65.190.230]: Preparing for NAS cluster join as ID 10.65.190.230

      [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

      [10.65.190.230]: Synchronization of local configuration with master 10.65.176.238 starting...

      [10.65.190.230]: Applying ads auth config sync settings ...

      [10.65.190.230]: Applying ads configuration settings ...

      [10.65.190.230]: Applying NAS cluster join settings ...

      [10.65.190.230]: Updating system NAS cluster configuration ...

      [10.65.190.230]: NFS: services (v4=yes, v3=yes, HA=no)

      [10.65.190.230]: Verifying local configuration with master 10.65.176.238 ...

      [10.65.190.230]: Cluster verification for 10.65.190.230 in-progress ...

      [10.65.190.230]: Node state: pnn:1 10.65.190.230 UNHEALTHY (THIS NODE), waiting ...

      [10.65.190.230]: Node state: pnn:1 10.65.190.230 OK (THIS NODE)

      [10.65.190.230]: Cluster verification of 10.65.190.230 successful ...

      [10.65.190.230]: Verify now joined to host.example.com as 1234567-MVIP ...

      [10.65.190.230]: Clearing ID map cache ...

      [10.65.190.230]: Restart SMB services to join with host.example.com...

      Successfully joined NAS cluster