Virtual Storage Console
The ability to quickly back up tens of hundreds of virtual machines without affecting production operations can accelerate the adoption of VMware within an organization.
The Virtual Storage Console (VSC) feature was formerly provided in a separate interface and was called SnapManager for Virtual Infrastructure (SMVI). It builds on the N series SnapManager portfolio by providing array-based backups. These consume only block-level changes to each VM and can provide multiple recovery points throughout the day. The backups are an integrated component within the storage array. Therefore, VSC provides recovery times that are faster than times provided by any other means.
14.1 Introduction to the Virtual Storage Console
The Virtual Storage Console (VSC) software is a single vCenter Server plug-in. It provides end-to-end virtual machine lifecycle management for VMware environments running N series storage. The plug-in provides these features:
•Storage configuration and monitoring, using the Monitoring and Host Configuration capability (previously called the Virtual Storage Console capability)
•Datastore provisioning and virtual machine cloning, using the Provisioning and Cloning capability
•Backup and recovery of virtual machines and datastores, using the Backup and Recovery capability
As a vCenter Server plug-in, shown in Figure 14-1, the VSC is available to all vSphere Clients that connect to the vCenter Server. This availability is different from a client-side plug-in that must be installed on every vSphere Client. You can install the VSC software on a Windows server in your data center, but you must not install it on a client computer.
Figure 14-1 Virtual Storage Console 2
Virtual Storage Console (VSC) integrates VSC storage discovery, health monitoring, capacity management, and best practice-based storage setting. It offers additional management capabilities with two capability options in a single vSphere™ client plug-in. Thus it enables centralized, end-to-end management of virtual server and desktop environments running on N series storage. VSC is composed of three main components:
•Virtual Storage Console Capability (base product): Provides a storage view of the VMware® environment with a VM administrator perspective. It automatically optimizes the customer’s host and storage configurations, including HBA timeouts, NFS tunables, and multipath configurations. Using the Virtual Storage Console, a VM administrator can quickly and easily view controller status and capacity information. Also, the administrator can accurately report back utilization information in order to make more informed decisions about VM object placement.
•Provisioning and Cloning Capability: Provides end-to-end datastore management (provisioning, resizing, and deletion). Also offers rapid, space-efficient VM server and desktop cloning, patching, and updating by using FlexClone® technology.
•Backup and Recovery capability (formerly SnapManager® for Virtual Infrastructure): Automates data protection processes by enabling VMware administrators to centrally manage backup and recovery of datastores and VMs. This can be done without impacting guest performance. The administrator can also rapidly recover from these backup copies at any level of granularity: datastore, VM, VMDK, or guest file.
VSC is designed to simplify storage management operations, improve efficiencies, enhance availability, and reduce storage costs in both SAN- and NAS-based VMware infrastructures.
It provides VMware administrators with a window into the storage domain. It also provides the tools to effectively and efficiently manage the lifecycle of virtual server and desktop environments running on N series storage.
It provides VMware administrators with a window into the storage domain. It also provides the tools to effectively and efficiently manage the lifecycle of virtual server and desktop environments running on N series storage.
14.1.1 License requirements
Table 14-1 summarizes the N series license requirements to perform different VSC functions.
Table 14-1 VSC license requirements
|
Task
|
License
|
|
Provision datastores
|
NFS, FCP, iSCSI
|
|
Restore datastores
|
SnapRestore
|
|
Use vFilers in Provisioning and Cloning operations
|
MultiStore
|
|
Clone virtual machines
|
FlexClone (NFS only)
|
|
Configure deduplication settings
|
A-SIS
|
|
Distribute templates to remote vCenters
|
SnapMirror
|
14.1.2 Architecture overview
Figure 14-2 illustrates the architecture for VSC. It also shows the components that work together to provide a comprehensive and powerful backup and recovery solution for VMware vSphere environments.
Figure 14-2 Architecture overview
14.1.3 Monitoring and host configuration
The Monitoring and Host Configuration capability enables you to manage ESX and ESXi servers connected to N series storage systems. You can set host timeout, NAS, and multipathing values, view storage details, and collect diagnostic data. You can use this capability to do the following tasks:
•View the status of storage controllers from a SAN (FC, FCoE, and iSCSI) perspective
•View the status of storage controllers from a NAS (NFS) perspective
•View SAN and NAS datastore capacity utilization
•View the status of VMware vStorage APIs for Array Integration (VAAI) support in the storage controller
•View the status of ESX hosts, including ESX version and overall status
•Check at a glance whether the following settings are configured correctly, and if not, automatically set the correct values:
– Storage adapter timeouts
– Multipathing settings
– NFS settings
•Set credentials to access storage controllers
•Launch the FilerView GUI to create LUNs and manage storage controllers
•Collect diagnostic information from the ESX hosts, storage controllers, and Fibre Channel switches
•Access tools to set guest operating system timeouts and to identify and correct misaligned disk partitions
When you click the N series icon in the vCenter Server and click Monitoring and Host Configuration in the navigation pane, the Overview panel displays. It is similar to Figure 14-3.
Figure 14-3 VSC overview
Alternatively, you can find the VSC plug-in under Solutions and Applications (Figure 14-4).
Figure 14-4 VSC location
14.1.4 Provisioning and Cloning
The Provisioning and Cloning capability of Virtual Storage Console helps you to provision datastores and quickly create multiple clones of virtual machines in the VMware environment. Using FlexClone technology, the Provisioning and Cloning capability allows you to efficiently create, deploy, and manage the lifecycle of virtual machines. These tasks can be done from an easy-to-use interface integrated into the VMware environment. It is ideal for virtual server, desktop, and cloud environments. You can use this capability for the following purposes:
•Clone individual virtual machines and place in new or existing datastores
•Create, resize, or delete datastores
•Apply guest customization specifications and power up new virtual machines
•Run deduplication operations
•Monitor storage savings
•Redeploy virtual machines from a baseline image
•Replicate NFS datastores across sites
•Import virtual machines into virtual desktop infrastructure connection brokers and management tools
Managing datastores and cloning virtual machines
To manage datastores and clone virtual machines, right-click an object in the Inventory panel of the vSphere Client and select N series → Provisioning and Cloning:
•Right-click a powered-down virtual machine or template to create clones.
•Right-click a datacenter, cluster, or host to provision datastores.
Figure 14-5 Accessing Provisioning and Cloning
Managing controllers, replicating datastores, and redeploying clones
Click the Inventory button in the navigation bar, and then select Solutions and Applications → N series. Use the following options:
•Select Storage controllers to add, remove, or modify properties of storage controllers.
•Select Connection brokers to add and remove connection broker definitions.
•Select DS Remote Replication to clone NFS datastore templates to multiple target sites.
•Select Redeploy to redeploy virtual machines.
14.2 Installing the Virtual Storage Console 2.0
The VSC provides full support for hosts running ESX/ESXi 4.0 and later. It provides limited reporting functionality with hosts running ESX/ESXi 3.5 and later.
14.2.1 Basic installation
Before downloading and installing the VSC, make sure that your deployment has the required components:
•You need a vCenter Server version 4.0 or later. The VSC can be installed on the vCenter Server or on another server or VM (see Figure 14-6).
•If installing on another server or VM, this system must run 32-bit or 64-bit Windows Server 2008, 2003 SP1 and later, or a 32-bit version of XP Professional SP2 and later.
•A storage array is required to run Data ONTAP 7.3.1.1 or later.
|
Attention: Before installing, verify supported storage adapters and firmware.
|
Figure 14-6 VSC possible deployments
|
Tip: To keep it simple, we suggest installing the VSC on the vCenter server.
|
Complete the following steps to install the VSC 2.0:
1. Download the installation program to the Windows server.
2. Run the installation wizard and select the features you would like to install as shown in Figure 14-7.
3. Follow the on-screen instructions.
During the installation process, a prompt displays to select the features of the VSC 2.0 to be enabled in the environment. The core VSC must be selected. The Provisioning and Cloning and Backup and Recovery features are the former RCU and the SMVI interfaces. Certain subfeatures might require licensing, as described previously.
4. Register the VSC as a plug-in, in the vCenter Server in the window that opens when the process is complete.
This final step requires a user with vCenter administrator credentials to complete the registration process.
Figure 14-7 Select VSC features
The installation process launches the vCenter registration process as shown in Figure 14-8.
Figure 14-8 vCenter registration process
5. Finally, register the VSC plug-in with a vCenter server (Figure 14-9).
Figure 14-9 VSC registration with vCenter server
Upon successful registration, the system confirms by issuing the following message on the web page: The registration process has completed successfully!
14.2.2 Upgrading the VSC
As of December 2011, an upgrade to VSC 2.1.1 is available that needs to be installed after installing and registering VSC2.0 (see Figure 14-10). Follow these steps:
1. Download the installer for VSC.
2. Double-click the installer icon, and click Run to start the installation wizard.
3. Click Yes on the confirmation prompt.
4. In the installation wizard, select the capabilities that you want to upgrade and click Next to start the installation. The installation might take several minutes.
5. Click Finish to complete the installation.
Figure 14-10 Upgrade to VSC 2.1.1
|
Support: VSC 2.1.1 supports upgrades from VSC 2.0 only. The VSC installer does not support upgrades from a version of VSC prior to 2.0 or from stand-alone versions of RCU or VSC (SMVI). If you have that software installed, you must uninstall it before you can install VSC 2.1.1. If the VSC installer finds one of those versions of VSC, RCU, or SMVI on the server, it prompts you to uninstall the software. Then the installer aborts.
|
The VSC installer checks the version numbers of each of the currently installed capabilities as shown in Figure 14-11. It lets you upgrade each capability that has an older version number than the one you are installing.
Figure 14-11 Select VSC upgrades
6. A web page displays when the installation is complete. You must register VSC with the vCenter Server. You must provide the vCenter Server host name or IP address and the administrative credentials.
|
Attention: After you finish, you must close the vSphere Client and restart it to display newly installed capabilities.
|
14.3 Adding storage controllers to the VSC
Adding the storage controllers that host the virtual infrastructure to the VSC is fairly simple:
1. Connect to vCenter by using the vSphere client.
2. Double-click the N series icon on the Home panel.
3. Select the Virtual Storage Console tab on the left.
After these steps are completed, the VSC launches and automatically identifies all storage controllers powered by Data ONTAP with the storage connected to the ESX/ESXi hosts in the environment. As an alternative to running discovery for the entire environment, you can select an ESX/ESXi host or cluster in the vSphere client and then select the NetApp tab in the left panel. The VSC then begins discovery of all storage controllers with storage connected to the host or cluster that was selected.
The Controller Credentials wizard starts, displayed in Figure 14-12, allowing you to enter the user or service account assigned for VSC management on the storage controller. This account can be the root account or one created specifically for the VSC core feature, as described previously.
Figure 14-12 Adding storage controller access in VSC
14.4 Optimal storage settings for ESX/ESXi hosts
The VSC enables the automated configuration of storage-related settings for all ESX/ESXi 4.x hosts connected to NetApp storage controllers. VMware administrators can right-click individual or multiple ESX/ESXi hosts and set the preferred values for these hosts. This functionality sets values for HBAs and CNAs, sets appropriate paths and path selection plug-ins, and provides appropriate settings for software-based I/O (NFS and iSCSI).
To perform the setting, go to the VSC pane, right-click the designated ESX server, and run the settings as shown in Figure 14-13.
Figure 14-13 Optimize ESX settings
After rebooting the ESX server, we can verify the improved settings. All status indicators are green (see Figure 14-14).
Figure 14-14 Optimized ESX adapter settings
14.5 SnapMirror integration
SnapMirror relationships cannot be configured through VSC. However, VSC can update an existing SnapMirror relationship on the volume underlying the datastore or virtual machine. Preferably, test the SnapMirror relationship from the storage system command line before updating through VSC. This method aids in identifying where any potential issues might occur. If the SnapMirror update is successful from the CLI, but fails from within VSC, the administrator has a better understanding of where to concentrate troubleshooting efforts.
Also, identify the destination storage within VSC in the same manner that the relationship is configured on the storage system. For example, if a SnapMirror relationship is configured on the storage system using IP addresses rather than a DNS name, identify the auxiliary storage to VSC by the IP address and vice versa.
Because its support is for SnapMirror volume only, map one volume per datastore.
During backup creation, SnapManager provides the option of updating an existing SnapMirror relationship. That way, every time a Snapshot is created, the data is transferred to a remote storage system. Whenever the backup of a virtual machine or datastore is initiated with the SnapMirror option, the update starts as soon as the backup completes, after of the current SnapMirror schedule.
For example, by configuring regular SnapMirror updates on a filter after the VSC schedule, you can cut down the time required to update the mirror, because it is done in the interim. However, keep in mind that the updates must be scheduled in such a way that they do not conflict with the SnapManager backup.
14.5.1 SnapMirror destinations
A single SnapMirror destination is supported per volume. If a SnapMirror update is selected as part of a backup on a volume with multiple destinations, the backup fails.
If multiple SnapMirror destinations are required, use a tiered approach when configuring the SnapMirror relationships. For example, if the data must be transferred to four destinations, configure one destination from the primary storage system supported to one destination. Then configure three additional destinations from the auxiliary storage through the storage system CLI.
14.5.2 SnapMirror and deduplication
Preferably, do not use deduplication with Sync SnapMirror. Although technically it works, the integration and scheduling of deduplication with Sync SnapMirror are complicated to implement in the type of rigorous real-world scenarios that demand synchronous replication.
When configuring volume SnapMirror and deduplication, consider the deduplication schedule and the volume SnapMirror schedule. Start volume SnapMirror transfers of a deduplicated volume after deduplication completes (that is, not during the deduplication process). This technique avoids sending undeduplicated data and additional temporary metadata files over the network. If the temporary metadata files in the source volume are locked in Snapshot copies, they also consume extra space in the source and destination volumes. Volume SnapMirror performance degradation can increase with deduplicated volumes.
The scenario described previously has a direct impact on backups configured within VSC when the SnapMirror update option was selected. Avoid scheduling a backup with the SnapMirror update option until a a confirmation of the volume deduplication completeness. Although a few hours must be scheduled to ensure avoiding this issue, the actual scheduling configuration is data and customer dependent.
14.6 VSC in an N series MetroCluster environment
N series MetroCluster configurations consist of a pair of active-active storage controllers. They are configured with mirrored aggregates and extended distance capabilities to create a high-availability solution. This type of configuration has the following benefits:
•Higher availability with geographic protection
•Minimal risk of lost data, easier management and recovery, and reduced system downtime
•Quicker recovery when a disaster occurs
•Minimal disruption to users and client applications
A MetroCluster (either Stretch or Fabric) behaves in most ways similar to an active-active configuration. All of the protection provided by core N series technology (RAID-DP, Snapshot copies, automatic controller failover) also exists in a MetroCluster configuration. However, MetroCluster adds complete synchronous mirroring along with the ability to perform a complete site failover from a storage perspective with a single command.
The following N series MetroCluster types exist and work seamlessly with the complete VMware vSphere and ESX server portfolio:
•Stretch MetroCluster (sometimes called a nonswitched cluster) is an active-active configuration that can extend up to 500 m depending on speed and cable type. It includes synchronous mirroring (SyncMirror) and the ability to do a site failover with a single command.
•Fabric MetroCluster (also called a switched cluster) uses four Fibre Channel switches in a dual-fabric configuration. It uses a separate cluster interconnect card to achieve an even greater distance (up to 100 km depending on speed and cable type) between primary and secondary locations.
The integration of the MetroCluster and VMware vSphere is seamless and provides storage and application redundancy. In addition to connecting to the vSphere environment using FCP, iSCSI, or NFS, this solution can serve other network clients with CIFS, HTTP, and FTP at the same time. The solution shown in Figure 14-15 provides a redundant VMware server, redundant N series heads, and redundant storage.
Figure 14-15 MetroCluster and VMware vSphere integrated solution
For more information about N series MetroCluster, see the “MetroCluster” chapter in the Redbooks publication, IBM System Storage N series Software Guide, SG24-7129.
14.7 Backup and recovery
This section provides examples of backing up a single virtual machine or the entire DataCenter. The Backup and Recovery capability of the Virtual Storage Console provides rapid backup and recovery of multi-host configurations running on N series storage systems.
You can use this capability to do the following tasks:
•Perform on-demand backups of individual virtual machines, datastores, or a datacenter
•Schedule automated backups of individual virtual machines, datastores, or a datacenter
•Support virtual machines and datastores that are located on either NFS directories or VMFS file systems
•Mount a backup to verify its content prior to restoration
•Restore datastores or virtual machines to the original location
•Restore virtual machine disks (VMDKs) to the original or an alternate location
•Restore one or more files to a guest VMDK without having to restore the entire virtual machine or VMDK using single file restore feature
To configure your storage systems, click the N series icon in the vCenter Server and click Setup under Backup and Recovery in the navigation pane. The Setup panel displays. Click Add on the left side and register your N series system as shown in Figure 14-16.
|
Important: You must register your N series system three times; first, for the VSC; second, for backup and recovery; and third, for Provisioning and Cloning.
|
Figure 14-16 N series registration for backup and restore
14.7.1 Data layout
Layout is indicated by N series best practices for vSphere environments. Move any transient and temporary data, such as the guest operating system swap file, temp files, and page files, to a separate virtual disk on another datastore. The reason is that snapshots of this data type can consume a large amount of storage in a short time because of the high rate of change.
When a backup is created for a virtual machine with VSC, VSC is aware of all VMDKs associated with the virtual machine. VSC initiates a Snapshot copy on all datastores upon which the VMDKs reside. For example, a virtual machine running Windows as the guest operating system has its C drive on datastore ds1, data on datastore ds2, and transient data on datastore td1. In this case, VSC creates a Snapshot copy against all three datastores at underlying volume level. It defeats the purpose of separating temporary and transient data.
Considerations for transient and temporary data
To exclude the datastore that contains the transient and temporary data from the VSC backup, configure the VMDKs residing in the datastore as “Independent Persistent” disks within the VMware Virtual Center (vCenter). After the transient and temporary data VMDKs are configured, they are excluded from both the VMware Virtual Center snapshot and the N series Snapshot copy initiated by VSC.
You must also create a datastore dedicated to transient and temporary data for all virtual machines with no other data types or virtual disks residing on it. This datastore avoids having a Snapshot copy taken against the underlying volume as part of the backup of another virtual machine. Do not deduplicate the data on this datastore.
SnapManager 2.0 for Virtual Infrastructure can include independent disks and exclude datastores from backup.
Including independent disks and excluding datastores
You can avoid having a Snapshot copy performed on the underlying volume as part of the backup of another virtual machine. In this case, preferably, create a datastore that is dedicated to transient and temporary data for all virtual machines. Exclude datastores that contain transient and temporary data from the backup. By excluding those datastores, snapshot space is not wasted on transient data with a high rate of change. In VSC 2.0, when selected entities in the backup span multiple datastores, one or more of the spanning datastores might be excluded from the backup.
After configuration, the transient and temporary data .vmdk are excluded from both the VMware vCenter Snapshot and the N series Snapshot copy initiated by VSC. In VSC 1.0, datastores with only independent disks were excluded from the backup. In VSC 2.0, an option is available to include them in the backup. Datastores with a mix of independent disks and normal disks or configuration files for a VM are included in the backup irrespective of this option.
If you have a normal disk and an independent disk for backup on the same datastore, it is always included for backup irrespective of the “include datastore with independent disk” option. Designate a separate datastore exclusively for swap data.
|
Restore from backup: If you exclude non-independent disks from the backup of a VM, that VM cannot be completely restored. You can perform only virtual disk restore and single file restore from such a backup.
|
14.7.2 Backup and recovery requirements
Your datastore and virtual machines must meet the following requirements before you can use the Backup and Recovery capability:
•In NFS environments, a FlexClone license is required to mount a datastore, restore guest files, and restore a VMDK to an alternate location.
•Snapshot protection is enabled in the volumes where those datastore and virtual machine images reside.
•SnapRestore is licensed for the storage systems where those datastore and virtual machine images reside.
14.7.3 Single wizard for creating backup jobs
With the wizard, you can create manual and scheduled backup jobs. In the right pane, you click Backup, name your new backup job, and select the per-backup job options:
•Initiate SnapMirror update.
•Perform VMware consistency snapshot.
•Include datastores with independent disks.
Virtual Machine backup
To back up individual VMs, follow these steps:
1. Right-click the VM Backup and drill down until you reach the selection to run or schedule a backup, as shown in Figure 14-17.
Figure 14-17 Adding a backup
2. Go to the Welcome panel, and then click Next.
3. Set a Name and Description, specify possible SnapMirror update, or include independent disks (see Figure 14-18), then click Next.
Figure 14-18 Backup options
4. Following, you can select scripts to be included in the backup job (see Figure 14-19).
Figure 14-19 Backup scripts
Figure 14-20 Backup schedule
Figure 14-21 Backup job credentials
7. Revise the information entered and click Finish on the Schedule a Backup Wizard and click Next.
8. Select to run your new backup job immediately if you want, as shown in Figure 14-22.
Figure 14-22 Revise scheduled backup job
Datacenter backup
Alternatively, you can also select to back up the whole datacenter as shown in Figure 14-23. Some options are then added to the previously described process.
Figure 14-23 Datacenter backup
The backup wizard adds the option to select the whole datacenter of backup individual datastores as displayed in Figure 14-24.
Figure 14-24 Datacenter backup options
Datastore backup
Alternatively, you can also select to back up an individual datastore as shown in Figure 14-24. Some options are then added to the previously described process.
Figure 14-25 Datastore backup
The backup wizard adds the option to select the whole datastore of backup individual datastores as displayed in Figure 14-26.
Figure 14-26 Datastore backup options
14.7.4 Granular restore options
The following granular restore options are available:
•Restore datastores or virtual machines to the original location.
•Restore virtual machine disks (VMDKs) to the original or an alternate location.
•Restore one or more files to a guest VMDK without having to restore the entire virtual machine or VMDK using single file restore feature.
You can access these options by the tabs as shown in Figure 14-27. Right-click the object that you want to restore.
Figure 14-27 Restore options
You can also select whether you want to restore the entire virtual machine or individual virtual disks, as in Figure 14-28. Furthermore, you can select the original or a new location.
Figure 14-28 VSC enhanced restore options
14.7.5 Other features
In addition, VSC offers these features:
•Consistent backup naming
•Serialization of VMware vSphere snapshots
•AutoSupport (ASUP) logging
•vFiler unit support for multiple IP addresses
•Advanced Find option to find specific backups
14.8 Provisioning and Cloning
This section provide information and examples of the Provisioning and Cloning functions integrated in VSC.
14.8.1 Features and functions
The provisioning features require al least Data ONTAP 7.3.3 to accomplish the following tasks:
•Creation, resizing, and deletion of VMFS/NFS datastores
•Ability to provision, clone, and resize volumes on secure vFiler units
•Adding storage system using a domain account
•Automation of pathing for both LUNs and NFS datastores
•Running deduplication operations
•Monitoring storage savings and performance
•Protection against failover of NFS mounts to non-redundant VMkernel ports by limiting multiple TCP sessions to iSCSI only
The cloning features allow you to perform the following tasks:
•Creation of multiple virtual machine clones in new or existing datastores (using FlexClone technology)
•Application of guest customization specifications and powering up of new virtual machines
•Redeployment of virtual machines from a baseline image
•Importing virtual machines into virtual desktop infrastructure connection brokers and management tools
•Clone misalignment alert and prevention:
– VM misalignment detection and user notification
– Support for VMFS- and NFS-based VMs
•Ability to import virtual machine settings from a file:
– Non-contiguous virtual machine names
– Guest customization specifications
– Computer name as virtual machine name
– Power-on settings
•Support for these products:
– VMware View 4.0, 4.5, 4.6 & 5.0
– Citrix XenDesktop 4.0 and 5.0
Further features are included:
•Space reclamation management
•Addition of new datastores to new ESX Servers within a cluster
•Service catalog-based provisioning API with enhanced SOAP API to support creation, deletion, and resizing of NFS/VMFS datastores by Storage Services in Provisioning Manager
•Space Reclamation Management
•Mounting of existing datastores when new ESX hosts are added to a cluster or datacenter with support for both NFS and VMFS datastores
•Capability for the user to mount any existing datastore to newly added ESX hosts:
– VDI One-click Golden Template distribution
– This feature allows the user to copy a datastore from a source vCenter to one or more target vCenters
•VMware Virtual Desktop Infrastructure (VDI) enhancements:
– XenDesktop/View import from API
– VDI One-click Golden Template distribution
– Saving of View credentials
– Soap API support for importing newly created clones into Citrix XenDesktop and VMware View
– Storing of View Server credentials
– Elimination of the need to add VMware View Server credentials each time by the cloning wizard
– Creation of multiple View Server pools
14.8.2 Provision datastores
The Provisioning and Cloning feature of the VSC 2.0 currently requires reauthentication of storage arrays by specifying the credentials necessary for communication.
|
Important: You must register your N series system three times; first, for the VSC, second, for backup and recovery; and third, for Provisioning and Cloning.
|
To do this using the vSphere client, complete the following steps (see Figure 14-29):
1. Connect to vCenter.
2. Select the N series icon on the Home panel.
3. Select the Provisioning and Cloning tab on the left side.
4. Click the Add button to begin the Controller Configuration wizard.
Figure 14-29 Provisioning and Cloning add controller
You can create new datastores at the datacenter, cluster, or host level. The new datastore displays on every host in the datacenter or the cluster.
This process launches the N series Datastore Provisioning wizard, which allows you to select the following features:
•Storage controller
•Type of datastore (VMFS or NFS)
•Datastore details, including storage protocol and block size (if deploying a VMFS datastore)
•Specifying whether the LUN should be thin-provisioned
The provisioning process connects the datastore to all nodes within the selected group. For iSCSI, FC, and FCoE datastores, the VSC handles storage access control as follows:
•Creating initiator groups
•Enabling ALUA
•Applying LUN masking
•Applying path selection policies
•Formatting the LUN with VMFS
For NFS datastores, the VSC handles storage access control by managing access rights in the exports file, and it balances the load across all available interfaces.
|
Tip: Remember, if you plan to enable data deduplication, then thin-provisioned LUNs are required to return storage to the free pool on the storage controller.
|
Follow these steps:
1. In the vSphere Client Inventory, right-click a datacenter, cluster, or host and select N series → Provisioning and Cloning → Provision datastore (see Figure 14-30).
Figure 14-30 Provision a datastore
2. Next specify the N series system to use (see Figure 14-31).
Figure 14-31 Select storage controller for provisioning
3. Following, select the protocol to use. Here we only have NFS available, as shown in Figure 14-32.
Figure 14-32 Specify datastore type
4. Now specify the new datastore details (see Figure 14-33).
Figure 14-33 New datastore details
5. Before applying your selection, verify the information as shown in Figure 14-34.
Figure 14-34 Review new datastore settings
The new datastore named newDatastore was created on the N series. It can now be mounted to the host you want. Figure 14-35 shows FilerView access and the NFS exports.
Figure 14-35 Verify NFS exports
14.8.3 Managing deduplication
Deduplication eliminates redundant objects on a selected datastore and only references the original object. Figure 14-36 shows how VSC is able to manage deduplication for each individual datastore.
Figure 14-36 Managing deduplication
Possible options to use N series advanced deduplication features are displayed in Figure 14-37. Click OK to apply your settings.
Figure 14-37 Manage deduplication features
14.8.4 Cloning virtual machines
The Provisioning and Cloning capability can theoretically create thousands of virtual machine clones and hundreds of datastores at one time. In practice, however, multiple executions of fewer requests are preferred. The exact size of these requests depends on the size of the vSphere deployment and the hardware configuration of the vSphere Client managing the ESX hosts.
Follow these steps:
1. In the vSphere Client Inventory, right-click a powered-down virtual machine (Figure 14-38) or template and select N series → Provisioning and Cloning → Create rapid clones.
Figure 14-38 Select VM for cloning
2. Next select the controller you want to use for cloning (see Figure 14-39).
Figure 14-39 Select controller for cloning
3. Following, select the destination N series system (see Figure 14-40).
Figure 14-40 Select clone target
4. Now specify the VM format for the clone as shown in Figure 14-41.
Figure 14-41 Clone VM format
5. In the following window, specify details for the new datastores as displayed in Figure 14-42.
Figure 14-42 Clone VM details
Figure 14-43 Summary for cloning
After successful completion of the cloning tasks, the new VMs are configured and ready for further use. Figure 14-44 shows the cloning results.
Figure 14-44 Clone results
14.9 SnapManager for Virtual Infrastructure commands
The SnapManager for Virtual Infrastructure (SMVI) command line interface is still part of the VSC. You can use the Virtual Storage Console command-line interface to perform specific Backup and Recovery capability tasks.
All VSC commands can be performed by using either the GUI or the CLI, with some exceptions. For example, only the creation of scheduled jobs and their associated retention policies and single file restore can be performed through the GUI.
Remember the following general information about the commands:
•SnapManager for Virtual Infrastructure commands are case-sensitive.
•There are no privilege levels; any user with a valid user name and password can run all commands.
You can launch the Virtual Storage Console CLI by using the desktop shortcut
or the Windows Start menu. Double-click the VSC CLI desktop icon or navigate to
Start → All Programs → IBM → Virtual Storage Console →IBM N series VSC CLI.
or the Windows Start menu. Double-click the VSC CLI desktop icon or navigate to
Start → All Programs → IBM → Virtual Storage Console →IBM N series VSC CLI.
14.10 Scripting
VSC provides users the ability to run pre, post, and failure backup phase scripts based on SMVI commands as stated in the previous section. These scripts are any executable process on the operating system in which the VSC is running. When defining the backup to run, the pre, post, and failure backup scripts can be chosen by using either the VSC GUI or CLI. The scripts must be saved in the <SMVI Installation>/server/scripts/ directory. Each chosen script runs as a pre, post, and failure backup script.
From the GUI, you can select multiple scripts by using the backup creation wizard or when editing an existing backup job as shown in Figure 14-19 on page 260. The UI lists all files found in the server/scripts/ directory. VSC runs the scripts before creating the VMware snapshots and after the cleanup of VMware snapshots.
When VSC starts each script, a progress message is logged indicating the start of the script. When the script completes, or is terminated by SAN volume controller because it was running too long, a progress message is logged. It indicates the completion of the script and states if the script was successful or failed. If a script is defined for a backup but is not found in the scripts directory, a message is logged stating that the script cannot be found.
The VSC maintains a global configuration value to indicate the amount of time that a script can execute. After a script runs for this length of time, the script is terminated by the VSC to prevent run-away processing by scripts. If VSC must terminate a script, it is implicitly recognized as a failed script and might force termination of the VSC backup in the pre-backup phase.
With the default settings, VSC waits for up to 30 minutes for each script to complete in each phase. This default setting can be configured by using the following entry in the server/etc/smvi.override file:
smvi.script.timeout.seconds=1800
VSC backup scripts receive input from the environment variables. This way, the input can be sent in a manner that avoids CLI line length limits. The set of variables varies based on the backup phase.
Sample scripts for VSC can be found in Appendix B, “Sample scripts for VSC” on page 281.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.