VCP6-DCV Section 3: Configure and Administer Advanced vSphere 6.x Storage

Section 3 of the VCP6-DCV blueprint

Objective 3.1: Manage vSphere Storage Virtualization

Identify storage adapters and devices

Adapters

SCSI adapter
iSCSI adapter
RAID adapter
Fibre Channel Adapter
Fibre Channel over Ethernet adapter
Ethernet adapter

Device drivers are part of the VMKernal and accessed directly by ESXi
Logical Unit Numbers (LUNs) are represented by a SCSI volume that is presented to the host.

Identify storage naming conventions

There are three different types of device identifiers used

SCSI INQUIRY Identifiers: Unique accross all hosts and are persistent. The host uses the SCSI INQUIRY command in order to use the age 83 information (Device Identification) to generate a unique identifier

naa.number
t10.number
eui.number

Path-based Identifier: When a device is queried and doe snot return device information, the host generates an mpx.path name. Path represents the path to that particular device. This is created for local devices during boot and is not unique or persistent.

mpx.vmhba1.C0.T0.L0

Legacy Identifier: ESXi also generates a legacy name as an alternative with the following format

vml.number: The Number are digits unique to the device and can be taken from a part of the device information.

Identify Hardware/dependant hardware/software iSCI initiator requirements

A hardware iSCSI adapter offloads the network and iSCSI processing form the host. There are two types of hardware iSCSI adapters; dependent hardware iSCSI adapter and independent hardware iSCSI adapter (Ensure these are listed on the HCL)
Dependent Hardware iSCSI Adapter

These types of adapters depend on VMware networking and the iSCSI management interfaces within VMware.
Dependant upon the host's network configuration for IP and MAC

Independent Hardware iSCSI Adapter

These types of adapters are independent from the host and VMware
Provides its own configuration management for IP and other network address.

The software iSCSI adapter is built into VMwares code, specifically the VMKernal. Using this type of adapter you can connect it iSCSI targets using a standard network adapter installed on the host. Since this is a software adapter, networking processing and ecapsulation are performed by the host, which does use host resources.

Discover new storage LUNs

Home -> Hosts and Clusters -> Rigth click host, datacenter or cluster -> Storage
Rescan Storage

Configure FC/iSCSI/FCoE LUNs as ESXi boot devices

General

You may need to do some configurations that are vendor specific, check the documentaiton
You need to setup access control from ESXi host to the boot LUN

From teh storage array, create a LUN to act as a boot LUN for each host that require boot from SAN
mask each LUN to their respective ESXi hosts. IE; create a LUN01 and mask it to host01, LUN02 to host02.
determine the WWPN for a front-end port on the SAN, vendor specific.

iSCSI

From storage array, create a LUN to act as a boot LUN for each host that require boot from SAN
Mask each LUN to their respective hosts (see above)
Determine iSCSI name an dIP addresses for the targets that are assigned for the particular ESXi host.
If using an independent hardware iSCSI adapter, you'll need to follow the vendor documentation for that adapter to conifgure it to boot from SAN. This configuration will include using the IPs you recorded earlier as send targets to discover the boot LUN
You can also use the software or depednent hardware iSCSI adapters. THis requires that the dependent adapter (or software adapter) supprt iBFT, or iSCSI Boot Firmware Table
When booting via iBFT the ESXi host will go through the following process:

BIOS finds the iSCSI firmware of the network adapter
The boot firmware will connect to the iSCSI target using its predefined parameters (The stuff set by vendor documentation)
Once the connection is made all the networking and boot parameters are stored in teh iBFT, which is then stored in memory.
The BIOS will then boot the boot device
from here the vmkernal takes over the boot operations and wil connect to the iSCSI target using the information from iBFT
ESXI will boot once the vmkernal establishes a connection to the target.

Here are the general steps to set this up on the ESXi host.

Configure your iSCSI boot parameters on the adapter, such as iSCSI IP address and CHAP informaiton
Change the boot sequence in the BIOS of the ESXi hosts so that iSCSI is the first option
install ESXi to the iSCSI target

If anythign on the SAN side changes, such as IP or CHAP, you will need to make that change on the ESXi Hosts iSCSI adapter
General Guidelines

Don't use DHCP for the iSCSI target IP
Ensure that the hosts can only see their corresponding boot LUN and not other hosts
Check vendor documentation for any iSCSI adapters you are using.

Create an NFS share for use with vSphere

Steps

Create a storage volume
Create a folder on that storage volume
Create a share for that folder
Allow the IP of your hose to access the storage
Give the IP of your host read/write access to the share

Enable/Configure/Disable vCenter Server storage filters

There are 4 different storage filters in vsphere

config.vpxd.filter.vmfsFilter: Filters out storage devices or LUNs that are already used by a VMFS datastore on any host managed by vCenter
config.vpxd.filter.rdmFilter: filters out any LUNs already referend as a RDM for any host managed by vCenter
config.vpxd.filter.SameHostandTransportsFilter: filters out LUNs that are unable to be used as a VMFS datastore extent

LUNs that aren't exposed on all the hosts that the datastore you are trying to extend is exposed to.
LUNs that are using a different storage type than the original datastore

config.vpxd.filter.hostRescanFilter: This filter automatically rescans and updates VMFS datastores after you perform datastore management operations.

Enable/Configure/Disable

Home -> Hosts and Clusters -> vCenter -> Settings -> Advanced Settings -> Edit
In the textbox labeled Key: enter in the value of the storage filter you want to enable or disable (.vmfsFilter, .rdmFilter, .SameHostAndTransportFilter or .hostRescanFilter) and in the textbox labeled Value: type True or False.

Configure/Edit hardware/dependent hardware initiators

Independent Hardware iSCSI adapters

Install the adapter based vendor documentation
Verify the adapter is installed correctly and configure it:

Home -> Hosts and Clusters ->Select host -> Manage,
Click Storage > Storage Adapters
If installed properly, you will see the new adapter in this list
Select the newly installed adapter and click the Properties...
From here you can change the default iSCSI name, alias and IP settings

Dependent Hardware iSCSI Adapters

The dependent iSCSI adapter will show up as a physical network adapter and as a physical storage adapter
Install the adapter based on vendor documentation
You’ll need to correlate the network adapter (vmnic) with the storage adapter in order to complete the configuration
Log into the vSphere Web client > click the Hosts and Clusters icon
From the left-hand tree select a host
On the right click Manage > click Storage Adapters
Find the adapter (vmhba) that corresponds to the physical network adapter of the dependent iSCSI adapter > click the Properties... hyperlink (under adapter details)
Click Edit
Enter in a name for the adapter (must be unique) > enter an alias if you’d like
Click the Network Binding tab > click Add
You should see the corresponding network adapter in the list. Bind that to the vmhba
If you don’t have a vmkernel adapter already set up on your iSCSI network, create one now
Ensure the networking policy for your iSCSI vmkernel adapter(s) has only one active physical adapter
Once you have setup your vmkernel adapter go back into the storage adapters tab (where you just were a few steps ago) and go under Network Binding
Click Add > bind the vmkernel adapter to the vmhba
You now need to discover the iSCSI target(s). While still in the Storage Adapters view select the iSCSI adapter > under Adapter Details click the Targets tab
You can do Dynamic or Static discovery, we’ll just talk about Dynamic discovery > click Dynamic Discovery and then click Add
Enter in the IP address or DNS name for the iSCSI target (on your storage system)
Click OK and rescan the iSCSI adapter

Enable/Disable software iSCSI initiator

Home -> Hosts and Clusters -> Select Host
Manage -> Storage -> Storage Adapters
Add icon -> Software iSCSI adapter

Configure/Edit software iSCSI initiator settings

Can configure iSCSI initiator for authentication, targets and other settings

Home -> Hosts and Clusters -> Select Host
Manage -> Storage -> Storage Adapters -> Select the adapter

Properties: Can change the name of the adapter
Targets: Can add dynamic or static iSCIS initiator targets

Configure iSCSI port binding

Home -> Hosts and Clusters -> Select Host
Manage -> Storage -> Storage Adapters -> select iSCSI adapter from the list
Click Network Port Binding -> Add button -> Select the port group/vkernal mapping that you want to bind together.

Enable/Configure/Disable iSCSI CHAP

Home -> Hosts and Clusters -> Select Host
Manage -> Storage -> Storage Adapters -> select iSCSI adapter from the list
Properties -> Authentication -> Edit
Authentication Methods

None
unidirectional CHAP if required by user
unidirectional CHAP unless prohibited by target
unidirectional CHAP
bidirectional CHAP

Select appropriate method and enter a name and secret configured on the target

Determine use cases for fiber channel zoning

Zoning:

Use single-initiator zoning or single-initiator-single-target zoning (more restrictive)
Defines which Host Bus Adapters (HBAs) can connect to which targets on the SAN. Objects that aren't zoned to one another, or outside of a prticular zone are not visible.
Reduces the number of LUNs and targets presented to a particular host
Controls/Isolates paths in your SAN fabric
Prevents unauthorized systems from accessing targets and LUNs

LUN Masking:

Exact same thing as zoning, but applied only for LUN-host mapping
Limits which hosts can see which LUNs
Can be done at the array layer or VMware layer

Compare and contrast array thin provisioning and virtual disk thin provisioning

Virtual Disk Thin Provisioning

Allows you to create virtual disks of a logical size that initially differs from the physical space used on a datastore. If you create a 40GB thin disk, It may initially use only 20GB of physical space and will expand as needed up to 40GB
Can lead to over-provisioning of storage.

Array Thin Provisioning

Thin provision a LUN at the array level
Allows you to create a LUN on your array with a logical size that initially differs from the physical space allocated - can expand up to logical size over time.
Array thin provisioning is not ESXi awarewithout using the storage APIs for array integration (VAAI) With a VAAI capable array, the array can integrate with ESXI, which at that point ESXI is aware that the underlying LUNs are thin provisioned.
Using VAAI you can monitor space on the thin provisioned LUNs and tell the array when files are freed (Deleted or removed) so the array can reclaim that free space)

Objective 3.2: Configure Software-defined Storage

Explain VSAN and VVOL architectural components

VSAN

Need at least 3 hosts for a VSAN cluster with minimum of 6GB memory
Need a mix of SAS and SSD drives or all SSD
Storgae Device must be local to the host, and have no preexisting partitions.
Each disk group will have one SAS and one flash drive
The latest format (2.0) of VSAN reuires 1% capacity per device.

VVOL

Home -> vCenter Inventory lists -> vCenter servers
Select vCenter from the list -> right click Manage -> Storage Providers -> Add
Type in Name for storage provider, URL for storage provider (Ususally VASA provider URL)
provide username and passwords
If you have a storage certificate check, use storage provider certificate.

Verify that the protocol endpoints exist
Change the PSP for the protocol endpoint

Determine the role of storage providers in VSAN

VSAN doesn't support multiple vkernal adapters on the same subnet
Multicast must be enabled.
Segment VSAN traffic on its own VLAN

Determine the role of storage providers in VVOLs

Explain VSAN failure domains functionality

Use fault domains to spread data across multiple hosts. a fault domain consists of one or more hosts. TO use fault domains you need a minimum of three hosts.

Configure/Manage VMware Virtual SAN

Configure

Home -> Hosts and Clusters -> right click on datacenter -> New Cluster

Give it a name, enable or disable HA or DRS, and mode (Auto or manual)
Add hosts to the VSAN (at least three)
Verify the datastore has been created.
Add a license to VSAN

under clusters still -> Manage -> Settings -> Configuration -> Licensing

Manage

From the inventory tree select the cluster you want to enable VSAN on
Manage -> Settings
Virtual SAN -> General -> Edit
Configure options

Create/Modify VMware Virtual Volumes (VVOLs)

Create

Home -> vCenter Inventory Lists -> vCenter Servers
Select vCenter from list -> right click Related Objects -> datastores -> create datastore
Select a location for the datastore, Select VVOL as type, enter a name for the new datastore and select backing storage container that will back the new datastore.
Select the hosts that will need access to the new datastore.

Verify

Home -> Hosts and Clusters -> select host
Manage -> Storage -> Protocol Endpoints

Configure Storage Policies

Configure

Home -> VM Storage Policies
If you see nothing, click on the Add button
If you want to create policies for datastores that aren't managed by a storage provider you can create a tag for datastore to define the capabilities

Browse to datastore
Manage -> Tags -> Add
Type in a name and description
Select a catagory, if none, create New Catagory
Enter in Category name and description and select objects to tag.

Create Policy for VM

Home -> VM Storage Policies -> Create new VM storage policy
Select vCenter server, type in name and description
Create a rule, you can create it based on data services (such as VSAN) or not.
Click Add tag-based rule
You can add another tag-based rule or you can add another rule set if you'd like.
Choose Comptabile or Incompatible, choose compatible
Ensure there are one or more datastores in the list

Apply Storage Policy to a VM

Home -> Hosts and Clusters -> Select host
Right click on VM -> VM Policies -> Edit VM Storage Policies
Selec tthe policy you'd liek to apply from teh dropdown and click Apply to All

Enable/Disable Virtual SAN Fault Domains

Home -> Hosts and Clusters -> Select VSAN cluster -> Right click Manage -> Settings
Under Virtual SAN click Fault domains -> Add
Type in a name, select hosts you want to put into the domain

Create Virtual Volumes given the workload and availability requirements

Collect VSAN Observer output

Create storage policies appropriate for given workloads and availability requirements

Configure VVOLs Protocol Endpoints

Objective 3.3: Configure vSphere Storage Multi-pathing and Failover

Explain common multi-pathing components

In case of a failure of any element in the SAN Network, such as an adapter, switch or cable, ESXi can switch to another physical path, which is not in a failed state. the process of path switching is known as path failover.

Differentiate APD and PDL states

Given a scenario, compare and contrast Active Optimized vs. Active non-Optimized port group states

Explain features of Pluggable Storage Architecture (PSA)

The Pluggable Storage Architecture consists of multiple components, top most being the multi-pathing plugin (MPP). The MPP can either be native (created by VMWare) or from a third party.
The VMware native multi-pathing plugin (NMP) provides two sub-plugins:

Path Selection Policy (PSP) plugin
Storage Array Type Plugin (SATP)

Here are some of the thing that the VMware NMP or third-aprty MPP are responsible for:

Provide logical and physical path I/O statistics
Loads and unloads multipathing plugins
Routes I/O requests for a specific logical device to the MPP managing that device.
Handles I/O queueing to the physical HBAs
Implements logical device bandwidth sharing between virtual machiens

The multi-pathing modules provide the following

Manage physical path claiming and unclaiming
Manage creation, registration and deregistration of logical devices
Associate physical paths with logical devices
Support path failure detection and remediation
Processes I/O requests to logical devices.

Understand the effects of a given claim rule on multipathing and failover

Explain the function of claim rule elements:

Vendor -
Model -
Device ID -
SATP -
PSP -

Change the Path Selection Policy using the UI

Determine required claim rule elements to change the default PSP

Determine the effect of changing PSP on Multipathing and failover

Determine the effects of changing SATP on relevant device behavior

Configure/Manage Storage Load Balancing

Storage load balancing is done by using multi-pathing. Multi-pathing is the ability to divide up I/O requests accross multiple paths to the same device.
To configure on an individual datastore

Home -> Storage -> Select Datastore -> Right click Manage -> Settings
Under Connectivity and Multipathing -> Select Host, see multipathing details
Click Edit Mutlipathing to change the type of load balancing/multi-pathing.

Differentiate available Storage Load Balancing options

Home -> Storage -> Datastore -> Right click Manage -> Settings -> Connectivity and Multipathing -> select host, see multipahting details -> Edit Multipathing
See teh different paths to that particular datastore and each paths status. Different statuses you might see are,

Active - used for active I/O. If an active path iscurrently accepting data it will be marked as Active (I/O)
Standby - This path will become active if the active path fails
Disabled - This means the path is disabled if the active path fails.
Dead - this path may have been one of the three aformentioned states but it currently has no connectivity to the datastore/device.

Differentiate available Storage Multi-pathing Policies

There are two types of multi-pathing policies that are available to storage device; Path Selection Policies (PSP) and Storage Array Type Policy (SATP)
Three types of PSPs available through Native Multipathing Plugin (NMP)

Round Robin - Most common PSP used, I/Os are sent down different available paths ar a set intervel. By default that set intervel is 1,000. Meaning 1000 I/Os are sent down one path and then the path is switched.
Most Recently Used (MRU) - I/Os all sent down the first working path that is discovered at boot time. Should the path go down or become disabled I/Os are sent down an alternative working path. There is no failback should teh old path become available again. This PSP is generally used for active/passive storage arrays
Fixed - sends all I/Os down the path that you set as the perferred path. If no path is set as the preferred path then the I/O is sent down the first working path that is discovered at boot time. If the perferred path coems back online then the I/O will then resume using the preferred path.

There are many SATPs that exist. A SATP is the plugin that gets associated with the different paths to a device. Typically the SATP relates to the storage vendor or to the type of storage array that teh devices are connected tp/ Such as VMW_SATP_SYMM is the storage array for an EMC Symmetrix array.

Configure Storage Policies

Home -> VM Storage Policies -> Add
or, if you want to create policies fro datastores that aren't managed by a storage provider you can create tags for datastores to define their capabilities. To Assign tags to a datastore

Browse to datastore
Manage -> Tags -> Add
Give name and description, select a category (or create new)

Create storage policy for a virtual machine

Home -> VM Storage Policies -> Create new policy
Select the vCenter server you want to create on from dropdown,
Type in name and description, Next, Next
Create a rule based on the data services (such as VSAN)
Select Category from doropdown list,
You can add more tags as needed.
Select Compatible or Incompatible,

Apply storage policy to Virtual Machine

Home -> Hosts and Clusters
Select host -> right click VM -> VM Policies -> Edit VM Storage Policies
Select the policy you'd like to apply.

Locate failover events in the UI

Objective 3.4: Perform Advanced VMFS and NFS Configurations and Upgrades

Describe VAAI primitives for block devices and NAS

Differentiate VMware file system technologies

VMFS5

Datastore capacity of 64TB
Block size standard at 1MB
Greater than 2TB storage devices for each VMFS5 extent
Supports VMs with greater than 2TB disks
Support for small files (1KB)
Ability to reclaim physical storage space on thin provisioned storage devices.

Upgrade VMFS3 to VMFS5

Compare functionality of newly created vs. upgraded VMFS5 datastores

Differentiate Physical Mode RDMs and Virtual Mode RDMs

RDM is a file that exists inside a VMFS volume which managed all the metadata fro teh raw device.
Some specific use cases for using RDMs

Storage resource management software
SAN management agents
Replication software
Microsoft failover clustering.

Physical Mode

When teh software needs direct access to the SCSI device.

Virtual Mode

When the software does not need direct access to the SCSI device.

Create a Virtual/Physical Mode RDM

Differentiate NFS 3.x and 4.1 capabilities

Compare and contrast VMFS and NFS datastore properties

Configure Bus Sharing

Configure Multi-writer locking

Connect an NFS 4.1 datastore using Kerberos

Create/Rename/Delete/Unmount VMFS datastores

Create

Home -> Storage -> datacenter -> related objects -> datastores
Add, Specify location, Select VMFS as type, give datastore a name,
Select host that has access to the LUN/device you want to create the datastore from
Select a partition configuraiton and select the size.

Rename

Home -> Storage -> datacenter -> related objects -> datastores
Right click ont he datastore you want and click Rename

Delete

Home -> Storage -> datacenter -> related objects -> datastores
Right click on the datastore you want, click Delete

Unmount

Home -> Storage -> datacenter -> related objects -> datastores
Right click on the datastore you want, click Unmount

Mount/Unmount an NFS datastore

Mount

Home -> Storage -> datacenter -> related objects -> datastores
Add, specify locaiton, selct NFS as type, select version of NFS, v3 or v4.1
Type in the name you want for the datastore
Enter int eh NFS share details

Specify the folder
Specify the server

Choose to mount as read only or not.
Select the host you wan tto mount the datastore to.

Unmount

Home -> Storage -> datacenter -> related objects -> datastores
Right click the datastore you want, click unmount.
Unselect the host you want to unmount from.

Extend/Expand VMFS datastores

Home -> Storage -> datacenter -> related objects -> datastores
Right click on your datastore you want to extend or expand
click Increase Datastore Capacity
Choose the device you want to use toexpand the datastore
From Parition Configuration, choose use all available partitions.

Place a VMFS datastore in Maintenance Mode

Home -> Storage -> datacenter -> related objects -> datastores
Right click on the datastore you want
Select Maintenance Mode -> Enter Maintenance Mode
You'll see faults under the faults tab, fix if needed.
The Migrations recommendations tab will list recomendations and are applied by default, you can choose to unapply them
Apply and continue

Select the Preferred Path/Disable a Path to a VMFS datastore

Home -> Storage -> datacenter -> related objects -> datastores
Connectivity and Multipathing -> Select host -> Edit Multipathing
select the path you want to be preferred.

Enable/Disable vStorage API for Array Integration (VAAI)

Home -> Hosts and Clusters -> select host -> right click host
Manage -> Settings
Change the following settings to 0 for disabled and 1 for enabled

HardwareAcceleratedMove
HardwareAcceleratedInit
HardwareAcceleratedLocking

Given a scenario, determine a proper use case for multiple VMFS/NFS datastores

Datastores sit on the backend storage that have physical disks configured in a particular way. If you have a requirement where some applications need more space, or need to be faster than others, creating multiple datastores with different characteristics will solve that requirement.
Disk contention could be a problem, having different datastore will allow you to spread those workloads over different physical disks
HA and resiliency - having multiple datastores allows you to spread your VMs accross them. If you lost a datastore all of your VMs won't go down, only VMs located on that particular datastore.

Objective 3.5: Setup and Configure Storage I/O Control

Describe the benefits of SIOC

Enable and configure SIOC

Home -> Storage -> Select datastore -> Manage -> Settings -> General
Next to Datastore Capabilities click Edit
Click the checkbox labeled Enable Storage I/O control
You can select the percentage of peak throughput and set a percentage at which point Storage I/O Control can kick in, or you can selecct manual and set a latency threshold in milliseconds

Configure/Manage SIOC

Home -> Storage -> Select datastore -> Manage -> Settings -> General
Next to Datastore Capabilities click Edit
You can manage the method of which Storage I/O control is implemented

Percentage of peak
Manual (Based in milliseconds)

You can also exclude I/O statistics from SDRS

Monitor SIOC

Home -> Storage -> Select datastore -> Manage -> Settings -> General
Monitor -> Performance
Timerange select realtime
You'll be able to see three different graphs to monitor Storage I/O control.

Storage I/O Control Normailized Latency
Storage I/O Control Aggregate IOPs
Storage I/O Control Activity

Differentiate between SIOC and Dynamic Queue Depth Throttling features

Given a scenario, determine a proper use case for SIOC

Compare and contrast the effects of I/O contention in environments with and without SIOC

Search This Blog

Ian Hood

VCP6-DCV Section 3: Configure and Administer Advanced vSphere 6.x Storage

Popular posts from this blog

VCP6-DCV Section 6: Backup and Recover a vSphere Deployment

VCP6-DCV Section 4: Upgrade a vSphere Deployment to 6.x

VCP6-DCV Section 7: Troubleshoot a vSphere Deployment