In this post we will continue covering the storage objective from the blueprint.To Better understand this objective I think we need to change the order and also some stuff in the list are redundant, so I will not follow subject by subject.
The following are the blueprint objective:
- Explain common multi-pathing components
- Differentiate APD and PDL states
- Compare and contrast Active Optimized vs. Active non-Optimized port group states
- Explain features of Pluggable Storage Architecture (PSA)
- Understand the effects of a given claim rule on multipathing and failover
- Explain the function of claim rule elements:
- Device ID
- 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
- Differentiate available Storage Load Balancing options
- Differentiate available Storage Multi-pathing Policies
- Locate failover events in the UI
PSA – Pluggable Storage Architecture :
Within VMkernel we have the following components:
- PSA -Which stand for Pluggable Storage Architecture
- NMP -Native Multipathing Plug-In , which is the VMware multipathing module
- MPP -Third-Party multipathing module
The VMware NMP includes:
- PSP -Path Selection Plug-In, which handles paths for for a given device
- SATP-Storage Array Type Plug-In, which handles path failover for a given storage array.
Lets start with PSP:
as mention PSP handle the path selection which we have three policies for that:
- MRU – Most recently used , When the path becomes unavailable, the host selects an alternative path. The host does not revert back to the original path when that path becomes available again. Default for ACTIVE-PASSIVE storage device.
- Fixed – The host uses the designated preferred path, if it has been configured.Otherwise, it selects the first working path discovered at system boot time.Default for ACTIVE-ACTIVE storage device.
- Round Robin (RR) – The host uses an automatic path selection algorithm rotating through all active paths
available paths when connecting to active-active arrays. RR is the only load balancing policy.
To assign a policy via the GUI: Under Storage >> Select datastore>>Setting >>Connectivity and multipathing >> host >> edit multipathing
To see it via the CLI , run the following command: esxcli storage nmp psp list
As mention before SATP is Storage Array Type Plug-In, which handles path failover for a given storage array.We can see the SATP policies only via the CLI (Get used to it).To see it via the CLI , run the following command: esxcli storage nmp satp list
When you start your ESXi host or rescan your storage adapter, the host discovers all physical paths to storage devices available to the host. Based on a set of claim rules, the host determines which multipathing plug-in (MPP) should claim the paths to a particular device and become responsible for managing the multipathing support for the device.
Before we change the policy lets see what devices are connected,run the following command: esxcli storage nmp device list .Here we can see that the SATP and PSP for out lab iSCSI device.
To see the rule list ,run the following command: esxcli storage core claimrule list.
Here you can see the rules for new devices.the rules are in order(think about it like a firewall if it meet it it not going down the list), you can see that usb/sata/ide all get the default plugin which is NMP, and at the end of the list , it will assign NMP plugin for anything else.(This can be changed)
Determine required claim rule elements to change the default PSP:
Basically you need to define new default policy: AS you can see in the pic below the default PSP for VMW_SATP_DEFAULT_AA is Fixed , to change it to MRU we ran the command : esxcli storage nmp satp set -s VMW_SATP_DEFAULT_AA -P VMW_PSP_MRU . We listed it again to confirm the change.
Compare and contrast Active Optimized vs. Active non-Optimized port group states:
I think VMware refers to ALUA (Asymmetric Logical Unit Access) connection.
When we have two paths to the storage and one is “Fast”(optimized) and the other one is “Slow”(non optimized)
- The defaults PSP for all VMW_SATP_ALUA is MRU
- The MRU select the optimized path from the VMW_SATP_ALUA or un-optimized path id there is no optimized path.
- will use the above path until better path available
Differentiate APD and PDL states:
APD: All Paths Down , happens when storage device loose all paths to the host.
- The operational state of the device changes to Dead or Error.
- All paths are shown as Dead.
- Datastores on the device are dimmed.
PDL:Permanent Device Loss, happen when storage device permanently fails or is administratively removed or excluded.
- The operational state of the device changes to Lost Communication.
- All paths are shown as Dead.
- Datastores on the device are grayed out.
Locate failover events in the UI:
You will need to “Decode” the /var/log/vmkernel.log . Please refer to this VMware KB 1003433 document : http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1003433 . to get more clues about the failover event.
Thanks for reading