Design, Storage, Uncategorized, Virtualization, vmware, VSAN

Core Knowledge vSAN HBA

The fundamentals cannot be over-emphasized. You need to ensure that the key components of your vSAN host is configured per recommendations.

Just a reminder of the HBA controller configuration.

  1. Make sure the device is on the Hardware Compatibility Guide (HCG) 
  2. And verify the firmware is up-to-date.

I have seen first hand what impact different firmware can have on your environment.

Example: Dell Perc H310

Controller queue depth impacts the rebuild/resync times. A low controller queue depth may impact the availability of your production VMs during rebuild/resync. A minimum queue depth of 256 is required in vSAN. Some vSAN Ready Node profiles require minimum queue depth of 512, All Flash configs.

For more details see this: vSAN Hardware Quick Reference Guide

The availability of vSAN and VMFS can be vying for the same resource; the HBA.

Do NOT mix Disk Access modes to your Host Bus Adapter (HBA) also called an I/O Controller. Pass through configuration is preferred, but RAID-0 can work. vSAN prefers to have a more direct access to the device attached to the I/O Controller.  So for example if the HBA is setup with some logic configuration the groups all the devices together before presenting to the ESXi host then you have some prep work to do. Several array controllers do not support pass through mode,  to use this type of controller for vSAN, we need to create a single disk RAID-0 group for every SSD and HDD.



Dell PERC 740


  • RAID levels access for the devices attached.
  • vSAN and VMFS devices on same HBA.

From the VMware KB:

  • Do not mix the controller mode for vSAN and non-vSAN disks.
    • If the vSAN disks are in pass-through/JBOD mode, the non-vSAN disks must also be in pass-through/JBOD mode.
    • If the vSAN disks are in RAID mode, the non-vSAN disks must also be in RAID mode.
    • Mixing the controller mode will mean that various disks will be handled in different ways by the storage controller. This introduces the possibility that issues affecting one configuration could also affect the other, with possible negative consequences for vSAN.

If you absolutely must use the same HBA:

  1. limit the use of the VMFS that is sharing the HBA with vSAN.
  2. AND DO NOT USE RDM for that shared device/HBA
  3. DO NOT have the boot device on the same vSAN controller
  • If the non-vSAN disks are in use for VMFS, the VMFS datastore should be used only for scratch, logging and coredumps.
    • Virtual machines should not be running from a disk or RAID group that shares its controller with vSAN disks or RAID groups.
    • ESXi host installation is permitted on non-vSAN disks attached to same controller.
  • Do not pass through non-vSAN disks to virtual machine guests as Raw Device Mappings (RDMs).

The number and type of drives plus their disk group configuration is not covered here but another topic of important discussion!




EMC, vmware, VSAN

Is your IT infrastructure an Oil Tanker?

I had the opportunity to attend an Avnet/EMC/VMware/Brocade sponsored for channel partners EVO-RAIL VSPEX Blue BootCamp.

In a nutshell it was all you can drink information from a firehose— about EVO-RAIL specifically the EMC VSPEX BLUE.

EVO RAIL is a new beast of an animal. It is a different breed. No, not in and single dimension you measure. The combination of technology presented is a synergy. Definition: Synergy is the creation of a whole that is greater than the simple sum of its parts.

Yes; you can get the the form factor for compute separately. You can also do the same for VSAN and ESXi vSphere 5.5 and networking. but you cannot get what the entire VSPEX BLUE offering of EVO RAIL provides TOGETHER.

I get ahead of myself.

You have to have perspective to understand where we are today. To me that means if you don’t know where you come from you  cannot know where are you today and where you will be tomorrow. It is all relative.

EVO RAIL is a clustered system. It is a Datacenter in a 2U form factor. Not just compute but a modern hyper converged solution.


Sure. Another IT buzzword. Is it just talk? I would say no.

Everyone is talking about it but only a few are “doing it”… more often than not the IT industry is a buzz with the new technology of the day.
In this case it is Hyper-converged.
To put it simply storage is local to compute. What a minute how is that different than 15 years ago when Client-Server model was the norm and storage was already local to the compute. Compute meaning the processing of the server CPU. Well lots has changed.
How is it better. A snapshot of what is the current available technology:
  • Compute is way, way faster and more dense.
  • Networks are 10 gigabit vs Fast Ethernet 100 Base-T or FDDI Optical rings are no longer the only viable choice.
  • Storage is IOPs crazy.
  • And add to that the agility of VMware Virtualization!
So look at the speed of compute, network and storage. Technology will continue to get faster and better (lower cost for the return on investment).
But what really hasn’t changed much is the complexity of the solution. There are many moving parts but how do you delivery your solution today to support legacy applications and have the agility to respond to changing business objectives.
The old phrase “turning an oil tanker on a dime”. Is IT today an Oil tanker? Does your private cloud have the agility your business requires? How will the current toolset respond? How will your staff? Oh what was that “IT staffing has been reduced and that is a trend that hasn’t gone away”
IT organizations are forced to do more with little… queue in viable alternatives…
Do you outsource.. the simplest short term gain, but not always the best long term investment.
I view EVO rail as a datacenter in a box. EVO-RAIL VPSEX Blue is Not the Cluster in a box solution but much much more.
This reminds me of the forerunning of MCSC cluster in a box. Like I said.. perspective. Where has the IT industry been before relative to where it is today. I recall back in the day when cluster in a box was a viable (and the best solution at the time) era 2000. That was only a high-availablity MSCS cluster with one node active at a time. 
Yes, I deployed and supported a few of these solutions. It was cutting edge back then.
Adjusted for inflation:
$1000.00 USD in 2000 is $1395.20
so the CL1850 was $27,864.00 (2000 $)
or $38,569 in 2015 $.
But what did you get for your IT dollar?
Each “node” in the CL1850
  • 1 GB RAM (128-MB 100-MHz registered ECC SDRAM memory)
  • 2 Pentium III processors @550MHz
  • 3 NICS (one dedicate for internode communication) 100 BaseT
  • 2 RAID controllers
Shared Storage System:
  • 218.4 GB (6 x 36.4-GB 1′′ Ultra3 10,000 drives)
Form Factor 10U
Fast forward to 2015….What does EVO-RAIL VSPEX BLUE PROVIDE:
Each EMC VSPEX BLUE appliance includes:
  • 4 nodes of integrated compute and storage, including flash (SSD) and HDD — 2U
  • VMware EVO:RAIL software including VMware Virtual SAN (VSAN), Log Insight
        12 cores @ 2.1GHz per node
  • 128 or 192 GB memory per node
  • Choice of 10 Gigabit Ethernet network connectivity: SFP+ or RJ45
  • Drives: up to 16 (four per node)
  • Drives per node: 1 x 2.5” SSD, 3 x 2.5” HDD
  • Drive capacities
    • HDD: 1.2TB (max total 14.4TB)
    • SSD for caching: 400GB (max total 1.6TB)
               14.4TB capacity RAW
Additional Software Solutions – Exclusive to VSPEX-BLUE
  • EMC VSPEX BLUE Manager providing a system health dashboard and support portal
  • EMC CloudArray to expand storage capacity into the cloud (license for 1 TB cache and 10 TB cloud storage included)
  • VMware vSphere Data Protection Advanced (VDPA) for centralized backup and recovery
  • EMC RecoverPoint for Virtual Machines for continuous data protection of VMs. Includes licenses for 15 VMs.

The question still remains. Is your IT infrastructure an Oil Tanker? OR Can you turn on a dime??

How does your IT respond the the ever changing business demands?

Is EVO RAIL for everyone? There are a lot of use cases that EVO RAIL VSPEX BLUE will work perfect for. But,  No it isn’t for everyone.  BUT what it does is usher in a new consumption of IT that is different manner. You will not have to be provision your datacenter in the same piece meal function. You can commoditize that to a pre-validated solution that is supported by a single vendor.


This graphic has a lot more details than this single blog post can explain! I will try to explain each section that helps to make VSPEX BLUE a different redefined EVO RAIL solution.


– VSPEX BLUE Manager users can conveniently access electronic services, such as the EMC knowledge base articles, access to the VSPEX Community for online and real-time information and EMC VSPEX BLUE best practices.


– ESRS is a two-way, secure remote connection between your EMC environment and EMC Customer Service that enables remote monitoring, diagnosis, and repair – assuring availability and optimization of your EMC products






– Built in Deduplicated Backups, powered by EMC Avamar


– Block and File upto 10 TB FREE. “EMC CloudArray software provided scalable cloud based storage with your choice of many leading cloud providers enabling limitless Network Attached Storage, offsite backup and disaster recovery and the ability to support both Block and File simply”


– Built into the VSPEX MANAGER dashboard. This unique feature enables customers to browse complementary EMC and 3rd party products that easily extend the capabilities of the appliance.

Again there are a lot of take aways for the VSPEX BLUE – EVO RAIL solution. Contact me if you more information.
Troubleshooting, Uncategorized, Virtualization, vmware, VSAN

VMWARE Virtual SAN networking

VSAN networking can be a bit tricky to troubleshoot. Before I go deeper into the topic here is a very important concept to remember about VSAN clusters.

Given any VSAN cluster remember the following:

** “Introduction to Virtual SAN Networking

Before getting into network in detail, it is important to understand the roles that nodes/hosts can play in Virtual SAN. There are three roles in Virtual SAN: master, agent and backup. There is one master that is responsible for getting CMMDS (clustering service) updates from all nodes, and distributing these updates to agents. Roles are applied during cluster discovery, when all nodes participating in Virtual SAN elect a master. A vSphere administrator has no control over roles.”

** from Cormac’s troubleshooting guide

That is a lot to digest but if break it down you can see some key principles about a VSAN cluster to remember.

The roles in VSAN:
A master
B agent
C backup.

There is one master.
If you see more than one master there is something not quite right with you VSAN CLUSTER.

The VSAN admin does not control which node will be the master.

Log into each node of a three node VSAN. The normal pre-req for troubleshooting make sure ssh is enabled.

Run the following command on each node:
~ # esxcli vsan cluster get

Cluster Information will output below.

Cluster Information
Enabled: true

Current Local Time: 2015-03-30T22:38:38Z
Local Node UUID: 55197cee-f530-4966-5ea6-a0369f58b8e4
Local Node State: MASTER
Local Node Health State: HEALTHY
Sub-Cluster Master UUID: 55197cee-f530-4966-5ea6-a0369f58b8e4
Sub-Cluster Backup UUID:
Sub-Cluster UUID: 551374b5-03f9-7bd6-6257-a0369f58b8e8
Sub-Cluster Membership Entry Revision: 0
Sub-Cluster Member UUIDs: 55197cee-f530-4966-5ea6-a0369f58b8e4
Sub-Cluster Membership UUID: a5ce1955-f5e5-5663-d338-a0369f58b8e4

Node 2
~ # esxcli vsan cluster get
Cluster Information
Enabled: true
Current Local Time: 2015-03-30T22:38:38Z
Local Node UUID: 55197cee-f530-4966-5ea6-a0369f58b8e4
Local Node State: MASTER
Local Node Health State: HEALTHY
Sub-Cluster Master UUID: 55197cee-f530-4966-5ea6-a0369f58b8e4
Sub-Cluster Backup UUID:
Sub-Cluster UUID: 551374b5-03f9-7bd6-6257-a0369f58b8e8
Sub-Cluster Membership Entry Revision: 0
Sub-Cluster Member UUIDs: 55197cee-f530-4966-5ea6-a0369f58b8e4
Sub-Cluster Membership UUID: a5ce1955-f5e5-5663-d338-a0369f58b8e4

Node 3
~ # esxcli vsan cluster get
Cluster Information
Enabled: true
Current Local Time: 2015-03-30T22:56:46Z
Local Node UUID: 54f9dc6f-8674-f412-364d-a0369f58b5a8
Local Node State: BACKUP
Local Node Health State: HEALTHY
Sub-Cluster Master UUID: 551374b5-03f9-7bd6-6257-a0369f58b8e8
Sub-Cluster Backup UUID: 54f9dc6f-8674-f412-364d-a0369f58b5a8
Sub-Cluster UUID: 551374b5-03f9-7bd6-6257-a0369f58b8e8
Sub-Cluster Membership Entry Revision: 1
Sub-Cluster Member UUIDs: 551374b5-03f9-7bd6-6257-a0369f58b8e8, 54f9dc6f-8674-f412-364d-a0369f58b5a8
Sub-Cluster Membership UUID: d6da1955-e2f8-38eb-d7f0-a0369f58b8e8

See the image below for the error seen in the web client.

From the output above can you see the problem?