Yellow Bricks

by Duncan Epping

update 3

Stretched cluster witness failure resilience in vSAN 7.0

· · 2 Comments

Cormac and I have been busy the past couple of weeks updating the vSAN Deep Dive to 7.0 U3. Yes, there is a lot to update and add, but we are actually going through it at a surprisingly rapid pace. I guess it helps that we had already written dozens of blog posts on the various topics we need to update or add. One of those topics is “witness failure resilience” which was introduced in vSAN 7.0 U3. I have discussed it before on this blog (here and here) but I wanted to share some of the findings with you folks as well before the book is published. (No, I do not know when the book will be available on Amazon just yet!)

In the scenario below, we failed the secondary site of our stretched cluster completely. We can examine the impact of this failure through RVC on vCenter Server. This will provide us with a better understanding of the situation and how the witness failure resilience mechanism actually works. Note that the below output has been truncated for readability reasons. Let’s take a look at the output of RVC for our VM directly after the failure.

VM R1-R1:
Disk backing:
[vsanDatastore] 0b013262-0c30-a8c4-a043-005056968de9/R1-R1.vmx
RAID_1
RAID_1
Component: 0b013262-c2da-84c5-1eee-005056968de9 , host: 10.202.25.221
votes: 1, usage: 0.1 GB, proxy component: false)
Component: 0b013262-3acf-88c5-a7ff-005056968de9 , host: 10.202.25.201
votes: 1, usage: 0.1 GB, proxy component: false)
RAID_1
Component: 0b013262-a687-8bc5-7d63-005056968de9 , host: 10.202.25.238
votes: 1, usage: 0.1 GB, proxy component: true)
Component: 0b013262-3cef-8dc5-9cc1-005056968de9 , host: 10.202.25.236
votes: 1, usage: 0.1 GB, proxy component: true)
Witness: 0b013262-4aa2-90c5-9504-005056968de9 , host: 10.202.25.231
votes: 3, usage: 0.0 GB, proxy component: false)
Witness: 47123362-c8ae-5aa4-dd53-005056962c93 , host: 10.202.25.214
votes: 1, usage: 0.0 GB, proxy component: false)
Witness: 0b013262-5616-95c5-8b52-005056968de9 , host: 10.202.25.228
votes: 1, usage: 0.0 GB, proxy component: false)

As can be seen, the witness component holds 3 votes, the components on the failed site (secondary) hold 2 votes, and the components on the surviving data site (preferred) hold 2 votes. After the full site failure has been detected, the votes are recalculated to ensure that a witness host failure does not impact the availability of the VMs. Below shows the output of RVC once again.

VM R1-R1:
Disk backing:
[vsanDatastore] 0b013262-0c30-a8c4-a043-005056968de9/R1-R1.vmx
RAID_1
RAID_1
Component: 0b013262-c2da-84c5-1eee-005056968de9 , host: 10.202.25.221
votes: 3, usage: 0.1 GB, proxy component: false)
Component: 0b013262-3acf-88c5-a7ff-005056968de9 , host: 10.202.25.201
votes: 3, usage: 0.1 GB, proxy component: false)
RAID_1
Component: 0b013262-a687-8bc5-7d63-005056968de9 , host: 10.202.25.238
votes: 1, usage: 0.1 GB, proxy component: false)
Component: 0b013262-3cef-8dc5-9cc1-005056968de9 , host: 10.202.25.236
votes: 1, usage: 0.1 GB, proxy component: false)
Witness: 0b013262-4aa2-90c5-9504-005056968de9 , host: 10.202.25.231
votes: 1, usage: 0.0 GB, proxy component: false)
Witness: 47123362-c8ae-5aa4-dd53-005056962c93 , host: 10.202.25.214
votes: 3, usage: 0.0 GB, proxy component: false)

As can be seen, the votes for the various components have changed, the data site now has 3 votes per component instead of 1, the witness on the witness host went from 3 votes to 1, and on top of that, the witness that is stored in the surviving fault domain now also has 3 votes instead of 1 vote. This now results in a situation where quorum would not be lost even if the witness component on the witness host is impacted by a failure. A very useful enhancement to vSAN 7.0 Update 3 for stretched cluster configurations if you ask me.

vSAN 7.0 U3 feature overview

· · 16 Comments

In this blog post, I want to go over the features which have been released for vSAN as part of 7.0 U3. It is not going to be a deep dive, just a simple overview as most features speak for themselves! Let’s list the feature first, and then discuss some of them individually.

  • Cluster Shutdown feature
  • vLCM support for Witness Appliance
  • Skyline Health Correlation
  • IO Trip Analyzer
  • Nested Fault Domains for 2-Node Clusters
  • Enhanced Stretched Cluster durability
  • Access Based Enumeration for SMB shares via vSAN File Services

I guess the Cluster Shutdown feature speaks for itself. It basically enables you to power off all the hosts in a cluster when doing maintenance. Even if those hosts contain vCenter Server! If you want to trigger a shutdown, just right click the cluster object, go to vSAN, select “shutdown cluster” and follow the 2-step wizard. Pretty straight forward. Do note, besides the agent VMs, you will need to power-off the other VMs first. (yes, I requested this to be handled by the process as well in the future!)

The Skyline Health Correlation feature is very useful for customers who are seeing multiple alarms being triggered and are not sure what to do. In this scenario, starting with 7.0 U3, vSAN will now understand the correlation between the events and inform you what the issue is (most likely) and show you which other tests it would impact. This should enable you to fix the problem faster than before.

IO Trip Analyzer is also brand new in 7.0 U3. I actually wrote a blog post on the subject separately and included a demo, I would recommend watching that one. But if you just want the short summary, the IO Trip Analyzer basically provides an overview of latency introduced at every layer in the form of a diagram!

Nested Fault Domains for 2-node clusters has been on the “wish list” of some of our customers for a while. It is a very useful feature for those customers who want to be able to tolerate multiple failures even in a 2-node configuration. The feature requires you to have at least 3 disk groups per host, in each of the 2 hosts, and will then enable you to have “RAID-1” across those two hosts and RAID-1 within the host (Or RAID-5 if you have sufficient disk groups). If a host fails, and then a disk fails in the surviving host, the VMs would still be available. Basically a feature for customers who don’t need a lot of compute power (3 hosts or more), but do need added resiliency!

Enhanced Stretched Cluster durability (also applies to 2-node) is a feature that Cormac and I requested a while back. We requested this feature as we had heard from a few customers that unfortunately, they had found themselves in a situation where a datacenter would go offline, followed by the witness going down. This would then result in the VMs (only those which were stretched of course) in the remaining location also be unusable, as 2 out of the 3 parts of the RAID tree would be gone. This would even be the case in a situation where you would have a fully available RAID-1 / RAID-5 / RAID-6 tree in the remaining datacenter. This new feature now prevents this scenario!

Last, but not least, we now have support for Access Based Enumeration for SMB shares via vSAN File Services. What does this mean? Pre vSAN 7.0 U3, if a user had access to a file share the user would be able to see all folders/directories in this share. Starting with 7.0 U3 when looking at the share, only the folders that you have the appropriate permissions for will be displayed! (More about ABE here)

Capacity Overview in vSAN 6.7 U3

· ·

I just wanted to do a short post before the weekend, in vSAN 6.7 U3 there’s a great capacity overview screen. It shows a couple of things, first of all, it provides a simple bar that shows “data written”, “reserved space” and “free space”. The second section provides you the ability to figure out what happens to your capacity consumption if you would change the policy on all VMs. The third section gives you a nice breakdown of the capacity per category and a great circular diagram which shows immediately what kind of data is consuming your capacity. Very useful!

vSAN 6.x customer? vSphere 6.0 Update 3 is out

· ·

Are you a vSAN 6.x customer? vSphere 6.0 Update 3 is out! There are a bunch of important fixes and improvements (checksumming performance for instance) in Update 3, so I would highly recommend looking in to it and testing it out.