VM FT supports only a single CPU??

Anyone heard about this?  Is it true?  Client saying they temporarily took it off (VM)while increased to multiple CPUs. Then did some maintenance to the VM and now the CPU is no longer an issue and can put it back and reinstate the Fault Tolerance for the VM.

I never heard of that. Thoughts?
LVL 17
Tiras25Asked:
Who is Participating?

[Product update] Infrastructure Analysis Tool is now available with Business Accounts.Learn More

x
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
Note: SMP (Multiprocessor) virtual machines are not supported with VMware Fault Tolerance at this time.

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1010631
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
also a reference here, to a recent document

http://communities.vmware.com/people/vmroyale/blog/2009/05/18/vmware-fault-tolerance-requirements-and-limitations

Ensure that the virtual machines are NOT using more than 1 vCPU. (SMP is not supported.)

So are you stating that you have a client that has more than 1 vCPU working in FT?

It would not surprise me, but what can be configured and what is supported by VMware specifications are different things. Officially if they have multiple vCPUs and FT working, it's not supported.
0
Tiras25Author Commented:
So if that multiple vCPU VM sitting on one physical Host, can it vMotion to another host in case the first fails?
Or is it a different kind of FT?
0
Defend Against the Q2 Top Security Threats

Were you aware that overall malware worldwide was down a surprising 42% from Q1'18? Every quarter, the WatchGuard Threat Lab releases an Internet Security Report that analyzes the top threat trends impacting companies worldwide. Learn more by viewing our on-demand webinar today!

Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
You can only vMotion if both host are up!

If a host fails, and you have VMware HA, the VM will start automatically on another Host.

0
Tiras25Author Commented:
What what kind of Fault Tolerance is not supported with SMP?
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
if you have enabled FT on a VM, it cannot be vMotioned.
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
FT
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
VMware vSphere Availability Features

VMware vMotion - Migrate Virtual Machines with Zero Downtime
http://www.vmware.com/products/vmotion/overview.html

VMware vMotion Product Brief
http://www.vmware.com/files/pdf/VMware-VMotion-DS-EN.pdf

VMware High Availability - Decrease Downtime and Reduce Risk
http://www.vmware.com/products/high-availability/overview.html

VMware High Availability (HA) Product Briefs
http://www.vmware.com/pdf/ha_datasheet.pdf
http://www.vmware.com/files/pdf/VMware-High-Availability-DS-EN.pdf

VMware Distributed Resource Scheduler (DRS) - Dynamic Resource Balancing
http://www.vmware.com/products/drs/overview.html

VMware Distributed Resource Scheduler (DRS) Product Briefs
http://www.vmware.com/files/pdf/VMware-Distributed-Resource-Scheduler-DRS-DS-EN.pdf
http://www.vmware.com/pdf/vmware_drs_wp.pdf

VMware Fault Tolerance - Deliver 24 X 7 Availability
http://www.vmware.com/products/fault-tolerance/overview.html

VMware Fault Tolerance Product Brief
http://www.vmware.com/files/pdf/VMware-Fault-Tolerance-FT-DS-EN.pdf

VMware vSphere Availability Guide ESX 4.1vSphere
http://www.vmware.com/pdf/vsphere4/r41/vsp_41_availability.pdf
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
Tiras25Author Commented:
Lots of information.  Thank you!!
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
No problems, lots of optins for Availability.
0
Tiras25Author Commented:
Exactly.  So what are other options to make VMs with multiple vCPUs Fault Tolerant or have some kind of redundancy?
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
VMware HA, DRS, vMotion.

Do you want Zero Downtime? you would have to look at Application based replication products like Doubletake;

Veeam Backup and Replication. (Replicate every 15 minutes)

(but you cannot use VMware Fault Tolerent Mode).
0
coolsport00Commented:
@Tiras25, MSCS is an option you can use for zero or near zero down time.
http://www.vmware.com/pdf/vsphere4/r41/vsp_41_mscs.pdf

With Replication, as in Veeam, there would be some downtime, albeit minimal. If using MSCS, like FT, there is no downtime at all.

Regards,
~coolsport00
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
@coolsport00: HA is faster than MSCS!
0
coolsport00Commented:
HA has downtime my friend...(again, albeit minimal)
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
@coolsport00: So does MSCS, test them!

Pull the plug on an SQL Clustered Server and compare to pull the plug on ESX host with HA. HA starts-up faster, than MSCS failover resources onto other nodes!
0
coolsport00Commented:
MSCS is "supposed" to have replicated data to a second node...no downtime
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
@coolsport00: it's also cheaper, because does not requrie two VMs (CPU, Memory, Storage) and 2 x Windows Server 2008 R2 Enterprise!

We have this debate all the time with SQL DBAs, that want SQL Clusters, when they have vSphere Enterprise Plus, with all the bells and whistles!

Single VM with HA, faster and not as complicated as SQL Clustered, 2 VMs! (with HA)
0
coolsport00Commented:
No extra OS license needed IF a datacenter license is used
0
coolsport00Commented:
Oh, and HA is not a "redundant" option. It only is for a host failure, not VM. So, if the orig VM is the thing that failed, MSCS (or FT) needs to be used.
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
coolsport00: Turn on VM and Application Monitoring, VM will be restarted if heartbeats are not received from VMware Tools.
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
@coolsport00:

 VM monitoring
off to watch TV, you have a good day! l8r
0
coolsport00Commented:
@hanccocka - that doesn't matter if the VM fails; I'm well aware of that. If there is some crash that a VM restart doesn't work, having a readily available 2nd VM, via FT or MSCS, is the recommended solution by VMware, not VM Monitoring and HA.
0
coolsport00Commented:
Oh, and @Tiras25, if it's vCenter (and a remote vCenter DB) VM you wanna make highly available, you can use vCenter Heartbeat, which works about like FT, but no restrictions like FT. Not sure why there are the FT restrictions there are....
0
markzzCommented:
Seems we have diverged from the original question.
Sorry Coolsport I go with Hanc Clustering is a wonderful idea but ha will generally give you recovery in around the same amount of time with no complications.
My general feeling is that clustering is near redundant  these days when your running in an ESX Enterprise or better farm environment.
Now what we want is SMP, vMotion support for FT..
Tiras25: forget FT for the moment until it matures. HA in a well designed environment will give you a 30second or so guest recovery. (and it's easy, painfree, and most importantly it works)
0
coolsport00Commented:
@markzz, as I mentioned, that may be well and good, but if the VM itself crashes, HA is a moot point. HA/VM Monitoring is not a redundant solution...it is a SPOF. For redundancy, you should have HA or MSCS. Yes, the question has diverged a bit, but for a highly available solution, the best is...well, it's probably FT, but a immature at the moment. The 2nd best is MSCS because it covers not only a host failure, but a VM failure as well.
0
coolsport00Commented:
I meant to say: "you should have 'FT' or MSCS" (not 'HA or MSCS') :)
0
Tiras25Author Commented:
So as I understand there are two typed of redundancy when it come to VMware.
FT and HA.  HA is MSCS clustering.  Correct?

What are the proc and cons for each?  PLease advice....
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
Not quite.

MSCS is Microsoft Cluster Services, or more recently, Failover clustering, the services failover on a Microsoft OS, to another Windows Host.

HA is a VMware Product and Feature called High Availability, when a VMware host server fails, all the VMs on that server, are restarted on other VMware hosts in what VMware call a Cluster,

Microsoft Clustering is at the Application Layer, VMware HA is at the Server layer, it's the VM which gets restarted because the host has failed.
0
Tiras25Author Commented:
Got it.  So with VMware HA there is another VM clone just sitting idle on another node.  What happens if the VM gets corrupted or crashes.  Would it restart a fresh VM on another physical host?
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
Not quite.

VMware HA -

1. Host A fails.
2. All VMs that were running on Host A, start executing on Host B.
(because Host A and Host B have access to the same storage, the VMs disks do not move anywhere, but where the VM process is moves)
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
VMware HA - can restart a VM if it detects it stops due to shutdown or crash via heartbeat.
0
Tiras25Author Commented:
Okay got it.  So not going away from my original question.  With VMware HA it supports only one single vCPU.
With MSCS it can support more multiple vCPUs but more expensive because need to double the OS licenses?  But almost no downtime right?
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
No, VMware HA supports multiple.

MSCS supports multiple.

Both VMware HA and MSCS you can expect an outage, when the Failover occurs.
0
Tiras25Author Commented:
So what's not supports multiple vCPU?
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
VMware FT (Fault Tolerence), which is another Feature.
0
Tiras25Author Commented:
Got it.  Sorry to bug.  Trying to see all the redundancy options for VMware.  Thank you.
0
Andrew Hancock (VMware vExpert / EE MVE^2)VMware and Virtualization ConsultantCommented:
No problems.
0
coolsport00Commented:
Ha...sorry for the late reply, but @hanccocka got it. To summarize:
HA => deals with VMs in a Cluster. If a host goes down, VMs will be restarted on remaining "live" (available) hosts in the cluster. No copies of the VMs are needed. Mutliple vCPUs supported.
FT => there is a copy of a VM that you want to be highly available. This copied VM cannot sit on the same host as the 'source' VM. Only 1 vCPU supported.
MSCS => basically like FT, but a MS OS solution. Multiple vCPUs supported.

Regards,
~coolsport00
0
Tiras25Author Commented:
I did work with DRS before.  Is DRS part of one of these or a completely separate redundancy control?
0
coolsport00Commented:
DRS is nothing more than a 'load balancing' solution, within a vSphere Cluster. What it does is balance VM resource load amongst all nodes (ESX/i hosts) in a cluster to best streamline performance. The Res Mgmt Guide further explains DRS, and the Availability Guide explains both FT and HA.
0
Tiras25Author Commented:
GOt it.  Thanks Cool!
0
coolsport00Commented:
No probs ;)
0
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
VMware

From novice to tech pro — start learning today.