Nutanix NCP-MCI (6.10) Exam - Topic 4 Question 5 Discussion

Actual exam question for Nutanix's NCP-MCI (6.10) exam

Question #: 5
Topic #: 4

An administrator has configured AHV Metro Availability with Witness and is testing failover scenarios.

During testing, the administrator disconnects the primary and recovery clusters but Prism Central remains connected to the recovery site.

What are two expected system behaviors? (Choose two.)

AGuest VM I/O operations pause (freeze) until connectivity is restored.

BGuest VM I/O operations pause (freeze) until connectivity between Prism Central and the primary site is restored.

CGuest VMs failover automatically to the recovery cluster.

DGuest VMs continue to run on the primary cluster.

Show Suggested Answer

Suggested Answer: A, C

When connectivity between Metro clusters is lost, Nutanix Metro Availability ensures data integrity using Witness for automatic failover.

Option A (Guest VM I/O operations pause until connectivity is restored) is correct:

Metro Availability enforces data consistency, so I/O operations pause until failover is confirmed.

Option C (Guest VMs failover automatically to the recovery cluster) is correct:

The Witness VM detects the failure and initiates an automatic failover to the secondary cluster.

Option B is incorrect:

Prism Central does not control VM failover in Metro Availability.

Option D is incorrect:

The primary cluster is unreachable, so VMs cannot continue running there.

Nutanix Metro Availability Guide How Witness Handles Failover Scenarios

Nutanix KB I/O Freezing and Failover Behavior in Metro Clusters

by Thora at Mar 20, 2025, 10:21 AM

Limited Time Offer

25%

Off

Get Premium NCP-MCI (6.10) Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Desiree

2 months ago

Totally agree with A, but C is a bit iffy for me.

upvoted 0 times

...

Karma

2 months ago

D seems wrong, they should failover if disconnected.

upvoted 0 times

...

Denise

3 months ago

Wait, how can VMs keep running if both clusters are down?

upvoted 0 times

...

Percy

3 months ago

I think B is misleading, connectivity to the primary site doesn't matter here.

upvoted 0 times

...

Stacey

3 months ago

A and C are correct!

upvoted 0 times

...

Soledad

3 months ago

I practiced a similar question, and I think the I/O operations would pause until the primary site is back online, but I’m not completely confident.

upvoted 0 times

...

Lina

4 months ago

I feel like the guest VMs would continue running on the primary cluster since Prism Central is still connected to the recovery site.

upvoted 0 times

...

Tina

4 months ago

I think the VMs might failover automatically to the recovery cluster, but I need to double-check how the witness setup affects that.

upvoted 0 times

...

Lajuana

4 months ago

I remember something about I/O operations freezing when there's a disconnect, but I'm not sure if it's just for the primary site or both.

upvoted 0 times

...

This is a tricky one. I'll need to think through the different options carefully and make sure I understand the expected system behaviors in this specific scenario. The wording of the question is important, so I'll read it a few times to make sure I don't miss anything.

upvoted 0 times

...

Huey

4 months ago

I think the key here is that the primary and recovery clusters are disconnected, but Prism Central is still connected to the recovery site. That should mean the guest VMs don't fail over, but their I/O operations might pause until connectivity is restored.

upvoted 0 times

...

Angella

5 months ago

Okay, let's see. If Prism Central is still connected to the recovery site, I'm guessing the guest VMs won't automatically fail over. But I'm not sure if the I/O operations will pause or if the VMs will continue running on the primary cluster.

upvoted 0 times

...

Cathrine

5 months ago

Hmm, this question seems to be testing our understanding of AHV Metro Availability with Witness. I'll need to carefully consider the different scenarios and system behaviors.

upvoted 0 times

...

Lou

11 months ago

I'm with Melda on this one. B and C make the most sense to me. Unless the admin has a teleportation device, I don't see how the VMs can keep running on the primary cluster.

upvoted 0 times

Rima

9 months ago

Yeah, it makes sense for the VMs to failover to the recovery cluster in this scenario.

upvoted 0 times

...

Kristeen

9 months ago

I agree, it seems like the most logical outcome.

upvoted 0 times

...

Artie

9 months ago

C) Guest VMs failover automatically to the recovery cluster.

upvoted 0 times

...

Hortencia

9 months ago

B) Guest VM I/O operations pause (freeze) until connectivity between Prism Central and the primary site is restored.

upvoted 0 times

...

Gearldine

9 months ago

So, it's safe to say that guest VM I/O operations would pause until connectivity is restored and the VMs would automatically failover to the recovery cluster.

upvoted 0 times

...

Nina

9 months ago

Exactly, it wouldn't make sense for the VMs to keep running on the primary cluster if there is no connectivity.

upvoted 0 times

...

Craig

9 months ago

Yeah, if the primary and recovery clusters are disconnected, it makes sense for the VMs to failover to the recovery cluster.

upvoted 0 times

...

Chantell

10 months ago

User B

upvoted 0 times

...

Rosann

10 months ago

User A

upvoted 0 times

...

Sheldon

10 months ago

I agree with you, B and C seem like the most logical options.

upvoted 0 times

...

Elfrieda

11 months ago

I believe option D is incorrect. VMs should not continue to run on the primary cluster if connectivity is lost.

upvoted 0 times

...

Jonell

11 months ago

I'm not sure about option C. I think VMs might not failover automatically unless configured to do so.

upvoted 0 times

...

Melda

11 months ago

Ha! Guess the admin forgot to bring their magic wand to restore connectivity. Jokes aside, I'd go with B and C.

upvoted 0 times

...

Truman

11 months ago

I agree with Alesia. Option B also makes sense because Prism Central needs to communicate with the primary site for VM operations.

upvoted 0 times

...

Edna

11 months ago

I'm not sure about D, since the question says the primary and recovery clusters are disconnected. I think A and C are the right answers.

upvoted 0 times

Vivienne

10 months ago

That makes sense. So D is not correct in this scenario.

upvoted 0 times

...

Sena

10 months ago

Yes, I agree. Guest VM I/O operations will pause and VMs will failover to the recovery cluster.

upvoted 0 times

...

Glory

10 months ago

I think A and C are the right answers.

upvoted 0 times

...

Alesia

11 months ago

I think option A is correct because VM I/O operations should pause when connectivity is lost.

upvoted 0 times

...

Fidelia

11 months ago

C and D seem like the correct answers. The VMs should fail over automatically to the recovery cluster, but they should also continue running on the primary cluster until connectivity is restored.

upvoted 0 times

Kristel

11 months ago

That makes sense. The VMs should switch over to the recovery cluster, but it's good that they keep running on the primary cluster too.

upvoted 0 times

...

Merlyn

11 months ago

D) Guest VMs continue to run on the primary cluster.

upvoted 0 times

...

Paris

11 months ago

C) Guest VMs failover automatically to the recovery cluster.

upvoted 0 times

...