Dear All,
Currently my Nutanix cluster consist of 3 Nutanix block, 5 node running on AHV 20170830.171
1 block of 3 NX-1065-G5 node
1 block of 1 NX6035-G5
1 block of 1 NX6035-G5
Recently, one of the NX6035-G5 is down due to hardware issue.
Somehow, all the VMs hosted on this node not able to auto failover to the others node.
Upon checking with Nutanix, we were informed that the failover failed as we don't have sufficient available memory.
We were told that AHV will perform the failover for all the fail VMs to one particular node which have sufficient resources to accommodate it; like moving the whole load from the fail node to another node.
Previously we are using VMware where it will failover VM by VM to all the available nodes.
Anyone encounter similar issue with AHV?
This topic has been closed for comments
Userlevel 3
+5
- Nutanix Employee
- 112 replies
-
24 May 2019
Hi @boonsoon.on-72320
Can you please share the case number than I can take a look and check additional information on the case.
But, keep in mind that there are different ways to reserve resources for HA on AHV and depending of how you configured the HA, you may expect different behaviors.
Take a look at the following links for a good introduction about AHV HA:
https://next.nutanix.com/blog-40/episode-1-acropolis-hypervisor-vm-ha-part-1-5595
https://next.nutanix.com/blog-40/episode-1-acropolis-hypervisor-vm-ha-part-2-5597
Can you please share the case number than I can take a look and check additional information on the case.
But, keep in mind that there are different ways to reserve resources for HA on AHV and depending of how you configured the HA, you may expect different behaviors.
Take a look at the following links for a good introduction about AHV HA:
https://next.nutanix.com/blog-40/episode-1-acropolis-hypervisor-vm-ha-part-1-5595
https://next.nutanix.com/blog-40/episode-1-acropolis-hypervisor-vm-ha-part-2-5597
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.