vCenter registered, but connection remains unstable | Nutanix Community
Skip to main content

Hi,

I am implementing a Nutanix Cluster with Lenovo HX5520, when I register the Prism in vCenter the process appears as completed but I cannot create virtual machines by Prism Element and when running the NCC I return communication errors with the vCenter is not stabilized.

 

I did the communication test of the CVMs with the vCenter on ports 80 and 443 and the connection worked successfully.

 

 

when checking by cli I see that the connection is ok, but it does not provide me with the settings of vcenter

 

follows the difference from another implementation

 

any idea what it might be?

Hey @Diego Azevedo Creating VMs from PE shouldn't be a problem. Any specific error message that you are getting?

 

Also, can you share the NCC report mentioning the problems with vCenter connectivity?

Hi @AnishWalia20 thanks for help, yes follow the NCC report

 

NCC Health Checks Summary

################################################################################
SUMMARY
################################################################################
+-----------------------+
| State         | Count |
+-----------------------+
| Pass          | 224   |
| Info          | 1     |
| Warning       | 2     |
| Fail          | 1     |
| Total Plugins | 228   |
+-----------------------+

################################################################################
DETAILED INFORMATION
################################################################################

Detailed information for check_vcenter_connection:
Node 10.83.141.14: 
FAIL: vCenter connection is not established. Unable to establish secure connection via vim to vCenter server 10.250.6.42 on port 80.
Refer to KB 3815 (http://portal.nutanix.com/kb/3815) for details on check_vcenter_connection

Detailed information for storage_container_mount_check:
Node 10.83.141.14: 
WARN: Failed to fetch mounted datastores with error vCenter connection could not be established.
Refer to KB 1888 (http://portal.nutanix.com/kb/1888) for details on storage_container_mount_check

Detailed information for vm_recovery_container_mount_check:
Node 10.83.141.14: 
WARN:  Failed to fetch mounted datastores with error vCenter connection could not be established.
Refer to KB 7911 (http://portal.nutanix.com/kb/7911) for details on vm_recovery_container_mount_check

Detailed information for cluster_version_check:
Node 10.83.141.14: 
INFO: This cluster is running AOS Version 5.11.2 which is End of Life. Kindly refer to the Nutanix support portal for EOL schedule.
Refer to KB 2720 (http://portal.nutanix.com/kb/2720) for details on cluster_version_check

 

Hey @Diego Azevedo, thanks a lot for the outputs. Right, as you mentioned NCC report flags the VC connectivity issues from PRISM.

 

I think due to “storage_container_mount_check” this check failing you are not able to create a VM as to create a VM we would need a container.

Also, another important thing is that the NTP settings on vCenter and PE match.

You can try out the below things:

a)Un-register and register the vCenter again.

b)Make sure the below are met:

  • vCenter requires port 80 for direct HTTP connections. Port 80 redirects requests to HTTPS port 443. A non-default port does not work.
  • You may see a recommendation by VMware to change the default vCenter port. Nutanix does not recommend changing the port for security reasons. This should be changed only for the special use case that Microsoft IIS is using port 80.
  • vCenter Registration done through Prism uses port 443. All other communication between Nutanix and vCenter Server occurs over port 80.
  • Nutanix cluster communicates with vCenter Server to obtain virtual machine information necessary for certain Nutanix cluster operations like Data Protection, VM Management, One-Click upgrades, Xtract, etc. If the vCenter Server is not registered or is not accessible, those operations may fail.
  • If vCenter Server is not registered or is not accessible, the configuration of the remote site as part of Data Protection may fail with the "Remote site is currently not reachable. Please try again later". Ensure both the primary and remote clusters are registered with vCenter Server. 
  • You may manually change every CVM's (Controller VM's) memory to 32 GB from vCenter/vSphere for passing the AOS 5.5 (or above) Pre-upgrade checks. Then perform the AOS upgrade. It may help the Nutanix cluster to skip the steps on changing CVM's memory automatically by using the standard port, which may cause errors if non-standard ports were being used.

 

Otherwise I would suggest you to engage vCenter on this or open a Support case with us. 

Terms & Conditions