Chnoili
Enthusiast
Enthusiast

Disable Lifecyle Manager for a Cluster / HA activation fails

We have activated Lifecyle Manager (Image Based) for a Cluster (Manage all Hosts in the cluster with a single image) - since then the activation of vSphere HA fails with the message - A general system error occurred: Image is not valid. and Cannot complete the configuration of the vSphere HA agent on the host. "Setting desired image spec for cluster failed".

Is there an option to remove / disable the hosts for the image based updates (like remove baseline in the older versions) or to disable lifecycle manager for this specific cluster. I only found the information where to activate image based update for a cluster but not how to disable. imagebased.PNG

Maybe it is because VMware removed the 7.02 image from the selection and my hosts are installed with the 7.0 u2 Image, so the image is older than the actual installation of my hosts. Nevertheless i want to activate HA for my cluster.

0 Kudos
2 Replies
rgb99
Enthusiast
Enthusiast

I have the same or very similar problem.

Remediation of cluster failed Cannot download VIB: 'https://hostupdate.vmware.com/software/VUM/PRODUCTION/addon-main/addon/CIS/vib20/igbn/INT_bootbank_i...'. This might be because of network issues or the specified VIB does NOT exist or does NOT have a proper 'read' privilege set. Please make sure the specified VIB exists and is accessible from vCenter Server.

I selected these in the Image:

  • ESXi Version: 7.0 U2d - 18538813
  • Vendor Addon: DellEMC addon for PowerEdge Servers running ESXi 7.0 U2 A07
  • Dell EMC OpenManage Server Administrator, 10.1.0.0.4561
  • VMware Tools Async Release, 11.3.5
 
Now that there's a failure downloading the vib above, HA fails on the entire cluster.
 
If I try to enable HA on the cluster (after disabling), this message pops: Cannot complete the configuration of the vSphere HA agent on the host. "Setting desired image spec for cluster failed".
 
So after the Image remediation shoots itself in the foot, the following messages are shown after a pre-check: 
2 issue (sic) on the cluster
  •  Fault Domain Manager vib is missing on some of the hosts in cluster 'ClusterName'. Ensure the vib is installed on all hosts or try re-enabling vSphere HA on the cluster.
  •  'vSphere HA' service, running on 'cluster 'ClusterName'', reported issue: The HA constraints in the image spec have version whereas the expected version is 7.0.2-18455215.

I already opened Support Request #21288748512 for this issue. Unfortunately, I have to wait for a call or email because it is next to impossible to talk to someone.

 
0 Kudos
rgb99
Enthusiast
Enthusiast

I ended up resolving the issue on my own.

I downloaded the missing driver from https://customerconnect.vmware.com/en/downloads/details?downloadGroup=DT-ESXI70-INTEL-IGBN-1520&prod... and then I imported it into LCM. That cleared the download error.

I then disabled HA on the cluster and remediated one host at a time. After each remediation, the compliance check for the host failed with another error: 

A failure occurred when starting a host compliance check operation on host 'host.domain.com' : Error: com.vmware.vapi.std.errors.internal_server_error Messages: com.vmware.esx.task.exec.error<Failed to start the task. Please ensure the system has enough resources and retry.>

What solved that for me was a 'services.sh restart'. Afterwards, I checked compliance against the cluster and it came back green. I was then able to enable HA on the cluster, and the agent was able to install on each host without any problems.

I still have a meeting with VMware Support on Monday to figure out why I can't export the image as JSON/ISO/ZIP. That, too, throws an error.

0 Kudos