Remote Console not accessing the VMs

[RodPin]

Hello!

I'm facing a quite strange issue as the Remote Session console cannot reach any of the VMs.

Initially, I thought that some policy implemented in the PC I'm using to access would be blocking it after hardening it, although, even after testing with another clean PC, I stil lcannot make the connection.

I think it's strange because I still can RDP to the Windows machines and also SSH into the Morpheus VM. I also can still Remote Console to the host computer.

This is the log when trying to connect to Morpheus VM.

==> /var/log/morpheus/guacd/current <==
2025-09-26_00:25:50.24630 guacd[23313]: INFO: Creating new client for protocol "vnc"
2025-09-26_00:25:50.24682 guacd[23313]: INFO: Connection ID is "$c19bf196-a88b-4471-8c15-e05c8e0644bb"
2025-09-26_00:25:50.25087 guacd[30330]: INFO: Cursor rendering: local
2025-09-26_00:25:50.25127 guacd[30330]: INFO: User "@c1b9fcca-fe27-40a0-9654-29d1fff91472" joined connection "$c19bf196-a88b-4471-8c15-e05c8e0644bb" (1 users now present)
2025-09-26_00:25:50.25154 guacd[30330]: ERROR: Unable to connect to VNC server.
2025-09-26_00:25:50.25356 guacd[30330]: INFO: User "@c1b9fcca-fe27-40a0-9654-29d1fff91472" disconnected (0 users remain)
2025-09-26_00:25:50.25366 guacd[30330]: INFO: Last user of connection "$c19bf196-a88b-4471-8c15-e05c8e0644bb" disconnected
2025-09-26_00:25:50.50462 guacd[23313]: INFO: Connection "$c19bf196-a88b-4471-8c15-e05c8e0644bb" removed.

Another thing that I've noticed is that if I try to clone a VM, I get an error due to low Cluster resources. I don't understand, as the Cluster has plenty of resources available.

Even if I try to Provison a VM with very low consuming resources, for example (1 CPU 512MB) Morpheus still will give me an error bur will create a instance with error.

I thanks in adance for any help provided.

[RodPin]

Adding some screenshots..

Cluster resources:

Cluster resources

Remote console window:

Remote Terminal windows

[RodPin]

I stil cannot connect to the Console, but at least found a workaround for the Cloning, so, please, don't close the post yet...

Basically, I did the clonning manually using virsh:

virt-clone --original=<source_vm_name> --name=<new_vm_name> --auto-clone

Then, I manually started the VM:

virsh start VM_NAME

Since I still could access the VMs using Windows RDP, I accessed the clone and adjusted the IPs as per my network configuration. Luckly, my VM had 2 interface cards configured, so I configured one first in the same network as another VM, and then from this other VM, RDP again and modified my RDP Network NIC IP.

Then, had to wait for the VM to show in the Cloud as "Unmanaged":

From Google Gemini:
- VM Discovery:
   Once the KVM host is successfully onboarded and the cloud is configured to "Inventory Existing Instances," Morpheus will automatically discover the VMs running on that KVM host during its regular cloud syncs (typically every five minutes by default).
   These discovered VMs will appear on the VMs tab of the Cloud detail page and also on the central Infrastructure > Compute > Virtual Machines list page. 

After that, just follow Morpheus instructions to allow it to manage the VM.

[RodPin]

Hi! Any news regarding this one? I'm still struggling to make it work...

Also, when I try to provison a new VM, Morpheus complains that I have reached the number of sockets, but I have only one host with one socket. I still need help...

[DA9]

I have the same problem.  No console access.  I am running 8.0.11 which was upgraded from 8.0.10.  Thanks.

[RodPin]

In my case, I'm still in 8.0.10 and managed to do most of the things using RDP, but looks like my windows broke and I have to troubleshoot...

[Craig_Manson]

Firstly have you changed the password of the user that the manager uses to connect to the hosts?

That was our problem with 8.0.10.  There is no logging or error messages for a bad password (annoyingly),  but you can tell that you have this form of the no console issue because the console screen will "Retry" 5 times before saying "Connection Failure" at the top left of the console tab.   This was about a month with support to sort this issue out. 

After upgrading to 8.0.11, the issue is back, but this time you just get a black console screen and no "connecting" messages.  There is an error in the Morpheus log. at least. 

We are working with support to resolve this one. 

[DA9]

No PW changes here and no "Retry" messages.  It looks like it connects, just to the black screen.  We'll be very interested in what Support tells you.  Thanks.

[RodPin]

Yesterday I had a call with HPE support. This is what we did to get access back to the Hypervisor:

1 - Using Putty to access the Host terminal: using "vi /etc/ssh/sshd_config", uncomment the line 97 ("AllowTCPForwad yes"). Then restart the ssh service ("systemctl restart ssh").

2 - Did the same thing as 1, but in the VME Linux. (Not sure if this helped, but didn't cause issues)

Those steps were recommended by @phan_hpe in his post (https://community.hpe.com/t5/hpe-morpheus-vm-essentials/vme-manager-console-issue/m-p/7257255#M1231)

3 - Go to "Compute > Virtual Machines > Edit", and clear the RPC IP and set the port to "5985". Do this for all VMs you have.

4 - Clear any "Guest Console" configuration and set it to "Disabled".

5 - Go to "Compute > Hosts > 'YOUR_HOST' > Edit" and make sure that the RPC Host IP, the username and password are correct.

6 - Still in the Host page, go to "Actions > Upgrade Agent". This may take a while and the progress can be followed using the tail command in the VME terminal ("tail -f current").

If you have more than one host, in the "Compute" page, just mark all of them and select "Upgrade Agent". Have in mind that this operation may take a while.

After upgrading the Agent, it was possible to restore Hypervisor access to the VMs.

7 - Upgrade the Agent for the other VMs. You can either do it by clicking "Actions > Upgrade Agent" or by downloading the script and running it in the VM using PowerShell with Admin rights.

Although, it was explained that version 8.0.10 had a bug that, if the VM is running a "broken" snapshot might be created. Restoring the VMs from this will break them. Unfortunately, it will be required to restore the VMs from a previous backup.  According to the support, this bug has been fixed in version 8.0.11.

Summarizing, it was a matter of upgrading the Hosts Agent. This way, whenever you upgrade Morpheus, it's also important to upgrade the Agents in the Host and in all the VMs you have. This part is not in the Upgrade steps, but is highly recommended as good practice. (https://support.hpe.com/hpesc/public/docDisplay?docId=sd00006670en_us&page=GUID-2F18C295-175A-4F00-BA7A-5DB2A068BA2B.html&docLocale=en_US)

Again, I have to express my gratitude to @PeterTzvetanov for his support! This ticket can be closed.

[DA9]

Wow!  That's some solution. There must have been an update along the way that made a major change.  I am slowing working through this list testing the steps.  Nothing has fixed it yet.  Thanks.