Provisioning instances greyed out, configuration issue?

[se7]

Hi,

When I'm trying to provision an instance within the GUI, both vmware and hpe vm is greyed out, from watching the guide within the "press material" there is a creation of a cluster, is this necessary if i just want to be running from 1 server? I listened to the explanation of creating a cluster due to whats been created so far is only the "physical cluster" meanwhile the "virtual cluster" also needs creation.

Assuming i want to provision just within one ubuntu system, is there a requirement of creating a cluster? I tried to do this but ran in to a configuration error of: "SSH Connectivity Failure for Host 192.168.1.254: Cannot set property 'pty' on null object"

This is just from testing, assuming i could create a virtual cluster on the same address the hpe-vm is ran from. This sounds wrong but i tried regardless.

Is there any insight in how the "backend" is structured, comparing to proxmox whereas theres no clustering involved within a one node setup. I fail to find documentation explaining point blank what's required to start provisioning VM's after installing. Or is there any logs that could tell me any potential errors besides the hpe-vm.log?

[PSPrakash]

You must create a Single Node Cluster for provisioning VMs.

Ref:    "Creating the first HPE VME cluster"     https://hpevm-docs.morpheusdata.com/en/latest/vme_getting_started/vme_getting_started.html 

[se7]

I've tried creating the cluster but when i put in the ip of the mgmt interface i get: "SSH Connectivity Failure for Host 192.168.1.254: Cannot set property 'pty' on null object"

I assume i should be able to put ens18 which is the mgmt interface of the hpe-vm, but it returns the error which googling leads me nowhere or i am simply not smart enough. Is this a correct configuration?

[DiegoDelgado]

Hi,

you need to add the interface name from your host, the same you added when deploying the VME manager VM

Hope this helps

[se7]

When deploying from hpe-vm i chose ens18 to be the management interface, its the one im trying to put in to the cluster also, but it returns an ssh error, ssh is installed and works sshing normally. Is there additional ssh configuration thats required? I don't understand what i'm missing, the video example does the same without ssh error. 

[DiegoDelgado]

Can you share a list of your /etc/netplan files? Sometimes cloud-init does some changes that confuse the underlying OS

[se7]

It seemed to have a bricked network config after rebooting as hpe gui is no longer available. 

network:
bridges:
mgmt:
addresses:
- 192.168.1.254/24
interfaces:
- ens18
nameservers:
addresses:
- 192.168.1.203
search: []
openvswitch: {}
routes:
- to: default
via: 192.168.1.240
version: 2

doing a sudo netplan apply gives:  

/etc/netplan/60-mvm-mgmt.yaml:7:9: Error in network definition: mgmt: interface 'ens18' is not defined
- ens18
^

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: ens18: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel master ovs-system state UP group default qlen 1000
link/ether bc:24:11:f6:a0:f4 brd ff:ff:ff:ff:ff:ff
altname enp0s18
3: ovs-system: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether 76:47:22:ce:36:de brd ff:ff:ff:ff:ff:ff
4: mgmt: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000
link/ether bc:24:11:f6:a0:f4 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.254/24 brd 192.168.1.255 scope global mgmt
valid_lft forever preferred_lft forever
inet6 fe80::be24:11ff:fef6:a0f4/64 scope link
valid_lft forever preferred_lft forever
5: virbr0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
link/ether 52:54:00:4f:8a:e4 brd ff:ff:ff:ff:ff:ff
inet 192.168.122.1/24 brd 192.168.122.255 scope global virbr0
valid_lft forever preferred_lft forever
6: vnet0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel master ovs-system state UNKNOWN group default qlen 1000
link/ether fe:54:00:6c:52:6b brd ff:ff:ff:ff:ff:ff
inet6 fe80::fc54:ff:fe6c:526b/64 scope link
valid_lft forever preferred_lft forever

it still has internet access and ssh access but netplan seems to have given up with the bridge config. How would one go about resolving this?

[se7]

Files in the directory are:

hpe@hpe:/etc/netplan$ ls
60-mvm-mgmt.yaml mvmbackup-20250310-112702

[DiegoDelgado]

Something's funny in that config.....
You can try adding a 01-base.yaml file with the following content:

[se7]

It's a incredibly confusing automation when trying to create the HpeVM cluster, it's incredibly buggy from my current experience, I'll elaborate with what i got "kind of working"

After getting to the configure menu in the creation of the hpevm cluster it can sometimes remove the 01-base.yaml file completely, overwrite it with an interface that does not exist (eth0) and refuse to change when i put in ens18, then it spits out an error saying ens18 doesn't exist. It overwrites my previous hpe-vm configuration with this:

hpe@hpe-vme:/etc/netplan$ sudo cat 01-base.yaml

network:
ethernets:
ens18: {}
version: 2
hpe@hpe-vme:/etc/netplan$ sudo cat 50-cloud-init.yaml
# This file is generated from information provided by the datasource. Changes
# to it will not persist across an instance reboot. To disable cloud-init's
# network configuration capabilities, write a file
# /etc/cloud/cloud.cfg.d/99-disable-network-config.cfg with the following:
# network: {config: disabled}
network:
version: 2
ethernets:
eth0:
addresses:
- 192.168.1.254/24
match:
macaddress: 52:54:00:02:ed:af
nameservers:
addresses:
- 1.1.1.1
- 8.8.8.8
routes:
- to: default
via: 192.168.1.240
set-name: eth0

This happens right when i land at the configure part, its a shame i didnt capture the "ip a" part where it removed all interfaces(mgmt, ovs-system etc) and  replaced them with only "lo" and "eth0". eth0 had correct ip which for me is 192.168.1.254, gui was still accessible. The solution for that current problem is manipulating the 50-cloud-init.yaml file and replacing eth0 with ens18, suddenly theres no issue creating the HpeVM cluster. During the initiliaztion of the cluster it disconnects ssh multiple times switching netplan files from 01-base.yaml & 50-cloud-init.yaml to 01-base.yaml & 60-mvm-mgmt.yaml (along with backup folders) a few times. I don't know if there is issues with nesting the hpe-vm with proxmox etc. 

Now what fails is the pacemaker configuration in the post-provision process where the Pacemaker Configuration script fails within 1 second, even after pressing the redo button. 

It seems to deploy vm's fine but its extremely RAM hungry slightly wish there was a more lightweight version as this feels like a vCenter with esxi built in, in terms of ram usage.

[DiegoDelgado]

Hello,

I get your comments and I know it can be confusing and require some better understanding, but as far as I can see, your specific issue might be related to both the following things:

- Most likely, cloud-init is wreaking havoc in the environment as this is something I've seen in the past. I recommend you to disable it completely by running "sudo touch /etc/cloud/cloud-init.disabled", this will disable it completely and that 50-cloud-init.yaml will stop appearing

- Having the interface renamed as eth0 sounds really strange, maybe you left this one in the Install Morpheus modal. Could you please try again making sure ens18 is entered there?

Maybe a fresh Ubuntu 22.04 server install would help if some deeper network configs are there that might interfere or you could run "/opt/hpe-vm/scripts/reset_host.sh" to try and clean it up before trying again with cloud init disabled