@samuelolavo
Yes, I do use NUT, but I don’t want to install any additional packet on XCP-ng. The way I found leverage the SSH authorized keys restrictions, and so you only need to change one file.
Thanks for the link anyway

From the xe help host-shutdown and xe help host-disable
"If no parameters to select hosts are given, the operation will be performed on all hosts."

I have a single host pool, so I can't test.

Oh I see, sorry for the confusion... I thought you were referring to xe.
And you're right the following is working like a charm:

command="xe host-disable; xe host-shutdown",no-agent-forwarding,no-port-forwarding,no-pty,no-X11-forwarding,no-user-rc,restrict a_dedicated_pub_key_here

Note: I intentionally omitted using any host selector. Therefore, if I’ve understood the documentation correctly, host-disable and host-shutdown will apply to all the hosts in the pool.

Sorry to revive this topic... I was just wondering: which force flag are you using?
Per the docs I couldn't find it for xe host-shutdown command.

FYI (in case it could help), my idea is to implement UPS shutdown order via SSH by using force command in /root/.ssh/authorized_keys.

command="the_force_shutdown_command_here",no-agent-forwarding,no-port-forwarding,no-pty,no-user-rc,no-X11-forwarding a_dedicated_pub_key_here

Then if you ssh in, it will force the command command= (other command will be ignored and you can't open interactive shell).

ssh -i /path/to/dedicated_private_key user@xcp-ng-host whatever_command

(whatever_command is just there to prevent error message)

@KPS Glad to hear

@KPS said in Linux VM (Ubuntu 22.04) - Grub-Menu invisible:

GRUB_TIMEOUT_STYLE

Set it to menu

If ‘GRUB_TIMEOUT_STYLE’ is set to ‘countdown’ or ‘hidden’, the timeout is instead counted before the menu is displayed.
Source: https://www.gnu.org/software/grub/manual/grub/html_node/Simple-configuration.html

I'm not sure to why you are setting up serial connection?

@KPS said in Linux VM (Ubuntu 22.04) - Grub-Menu invisible:

GRUB_TERMINAL="serial"
GRUB_CMDLINE_LINUX="console=tty0 console=ttyS0,115200"
GRUB_SERIAL_COMMAND="serial --speed=115200 --unit=0 --word=8 --parity=no --stop=1"

I'm pretty new on XCP-ng... but I'm pretty sure the "console" of a VM is not a "serial console".

FYI, this is my /etc/default/grub

GRUB_DEFAULT=0
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT="quiet"
GRUB_CMDLINE_LINUX=""

2 ideas:

• What is the value of GRUB_TIMEOUT_STYLE?
• Do you have a GPU connected? it's a shot in the dark but I got this kind of issue because of GPU drivers... but on bare-metal machines

Hi,
Have you run sudo update-grub after changing your configuration?

@olivierlambert
Happy to report that it worked perfectly. VM booted without an issue.

@AtaxyaNetwork
Thanks for such quick compile.

No problem I’ll update when it’s done and working.

@AtaxyaNetwork seems to work fine. I can mount my VMFS6 and the conversion is in progress.
Do you need specific test?
Note: did it on a fresh install of XCP-ng and after a yum update.

@AtaxyaNetwork Sure I can test and I'll report back if it was successful.
Thanks a lot.

You're right. It's certainly easier that way

What about using XOA VM as it is Debian-based and the packages are already part of Debian repo.
Here is an idea:

• Install XCP-ng on a separate drive (like in the procedure).
• Deploy or migrate XOA on the host.
• Expose the VMFS devices to XOA (as passthrough) with something like:

mkdir /srv/vmfs 
ln -s /dev/disk/by-id/<DISK_ID> /srv/vmfs`
xe sr-create name-label=VMFS type=udev content-type=disk device-config:location= /srv/vmfs host-uuid=<HOST UUID>`

• Attach the new SR to XOA.
• Mount the VMFS within XOA VM.
• Use XOA to convert and send the vhd to an SR via the API.

The part of sending the VHD to an SR is what XO does already... but I'm really not familiar with how it work internally.

This can be done manually or could (if someone want to code it) be integrated in XOA web interface to provide super fast VMware conversion.

Hi there,
First post in XCP-ng community and it's a question... and my workaround.
Migrating huge VMDK from ESXi 7 is a pain... the V2V is great but it seems that the VMkernel is capping my 10Gbps SFP to 200Mbps.
So I wanted to give a shot to the local migration procedures: https://docs.xcp-ng.org/installation/migrate-to-xcp-ng/#local-migration-same-host

Buuuut I got stuck on

wget https://forensics.cert.org/centos/cert/7/x86_64/vmfs6-tools-0.2.1-1.el7.x86_64.rpm

404 error... so I tried to find an equivalent rpm... but no luck (note I'm not familiar with CentOS package management)! So here is the question: do somebody know where to find it?

Now I was not going to be stopped... so I booted a LinuxMint (Debian Edition) Live CD.... a GUI is not needed but nice in case you want to browse the web.

sudo apt update
sudo apt install qemu-utils vmfs6-tools

Then I mounted the LVM XCP-ng SR and followed the rest of the procedure for converting.