Posts made by Schmarvin

For starters, sorry if this is in the wrong area.

I searched the forums and found moving from local to shared storage might give problems.
I'm currently looking to move 3 VDI/Disks from a shared storage iSCSI to local SSD. I have the folllwing:
1 Host

• VM running with 16 CPUs, 32GB RAM, and 3 disks listed below:
  -->> 64GB Disk for OS, 1TB Disk for Database, 24GB Disk for Swap/Limp
• Running Rocky Linux 9

I have 4TB available on the local storage I'm trying to move to. But everytime I try, it says:

HOST_NOT_ENOUGH_FREE_MEMORY(34638659584, 9544155136)
This is a XenServer/XCP-ng error

I'm not trying to move the host, just the disks under it. Its literally moving just the shared storage disks to local. Maybe I missed a step?

@olivierlambert Thank you.

Ran into an odd issue on one of the test benches. I get the following error log when trying to update:

+ command -v git
/usr/bin/git
+ cd . && git config --get remote.origin.url
https://github.com/ronivay/XenOrchestraInstallerUpdater.git
+ cd . && git status --porcelain
?? node_modules/
?? yarn.lock
Running script version a930e94 with config:

PORT="80"
INSTALLDIR="/opt/xo"
SELFUPGRADE=true
CONFIGUPDATE=true
REPOSITORY="https://github.com/vatesfr/xen-orchestra"
BRANCH="master"
PLUGINS="all"
AUTOUPDATE="true"
OS_CHECK="true"
ARCH_CHECK="true"
PRESERVE="3"

+ id -u
0
+ uname -m
x86_64
+ command -v xe
+ grep ^VERSION_ID /etc/os-release | cut -d'=' -f2 | grep -Eo '[0-9]{1,2}' | head -1
10
+ grep ^NAME /etc/os-release | cut -d'=' -f2 | sed 's/"//g' | awk '{print $1}'
Debian
+ grep ^GRUB_DISTRIBUTOR /etc/default/grub | grep 'Xen Orchestra'
+ command -v yum
+ command -v apt-get
/usr/bin/apt-get
+ command -v systemctl
/usr/bin/systemctl
+ df -P -k '/opt' | tail -1 | awk '{print $4}'
25538416
+ grep MemTotal /proc/meminfo | awk '{print $2}'
4020612
+ node -v 2>/dev/null| grep -Eo '[0-9.]+' | cut -d'.' -f1
18
+ apt-get update
Hit:1 http://deb.debian.org/debian buster InRelease
Get:2 http://security.debian.org/debian-security buster/updates InRelease [34.8 kB]
Hit:3 https://dl.yarnpkg.com/debian stable InRelease
Hit:4 https://deb.nodesource.com/node_18.x nodistro InRelease
Get:5 http://security.debian.org/debian-security buster/updates/main Sources [375 kB]
Get:6 http://security.debian.org/debian-security buster/updates/main amd64 Packages [592 kB]
Get:7 http://security.debian.org/debian-security buster/updates/main Translation-en [321 kB]
Fetched 1,323 kB in 2s (795 kB/s)
Reading package lists...
+ apt-get install -y --only-upgrade nodejs yarn
Reading package lists...
Building dependency tree...
Reading state information...
nodejs is already the newest version (18.20.1-1nodesource1).
yarn is already the newest version (1.22.19-1).
The following package was automatically installed and is no longer required:
  linux-image-4.19.0-17-amd64
Use 'apt autoremove' to remove it.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
+ cd "/opt/xo/xo-src/xen-orchestra" && git remote set-url origin "https://github.com/vatesfr/xen-orchestra" &&             git fetch --prune &&             git reset --hard origin/$
HEAD is now at 74fc4dcac refactor(xo-server/importFromEsxi): don't duplicate code
+ rm -rf "/opt/xo/xo-builds/xen-orchestra-202404061141"
+ rm -rf "/opt/xo/xo-builds/xen-orchestra-202404061141"
+ cp -r "/opt/xo/xo-src/xen-orchestra" "/opt/xo/xo-builds/xen-orchestra-202404061141"
+ cd /opt/xo/xo-builds/xen-orchestra-202404061141 && git rev-parse HEAD
74fc4dcacaa388d17329b2153ad2c09a915ba7d9
+ cd /opt/xo/xo-builds/xen-orchestra-202404061141 && git rev-parse --short HEAD
74fc4dcac
+ cd .
+ readlink -e /opt/xo/xo-server
/opt/xo/xo-builds/xen-orchestra-202401032158/packages/xo-server
+ cd /opt/xo/xo-server && git rev-parse HEAD
8b0b2d7c31392fab71f3e4715d188281451dd74f
+ cd /opt/xo/xo-server && git rev-parse --short HEAD
8b0b2d7c3
+ cd .
Updating Xen Orchestra from '8b0b2d7c3' to '74fc4dcac'
+ cd /opt/xo/xo-builds/xen-orchestra-202404061141 && yarn --network-timeout 300000 && yarn --network-timeout 300000 build
yarn install v1.22.19
[1/5] Validating package.json...
[2/5] Resolving packages...
[3/5] Fetching packages...
info There appears to be trouble with your network connection. Retrying...
info There appears to be trouble with your network connection. Retrying...
info There appears to be trouble with your network connection. Retrying...
info There appears to be trouble with your network connection. Retrying...
error An unexpected error occurred: "https://registry.yarnpkg.com/@babel/parser/-/parser-7.24.0.tgz: connect ETIMEDOUT 2606:4700::6810:1922:443".
info If you think this is a bug, please open a bug report with the information provided in "/opt/xo/xo-builds/xen-orchestra-202404061141/yarn-error.log".
info Visit https://yarnpkg.com/en/docs/cli/install for documentation about this command.
+ rm -rf /opt/xo/xo-builds/xen-orchestra-202404061141

I tried the following from this topic as well, no success: https://xcp-ng.org/forum/post/38138

XO is in a Debian 10 VM on XCP-NG running version 8.2
Debian is up-to-date at this point as well.

@olivierlambert

Is there a way to add another permission group? So that they could have select permissions?

@DustinB

Under users, I only have "User" and "Admin" for permissions.
Setting as user, they can't change the password or setup OTP.

@DustinB @olivierlambert

Awesome, thanks for the assistance. It works! Though, noted an issue. Might be user error, meaning me. But, when setting the user to "User" instead of "Admin" they can't change their password. Says "not enough permissions"

Is there a way to fix it where they can still change their password?

Just an off the wall question to see if someone has something similar or some similar ideas.

Would it be possible to have Xen Orchestra public facing, and setup to give lets say, customers, permissions only to their VMs/servers?

Looking to see if this would be a good setup for host cloud hosting providers have this setup, but in a small capacity. Less than 20 users, maybe 40 VMs total.

@Danp Ninja Backups, its similar to Comet Backups I believe. Does full image backups via WindowsPE.

@Danp Again, my apologies, Trying to figure out how to make it easy.

• When starting the backup software ISO, like you would start from an ISO to load an OS, it only detects the first Hard Disk attached to the VM.
• In Xen Orchestra, during the VM creation, I added all 5 drives with their storage sizes. These are brand new hard disks created from within Xen Orchestra.
• All 5 hard drives appears under the Disk's tab in XO.

As a test, I also tried this on a different newly created VM and attempted to install Windows 10 with two hard drives (created with XO), and same issue. It only shows 1 drive.

I know I had a similar issue a long time ago with VMware where I had to change the hard drive type from iSCSI to SATA or IDE. Is that possible for XCP-NG/XO or not?

@Danp Can't install guest tools on a VM that has no operating system. Backup software copies the image from the source (the old server) and lets us download to the new target (new VM).

Since we are coming from a server that is on-site at a customer to a VM hosted in our servers, we can't just set the VM as a new server and config and move it over.

@Danp Added during the creation of the VM.

The setup doesn't need raid, the old server/setup had raid, but I don't care about that. There are 5 Disks/Volumes that I need to carry from that server to the new VM.

@Danp

My apologies.
New VM from scratch. Running Server 2016.
No, we have a backup software that does that.

**as an FYI, we did this all the time with VMware. Trying to re-create the procedure with XCP-NG.

Hello,

Just thought I'd ask, as this was the first time I've encountered this issue and couldn't find the answer from a few hours of my Google-Fu.

Migrating a customer server into a VM. Current server has multple hard drives, not partitions. Mind you, these drives are set from the RAID controller, can't change these. I believe there are 5 total "drives" and not partitions.

Starting up the new VM in Xen Orchestra and noticed only 1 of the 5 drives added are showing up. This is a Windows Server machine. Is there a reason this is happening? Or a work around or solution to this?

As a last FYI, the number of these "drives" can't be reduced or put into partitions. This server runs a huge medical program for a customer of ours and it can't be reconfigured properly.

@DustinB No, not really. I just like hearing peoples opinions from a personal perspective and not corporate lingo.

@olivierlambert I know it is all based on physical hosts and hardware. I just like hearing stories of how massive these networks are on XCP-NG.

@Andrew That's awesome. I'm just impressed. Moving from VMware/Vsphere, not ESXI, to XCP-NG w/ XOA, everything just works and is lightweight. The issues with VMware is the resources needed. Just the base system requires 4-8GB of RAM, each add-on, such as vsphere is an additional 6-8GB of RAM. And yes, the UI is great, but sometimes, it sticks and doesn't work.

@Andrew Yea, I know it depends on all that. I'm just curious on the most crazy environments. We don't have that many VMs, I don't think. But I'm curious what others have, no necessarily what the hardware running is...just the number of Virtual Machines in the Pool and how many servers run those many VMs.

Hello all,

Question for anyone who uses XCP-NG:

Whats the most VMs you have running?

We have a 3 server pool of XCP-NG managed by XOA. Migrated from VMware due to the Broadcom stuff going on. Everything is stable. Running 22 VMs at the moment, about to add 32 more.