Posts made by laurentm

@john-c I know already that, bios was updated and it was easy to boot in UEFI but the install failed and even complains about not enough space on EFI boot part.
I tried after wiping all disks but could not perform the install all the way.
Either with SATA SSDs or NVMe (with an upgraded DELL backplane designed for NVMe) it never works.

@Ajmind-0 Same problem with DELL PowerEdge R820 and DELL PowerEdge R520:

I had to upgrade in BIOS mode and it was the same thing for a fresh install.

With DELL PowerEdge R240, no problem to upgfrade in UEFI.

@AbyssMoon
My release vNext 24158 still works with XCP-NG 8.3 Final but I cannot use it anymore for Live Migration and Cold Migration due to a cert error between servers.
It was working with XCP-NG 8.3 RC2.

Three standalone servers (no pool) running XCP-NG 8.3 in the same LAN: everything ok

After upgrade to 8.3 Final (on DELL R820 & R520 I had to use bios method instead of UEFI)
all VMs run smoothly but cannot use Live Migration anymore (that was possible on 8.3 Beta).

I tried to de-activate TLS on every host without success.

Finally, Live Migration works again via XOA after updating XOA to the latest.

XCP-NG Center still works with 8.3 final but cannot do LIve Migration now.
I know that it is deprecated but it allows to see easily more than one item in the same screen.

@olivierlambert
Thanks for your reply.

Since this is just a spare server, I decide to perform a new fresh install.
Previous install of Beta was easy (I think that it was via Bios) but this final release causes a lot of trouble:

• in UEFI no success, even after wiping all the SSDs with Gparted-live-cd (because XenServer and XCP-NG have sometimes trouble to remove flags from ZFS, Raid, LVM or others, according to olders posts.

• in Bios mode, install works but one NVMe of two is not detected (the one seated on a U2-pcie adapter)

Now I'm facing a hardware problem known of these DELL Rx20 (server halted at boot) with a pcie adapter. So I'm trying to shutdown for a while the machine via the IDRAC and try a cold boot.
If not working I have to reseat PCI riser as it is advised in many forums.
This prevent me to upgrade via iso two productions servers (DELL R520 and R820) before having solved the current issue on the backup machine.

Re: XCP-ng 8.3 betas and RCs feedback


My upgrade to 8.3 final via iso looks fine at first glance. But after startup I was able to start up only one VM, all the other have no more storage. The local storage was gone away !

So I decided to build a new setup from scratch but either my SATA system SSD or a NVMe SSD on the next attempt leads to this error: no space left on device.

I never encounted this error before in either XenServer or XCP-NG from the last fifteen years.

About my hardware: it is a quite unusual setup, this is an old quad cpu DELL R820 renewed with a NVMe backplane and two U2 Kingston Enterprise SSDs (gen 4 SSDs running on gen3 pcie slots).
I keep it just for remote backups (using XenOrchestra from sources) since power consumption is too expensive at the datacenter energy rates. This server is able to boot in UEFI or Bios mode.

@olivierlambert

Thanks for your answer.

The CPU are absolutely identical (no rev mismatch). The two servers are stand alone, not in a pool, same 10 Gb LAN.

I'm able to migrate if I shutdown the VM or if I export it.

The only difference between servers is the bios revision and the uptime. The xcp-ng updates have all been installed but maybe a reboot is missing on one server in order to apply them properly.

I'm planning to live migrate the most important VMs from the server with the older bios to the one with the newer bios (because it works in this way) and shutdown all the remaining VMs to update the newer bios. I supposed that DELL has changed quite a lot of things in their new release but this important notice was already present in previous bios releases.

Disabling the CPU Physical Address Limit option from the BIOS setup menu enables 5-level paging. However, enabling the CPU Physical Address Limit option may cause the operating system to stop at the driver verifier DMA violation test with a blue screen error when booting from an operating system that does not support 5-level paging, such as Microsoft Windows Server 2019, Microsoft Windows Server 2016 and so on.

I do not know how XCP-NG deals with 5-level paging

I have two DELL R650XS similar (bi-XEON 4309Y) and I'm no more able to migrate VMs between those servers.

XCP-NG release is 8.2 LTS with the latest updates. I use only local storage (because I have U2 NVMe SSDs)

Attempts to migrate from XOA or XCP-NG Center display the same warning: some CPU features are missing on this server.

In the opposite way from server 2 to server 1, it works. Server 1 has 46 days uptime and Server 2 has 180 days uptime.

Maybe I shoud reboot both hosts but it's annoying for users to shutdown 100+ VMs !

@stormi

Thanks for your answer.

I'll give a try later, bypassing the PFBlocker module of the router.

Github is not currently on any blacklist, but I block Digital Ocean and some others who host many attackers.

@stormi

https://xcp-ng.org/forum/topic/8527/fatal-error-while-updating-xoa-installed-from-source/7

@HeMaN

The XOA running on 8.3 was a clone of the XOA running on 8.2 and has been fine for more than four months. It still works, just the last update cannot be performed.
The VM OS is the same Ubuntu 22.04 LTS with the latest updates.

I cannot right now update the 8.2 because it is used by 100+ VMs.

@Danp I have only one XOA running on XCP-NG 8.3, the other one is on 8.2 and had no issues. I do not know if others 8.3 users have success or not.
The previous updates on 8.3 had no problems.

Both XOA VMs are build from source on Ubuntu server 22.04 LTS with 16 GB ram
and 4 vCPU.

Hardware:

• DELL Poweredge R820 quad cpu Xeon E4610 for XCP-NG 8.3
• DELL Poweredge R650 two cpu Xeon 4309Y for XCP-NG 8.2.

@Danp

cd /root/XenOrchestraInstallerUpdater/

 ./xo-install.sh

Welcome to automated Xen Orchestra install

Following options will be used for installation:

OS: Ubuntu 22
Basedir: /opt/xo
User: root
Port: 80
HTTPS: false
Git Branch for source: master
Following plugins will be installed: all
Number of previous installations to preserve: 3
Node.js and yarn auto update: true

Errorlog is stored to /root/XenOrchestraInstallerUpdater/logs/xo-install.log-202403021931 for debug purposes

Depending on which installation is chosen:

Xen Orchestra configuration will be stored to /root/.config/xo-server/config.toml, if 2a74b636-cb2b-e63d-f4da-76aa3789d6d5you don't want it to be replaced with every update, set CONFIGUPDATE to false in xo-install.cfg
Xen Orchestra Proxy configuration will be stored to /root/.config/xo-proxy/config.toml. Config won't be overwritten during update, ever
-----------------------------------------

1. Install
2. Update
3. Rollback
4. Install proxy
5. Update proxy
6. Exit




**choice 2 is selected**


: 2
[info] Installing nodesource repository

[info] Checking current node.js version

[ok] node.js version already on 18, checking updates

[info] Fetching Xen Orchestra source code


[fail] Something went wrong, exiting. Check /root/XenOrchestraInstallerUpdater/logs/xo-install.log-202403021931 for more details and use rollback feature if needed

XenOrchestra (from source) update failed with the latest updates of XCP-NG 8.3.
I opened a topic in XOA forum, but I succeeded to update in XCP-NG 8.2, so the
error is coming from 8.3.

fatal: unable to access 'https://github.com/vatesfr/xen-orchestra/': server certificate verification failed. CAfile: none CRLfile: none

Finally, I have bypassed the PFBlocker module during the update and no more error happened.

I'm running XCP-NG 8.3 Beta (with last updates availables) and when I try to update XOA (installed from source), I have this error:

fatal: unable to access 'https://github.com/vatesfr/xen-orchestra/': server certificate verification failed. CAfile: none CRLfile: none

EDIT: same update with XCP-NG 8.2 LTS works perfectly

@borzel

Thanks for your valuable work.

I did not found any information about how Citrix users will manage XenServer after the end of XenCenter.
I doubt they recommend XOlite
Do they plan to rely only on XSconsole or CLI ?

I'm still using XOA and XCP-NG Center together because for basic tasks (like start/stop a VM) XCP-NG Center is faster and simple.

XOA is fantastic for advanced use and backups, and does not need a windows machine but you have no overall view at a glance.

@Danp Yes I did that and it looks good now.

@davemcl said in NVMe storage wrong detection:

List the LBA format of drive

Thanks, I'm new to NVMe storage in XCP-NG and did not even know the nvme-cli (which is not installed by default in CentOS and XCP-NG). It looks like I'm already using 512b sector size:

/dev/nvme0n1 50026B72830CEBE9 KINGSTON SEDC1500M7680G 1 7.68 TB / 7.68 TB 512 B + 0 B S67F0103
/dev/nvme1n1 50026B72830167A0 KINGSTON SEDC1500M3840G 1 3.84 TB / 3.84 TB 512 B + 0 B S67F0103

In the last 15 years, with my previous DELL servers, the RAID controller was doing everything and volumes were perfectly seen by XenServer and XCP-NG. When I bought a full NVMe RS650XS, I did not ordered any RAID controller and only one DELL U2 960 GB to save money because DELL U2 prices are crazy. Instead I bought 4 really cheap KINGSTON DC1500M, which are less expensive than their SATA counterpart (DC600M). Unfortunately, the LVM choice at install leads me to crash half of the VMs after one month !

Now, I'm using EXT and many separate disks, this is far for ideal because I waste room for more VMs but it seems reliable
and these large U2 SSDs are really cheap, while the slower SAS SSDs are still pricey for an unknown reason.

Re: XCP-ng 8.3 beta

I have a DELL R820 retrofitted with a U2 NVMe backplane (with one single U2 SSD for now) and an add-on cheap PCIe U2 adapter (with one other U2 SSD). I did that for testing the perfs, which are twice better than a SATA SSD but the same for the backplane and the add-on card, restricted to gen2 lanes.

I installed XCP-NG 8.3beta1 via IDRAC on a small SATA SSD (in a false cdrom drawer adapter) and, during installation, both U2 NVMe were properly detected and choosen as EXT storage for VMs.

In XOA, only one storage (7,6 TB) appears with this combined path
/dev/disk/by-id/nvme-KINGSTON_SEDC1500M7680G_50026B72830CEBE9,/dev/disk/by-id/nvme-KINGSTON_SEDC1500M3840G_50026B72830167A0
but the 3,8 TB disk is not showing its capacity.

So, I had to forget this storage and recreate one after one two EXT storage.

I know that my use case is outside an average setup but I do not understand why XCP-NG is aware of boths disks
but only takes in consideration the larger one.

Hi, I would like to get an advice to solve an annoying migration.
The VM has a 700 GB system disk on local storage and a 24 TB i-scsi storage (with 12 x 2 TB virtual disks).

I want to move both VM and datas to a newer App server and a newer NAS storage. Everything is connected via 10 GB Nics.

All my attempts of live migration have failed after 3h30mn with an error message (insufficient temporary space on source).
I found on an old Citrix board that 200% free on source and 300% free on destination are required to perform successfully a live migration.

So, the remaining option is cold migration. The downtime could be 3 or 4 days because the Synology rackstation is not that fast on contents with small files. This appears unacceptable to the customer.

Can I use a warm migration instead ? or does it requires a lot of temporary space ?

Regards,

Laurent