@mhomoky Thanks, worked perfectly fine.
deb [trusted=yes] https://gitlab.com/api/v4/projects/xen-project\%252Fxen-guest-agent/packages/generic/deb-amd64/ release/
No issues in Cloud-config
Also for hostname, I just used {name}without any %
The only thing missing is being able to reapply cloud-init configs or easy reconfigure from the Ui. But that's for another topic.
@olivierlambert also if you use hostname with
hostname: {name}%
The hostname of the vm would end in 0
@olivierlambert said in New Rust Xen guest tools:
Does it work if you replace
%252Fby a/?
I did that, and the / was parsed correctly, but apt didn't update with it.
EDIT:
I just try. Ubuntu 24.04 cloudimage, if you add manually with / instead of encoding it as %252F would give errors.
Like other people the 401 unathorized
Same cloud data file works without issues on other hypervisors.
A few notes about the tools and install through Cloud-init.
In Xen Orchestra for cloud-init user data if you add the debian source:
deb [trusted=yes] https://gitlab.com/api/v4/projects/xen-project%252Fxen-guest-agent/packages/generic/deb-amd64/ release/
There's an error parsing:
%252F
Would parse a:
% = 0
Would result in the following:
deb [trusted=yes] https://gitlab.com/api/v4/projects/xen-project0252Fxen-guest-agent/packages/generic/deb-amd64/ release/
The error would only apear on the parsed data in the VM. Doesn't matter single, double, without quotes, double % would be 00.
#cloud-config
apt:
sources:
xen-guest-agent:
source: "deb [trusted=yes] https://gitlab.com/api/v4/projects/xen-project%252Fxen-guest-agent/packages/generic/deb-amd64/ release/"
packages:
- xen-guest-agent
@olivierlambert will it be fixed all of the IOMMU?
I'm still very running very stable 8.3, but I'm not using right now pcie passthroguh and SR-IOV. But I'm planning to in the next month or so.
Also some issue I've got from the internal 2,5gb nic. Realtek one.
Not having much trouble with the ASUS board so far.
I woule like to move to 2x48GB, 4x32, or 4x48gb. Affordable performance/price ratio.
For me it's working without issues since it was released.
ASUS Prime X670-P , 7900X and 2x32 GB 5600 (QVL).
The realtek nic, gave me some trouble, so I move on from it to a i350-T4 temporary.
The ASUS motherboard was the only one with 3 mechancal 16x ports.
I use them with 3 cards (2xi350-T4 and 1 x520-DA2).
I left a VM for a week with 20 cores doing ML and seems to be stable.
I would prefer a more enterprise solution but I had this HW for another purpose and did some testing.,
I chose the AM5 route so I don't spend money on a dedicate x1 GPU in a headless computer. And had to deal with many issues in the BIOS settings to get things properly running.
Today I tested a Corsair 6200 MT 32-38-38 Memory, lot's of issues @ 6.2 (just XCM profile Auto). Lot's of issues, @ 6000 MT seems to be working fine.
@olivierlambert I just cloned the drive, the issue is the same. I was thinking of cleaning the DB or trying to recover that instance.
I think I also saw on the screen the menu flashing with some colours.
@olivierlambert that was my thought. I cloned live to a nother SSD, and that SSD reinstall the 8.3 on top of it and updated. But same result. I wanted also to reset the BM but keep the SR with the disks (thin) LVM.
This is a test, for a DR. The VM are not important. I can clone that SSD (nvme) and leave it for testing on the same HW.
If you want to test something, be my guest.
I've got some logs:
xapi: [error||0 |starting up database engine D:93df1d58153e|backtrace] dbsync (update_env) R:ff17b7372476 failed with exception Stdlib.Scanf.Scan_failure("scanf: bad input at char number 6: looking for ':', found '.'")
Also the xapi keep restarting, and:
Cannot add dependency job for unit lvm2-activation-early.service, ignoring: Unit is masked
Cannont add deppendency job for unit lvm2-activation.service, ignoring: Unit is masked
This is a test on this machine. I'll clone the disk if you need further info of how to recover.
Today I installed a i350-T4 for testing and not depending on the realtek 2.5g card.
As I guess the computer didn't boot after installing the card.
DDR5 5600@C40 max supported on the QVL. Had to clear cmos and reconfigure again the bios so the computer can post again. Set up the L2APIC again, and SR-IOV, IOMMU.
Got access to the server through SSH, but the server can't start any VM and have issues with the API not working locally. (XOA LITE) not working neither VMS.
With xe vm-list sometimes list the vms, even with xe-toolstack-restart.
The error: Error : Connection Refused (calling connect)
On the display in the vm list : ("'NoneType' object has no attribute 'xenapi' ", )
Also in the xensource.log some erros on the string.
I will dump the logs and copy some over here.
@tmservo433 I'm testing a 7900X with Asus Prime X670-P with the latest patches on 8.3. When the patch was released I left a baisc configuration only deploying a W10 test machine and cloned snapshots to test. Was great.
Today I was setting the W10 machine and started to go very slow and painful. I just remember your post. DId a reboot, and started working fast again.
I'll be testing and installing a 10G nic on that board. The Realtek nic was working at 2.5Gb (I need to fire some iperfs to test). Connected to a 5GbE Switch.
I have a few machines with this combo with multiple NICs to use with QEMU on Ubuntu.
Not liking the ASUS expirience, I would love to have an AM5 board with 4 PCie 4X mechanical or with bifurcation.
I also tested unsucessfully to passtrough the NVME M.2 PCiev4 Driver with W11 installed. ( I managed to boot another system with Proxmox) but with older bios and many inestabilites.
@stormi today I installed fresh on asus motherboard:
7900X 2x32GB 5600 CL40 , X670-P-Wifi bios 1616 from May 16 2023.
Without forcing X2APIC on the bios, the 8.3 installer gave KP.
Installed to an nvme SSD and rebooted without issues. Added the repo and updated the system.
Install XOA through https://xen-orchestra.com/#!/xoa And made a simple Windows 10 Client machine, with 2 cores, the installl went smooth.
So good that I upgrade the VM with 24 cores just for testing and got a very nice surprise:
Almost the same performance I've got through Windows 11 baremetal!
Note this CPU only gets about 5.04 Ghz in one CCD and 4.85 Ghz in the other. When fully loaded. Not the best CPU I've got and runs 95ºC (normal for this CPU).
HTOP only shows 16 threads when X2apic enabled.
I need to test the integrated 2.5Gb NIC and the PCIe passtrough:
--> NVME pcie 4.0 directly to a VM (W11)
--> i350-T4 SR-IOV testing
---> i350-T4 directly to a VM.
---> X710-DA2 / X520-DA2
@stormi Good news! if you need some testing to be done, I've a 7900X avaliable for testing.
@olivierlambert said in Non-server CPU compatibility - Ryzen and Intel:
It's possible we got a fix, but we need sometime to generate the right build/RPM for you to test
I have to identical setups for testing. With all the storage being pcie and interrupts. I also got a instlall with proxmox, but kind of same issues with IOMMU, trying to passtrough the 4.0 nvme. The disk was very slow, CPU wise performance on W11 machine was great.
I'm testing this combo:
With setting: Local APIC Mode = X2APIC and UEFI set to Other OS. Installed 8.3 alpha and updated, got errors. Test installing XOA took too much time and booting was painfully slow using only 1 SSD on chipset NVME.
Tried disabling IOMMU, but the same issue.
