Search results

Hi Stefan, well, from host perspective everything is working, but I haven't analyzed in detail the health of all VMs/LXCs as we have some tasks to delivered ASAP on the hosts and then I can dig deeper to the host system itself. CPU: AMD Ryzen Threadripper 2950X - snippet of one core from...

Storage is local, named identically on both hosts to allow cloning/migration between the hosts. TIER0 and TIER2 are ZFS pools and t0-{id} / t2-{id} are directories on it. Problem is I don't see the storage of the other host, so I cannot execute the cloning as no storage on Destination host...

Based on my experience pveproxy is binded to 0.0.0.0:8006 by default. It means it's binded to all IPv4 interfaces available on all interfaces with set IPv4 address. What is Your output of netstat -plntu | grep 8006 command?

Well, that's an interesting Use Case. Generally KVM is complaining about one CPU flag missing on the host OS level presented to it by the installed CPU. CPU flag is either missing or not supported by QEMU version 6.0, which I doubt if you're not running any "ancient" CPU. Can you please share...

Yes, that makes sense. I will later check what CPU instructions are presented to FreeBSD underlying pfSense. AES-NI definitelly impacts performance of OpenVPN. Let's dig it deeper. Sure, it is a must in Nested Virtualization Use Case. Just a little note: in documentation is mentioned that in...

Hosts are in same DC? Share your /etc/apt/sources.list and any Proxmox lists in /etc/apt/sources.list.d/ in order to allow us to investigate RCOTI in detail.

Well, in mentioned doc is HTTP used for pve-no-subscription and HTTPS for pve-enterprise.

I've experienced issues with system hungs of pfSense 2.4.4-RELEASE-p3 VMs on kernels from 5.0.21-3-pve up to 5.0.21-10-pve, so I have deployed 5.3.7-1-pve and it fixed my issue for the moment. When referring to pve-kernel-5.3, there are still there versions of kernel 5.3.x available in the...

Why are you setting CPU Type to host for pfSense VM? Based on my current knowledge, host CPU Type is required (and meant) for Nested Virtualization Use Cases. I haven't tested any "non-virtualizing" VM to run with host CPU Type, but I assume it negatively impacts the instruction set...

Please provide further information related to Your network setup and try to describe Your L3 scenario in order to be capable to help more directly. Proving output of ip a s && ip r l should be a good start. Output of traceroute {destination} from Your Proxmox may give us some hint as well...

Currently I am facing the issue I cannot clone VM from one host to another due to no Target Storage available on destination hosts. Both hosts has same local storage mapped identically on both hosts, but this look likes to more like a missing privilege issue. Here is my /etc/pve/storage.cfg...

Yes, clone means full clone. So all OS related settings is identical on the original and cloned VM. At least IP addresses should be changed if the clone is connected to the same bridge as the original one. If you are cloning the VM from one host to another, you should have same bridge on the...

Usually I do not touch /etc/apt/sources.list when installing Proxmox. I do rather create a new APT list (/etc/apt/sources.list.d/pve-enterprise.list in Your case as You have a subscription) with just one line as described in the docs. deb https://enterprise.proxmox.com/debian/pve buster...

My /var/log/kern.log is flooded by the kernel errors see below. Can anyone help me how should I investigate the RCOTI? I would like to figure out what is causing these traces. Thanks! Nov 25 16:12:34 hetzner-srv1 kernel: [797845.641703] ------------[ cut here ]------------ Nov 25 16:12:34...

If anyone is facing this Connection Error 595 after creating the Proxmox Cluster, this thread may help you to resolve the RCOTI.

Sorry, my bad - You are right that my host behavior is different, so I have opened a new thread here.

Adam, I have followed Your hint, edited /etc/hosts on both hosts with cluster network IPs, did systemctl restart pve-cluster && systemctl restart corosync and the RCOTI has been resolved. /etc/pve/.members contains expected values and networking between the hosts seems to be OK. Many thanks for...

Yes, I chose the cluster network, Link 0 has IPs assigned from that L3 network, but communication between the hosts goes over the WAN interfaces as present in Cluster Join Information.

In /etc/hosts I do have the WAN address from the time before creating the cluster network. I understand I should change the IPs to the cluster network, right? Would it affect the IP address listed in /etc/pve/.members?

Hi, I have created a Proxmox cluster via Web GUI and that was probably my fault as in the Cluster Join Information I do see my WAN IP instead of the one dedicated for cluster LAN. My Link 0 IP addresses were set correct, but my /etc/pve/.members is faulty see: { "nodename": "hetzner-srv2"...