Recent content by weehooey-bh

When moving from standalone nodes to a cluster, the network you have to be most concerned about is the cluster network. The cluster network carries the Corosync traffic, and best practice is to have a dedicated 1Gbps network for just the cluster network (i.e. Corosync traffic). Note, a VLAN on a...

Some suggestions: You can use bmon on your PVE host to determine which interface is handling the traffic. When you run iperf3, use port numbers other than the default. Disconnect one of the links and test again. Do you get the speeds you are expecting with just one link? Based on what you are...

Do you mean mapping VMID to the VNC port number somehow? Or is there another method?

@bytegenius You can file a feature request here: https://bugzilla.proxmox.com/ Once you do that, please post the link back here so people can follow it. @lost_avocado where did you see this? Can you post a link to it? How do you get the port number? There is no automatic fix. You need to...

Re-review your configuration. Please post your /etc/network/interfaces file. Confirm that the packets are flowing as you expect them to. A packet capture may provide some insights. What is your definition of "moderate" traffic? Is this the problem you are trying to solve? host-vm network...

Corosync can run over a bond, but it is not recommended. Smaller clusters and blade servers regularly use bonded interfaces out of necessity. Having multiple IP addresses per host is a standard deployment for a PVE cluster. In fact, it is recommended. Each machine will only have one hostname...

Hi All, Proxmox has fixed this issue: https://bugzilla.proxmox.com/show_bug.cgi?id=5415 Thank you to @cheiss !

@bravo0916 Not sure. What I can tell you is if you are going to find some information about capability, Ubuntu is the place you will find it.

Hi @bravo0916 Sorry, I gave you an incomplete answer. PVE is based on Debian with a Ubuntu kernel. There isn’t a Debian HCL but there is some compatibility information available from Ubuntu. Others have asked similar questions: https://forum.proxmox.com/threads/compatibility-matrix.168096/...

Proxmox does not publish hardware compatibility lists (HCL). Under the hood, Proxmox products are Debian. If the NIC will work with Debian, it will work with Proxmox products. We have found that Intel is a very reliable network interface card manufacturer.

Right now, restores are not included in the built-in notifications. As a starting point, you could monitor journalctl for restore entries and act upon that event (i.e. send an email). /usr/bin/journalctl -f -n 0 | /usr/bin/grep --line-buffered "vzrestore" Hope this helps.

Carefully read and follow: Corosync Redundancy

Yes, you can configure up to eight (8) separate Corosync links. Just make the primary link (link 0) entirely separate from all other traffic.

There are different types of "cluster" traffic. By default, it all goes over the host network. Corosync is not migration traffic; the Corosync traffic should be moved to its own network. You can leave the migration traffic on the host network to take advantage of the 10G link. Setup your...

No worries. Perfect! Corosync is what makes a cluster work. You will be unhappy if your Corosync is unhappy. :) Cluster Manager