Search results

NUMA was enabled in the BIOS, and node went from showing single numa node to showing 4. VirtIO issues disappeared. Went from 30Gbps vm to vm to 93Gbps. You could also use multiqueue for the virtio nic, but even then, without numa, it was slow. After numa, it was fast.

Tom, The issue was already identified earlier by LnxBil pointing me in the right direction, NUMA was disabled in BIOS. Simple fix, nothing exotic. Your comment about “where I got the hardware” or “who set up the BIOS” doesn’t really contribute anything at this point, especially after the...

Hello, I appreciate your response LnxBil! And it just so happen that this resolved it too. Prior to BIOS changes, we were only seeing a single numa node. After some changes to the BIOS, we are now seeing 4 numa nodes and performance issue is gone. Thank you so much! Cheers, rjoensen

Hello, Anyone noticed slow network performance on AMD EPYC CPUs? I have now confirmed this on two generations. - AMD EPYC 9454P (single socket) - AMD EPYC 7502 (dual socket) Network configuration: - vmbr bridge dedicated for private VM traffic VMs config: - CPU type: host - 10 cores...

Hello, I am in the process of upgrading my cluster, both from 7.4 to 8.2, but also to newer hardware. I have added 2 new nodes to the cluster that previously was 4 nodes, so I have a total of 6 nodes right now. All nodes are participating in a Ceph cluster as well. Ceph wise, I have outed the...

Is that not the same principle? You will end up with a node that is 8.3 and the others are 7.4?

Hello, I am currently running a cluster of 4 nodes that are running 7.4-17 I acquired a new node, that I wanted to add to the cluster, and then gradually upgrade the other 4 nodes to 8.3. Has anyone done this? Any issues? The current cluster is also running ceph.

Hello, I disabled ipv6 in sysctl, restarted corosync on all nodes, and then did a pve-cluster restart on node3 and they all came back into a cluster. Continuing to monitor. This issue randomly happened after 6 months of uptime and no issues whatsoever. Thanks, Ragnar

Hello 1st node auto lo iface lo inet loopback iface eno1 inet manual #PUBLIC iface eno2 inet manual #PRIVATE auto vmbr0 iface vmbr0 inet static address <snipped> netmask 29 gateway <snipped> bridge-ports eno1 bridge-stp off bridge-fd 0 #PUBLIC auto vmbr10 iface...

Hello, 2nd node: auto lo iface lo inet loopback iface eno1 inet manual #PUBLIC iface eno2 inet manual #PRIVATE auto vmbr0 iface vmbr0 inet static address <snipped> netmask 29 gateway <snipped> bridge-ports eno1 bridge-stp off bridge-fd 0 #PUBLIC auto vmbr10 iface...

Hello, third node; auto lo iface lo inet loopback iface eno1 inet manual #PUBLIC iface eno2 inet manual #PRIVATE auto vmbr0 iface vmbr0 inet static address <snipped> netmask 29 gateway <snipped> bridge-ports eno1 bridge-stp off bridge-fd 0 #PUBLIC auto vmbr10 iface...

Hello, I have got 3x proxmox nodes in a cluster. This cluster worked fine up until some 4 hours ago, when the bond between them broke. I have pinned the issue down to the 3rd node that is causing havoc on the corosync level. If I stop corosync, the two other nodes join together just...

Hello, I have had absolutely no problems with zfs replication between my two node cluster for the past 5 months. This morning when I logged into the cluster (usually hv01) and just check basic things (vm usage and such) and to check what kind of VMs I was doing replication for and backups for...

Hello, In previous verious of Proxmox you could run /etc/init.d/pvenetcommit start and you wouldnt need to reboot your system after making changes to the network stack on a proxmox server. I see this script is now gone in newer versions. Is there an equivalent feature that can be used...