I solved this for my cluster. The steps were thus:
Make sure all your monitor keyring files are identical (all mon use the same key!) and of this structure (use your real client-admin key)
[mon.]
key = <somekey>
caps mon =...
What we do is simply edit ceph.conf once the first MON has been provisioned to move both Public and Cluster networks from /32 to /24 (sed -i 's/\/32$/\/24/g' /etc/pve/ceph.conf), then you can keep provisioning Ceph as usual.
Edit: forgot to...
I know that this is for a slight different model network adapter, however this may be useful:
https://forum.proxmox.com/threads/proxmox-ve-9-2-21-bcm57412-bnxt_en-network-failure-after-upgrade-to-kernel-7-0-14-20-pve.186749/
In that thread, the...
One more GX10 post, the last one for now. After battling the black screen, the hard resets and the power cap, we were finally able to see what this box and PVE on arm64 can do.
Out of the box, the kernel initiator managed NVMe/TCP reads at 6.3...
1. I jump because using a template works. I might not want to use templates as it kinda locks me to specific nodes or forces me to share all templates on all nodes.... The connection is that the official guide show a step by step to a TEMPLATE...
If you habe 4 VM with 8 GB each, then the or my calculation is 2 gb Proxmox VE Host, 2 gb some other Linux Cache, 2-16 gb zfs arc cache and 4 * 8 GB for the VM.
Get real numbers and you are fine.
Thanks for the reply. But some of the windows vm where taken only consumed memory. But some of the vm were consuming full allocated memory. This is the problem. Though if it is taking lkke full allocated size means we can't able to determine that...
This is what i'm expecting, i simulate a real pc with x GB Memory, so the Hostsystem must provide x GB Memory und WinOS has the choice to take them all.
Dear team,
I am facing an issue with memory usage. When I install windows server 2022, 2016, 2019, 2025 the host memory usage was taking as same what is have allocated for the vm. Some of the windows vm were consuming low host memory usage than...
A few points I would add to the answers above:
Single socket does not automatically mean “one NUMA node”. On EPYC this depends on the NPS setting (NPS1/2/4), and Intel has a similar concept with SNC. Since you specifically require at least 64...
Danke! Das verlinkte Howto von @UdoB notiere ich mir gleich für später. Ob NAS oder Debian-Server, mal sehen. Einer der Debian-Server hat bereits QEMU/KVM drauf, allerdings bin ich noch nicht sicher, ob der in Betrieb bleibt usw. / Ich drehe...
Das ist insofern problematisch, weil man dann vom qdevice mit ssh als Root auf die Cluster-Knoten kann und umgekehrt.
@UdoB hat beschrieben, wie sich das absichern lässt...
Update: I can confirm that the E5-2630L v4 was the cause of the reboot behavior. The issue went away with the v3. Unfortunately, the v3 runs hotter and noisier than the v4, but the stability is worth the noise.
Update: I can confirm that the E5-2630L v4 was the cause of the reboot behavior. The issue went away with the v3. Unfortunately, the v3 runs hotter and noisier than the v4, but the stability is worth the noise.
well in my case, it turned out to be a HW issue. The Tesla K80 that I have has 2 dies and it basically is 2 GPUs on 1 card but in many ways still acts and looks like 2 GPUs. GPU 1 is good but GPU 2 has a HW problem so as long as I was only...