Recent content by alexskysilk

the gold standard for volume partitioning is the same regardless of number of disks: part 1 (optional, default using pve installer) legacy grub part 2 efi part 3 root part 4 data/payload part4 layout and usage will depends on how you partition it. for this use case I'd format it btrfs, and you...

It has been ready for years. A high-friction CLI/UX workflow is not the same thing as a system being "unusable" or "not production-grade." Thousands of enterprise nodes run multipath iSCSI in production every day—it just requires Linux systems engineering rather than clicking through a guided...

DO you have a load balancer? what provides your internet service?

Or more. all designs based on chiplets (eg, all modern AMD) will have more then one per socket. /pedantic

From scratch? the same as any other restoration RTO calculation- the longest of (- time to read from source - time to transfer - time to write to destination) * processing coefficient (1.2 is safe with typical current hardware.) HOWEVER, PBS has live restore which means you can be operational...

Absence of evidence isnt evidence of absence. Searching for evidence after making a claim is backwards. your google searches arent even informative- the toms hardware is breathless marketing without ANY substance. the Unihost article discusses theory, not actual metal. and I wont even dignify...

You might want to cite your sources. and see what they were comparing.

While it is possible to install Trixie on Mac hardware, it is dependent on Asahi linux patches. I doubt the pve Kernel would even work. If this is something of specific interest to you, please try it and let us know :) this would be enlightening reading...

and yet you asked all the right questions anyway, correctly unwinding the "ceph" part from the "storage" part.

there are very few options for this. and because its all low volume, they are not at all cost or performance competitive with comparable x86. No. and even if it could run, it wouldnt support mac guests, support MLX either directly or indirectly. see https://asahilinux.org/

PVE isnt very good for/at multitenant applications. Since KVM works natively on these platforms, an operator for such usecase is more likely to just deploy them to a cloud native orchestrator, eg Openstack.

Dont conflate Nvidia hardware support as "AI." All this means in practice that nvidia "incentivized" Proxmox GMBH (probably with an NRE paid with nvidia couch change) to port PVE to their ARM cpus. Not really sure what for since the majority of AI workloads are dockerized but there you go. From...

You dont need a qdevice for a 3 node cluster; it will just serve as a 4th vote that doesnt have any consequence since its not a tiebreaker anyway. The lessons here are well learned, namely: 1. your cluster is only as stable as its network design 2. without share storage you're dependent on...

seems like you have some leftovers from whatever you tried earlier. begin with ceph-volume lvm disk zap /dev/nvme0n1 IF this completes successfully, check for the presence of ceph-osd@0.service and ceph-osd@1.service. if either exist, delete it. IF it doesnt, you'll need to use pvs to find and...

Here is the uncomfortable truth for people wanting to deploy ceph: Its not a black box. you WILL need to understand what it (and by extension, you) are doing. If a key file in the preconfigured location is missing, you need to either create/obtain it and place it in its expected location, OR...