Do I need to choose a mirror for a server or a not moving computer to reduce the load on the central servers?

deb.debian.org is somehow load-balanced using a global CDN. Will this CDN-backed service pick the fasted mirror for me every time I update or will it just load-balance within the infrastructure of Debians main servers? The connection and download speed is not very different when choosing a mirror near me or just using the default.

I want to do what is best for the project.

I want to make windows clients at my workplace more secure by using software obtained with winget and have it automatically updated on a regular schedule. I have a Linux (Gentoo and Debian) background.

In the majority of cases the users are AD users without Administrator rights, so they cannot do winget upgrade --all in PowerShell. My idea was to create a scheduled task which runs as the SYSTEM user, but unfortunately, a PowerShell spawned that way cannot access winget, reporting that this Cmdlet cannot be found.

I recently saw WAU (Winget-AutoUpdate). I did not try it myself yet. Can it do the job? What are you doing to maintain 50+ windows clients with users that are not Administrators on their system and lack the knowledge to update software besides what Windows 11 does for them out-of-the-box.

Interestingly, there does not seem to exist anything on Windows that is as easy as cron, systemd.timers or unattended-updates. And, in most cases users of Linux clients get sudo rights, because you can expect some basic knowledge about the package manager. On the other hand it wouldn't strictly be neccessary if they are not devs and need only a static set of software. The beauty of having it all in one repo + flatpaks in user space makes it all possible on Linux.

Even with winget which is a great relieve on Windows, btw., OS updates are seperate from app updates; basically only "flatpak", but without native auto-updates.

One additional remark: The apps need to be preinstalled before a new AD user logs on; I have to use --scope machine with winget. Users should not be bothered installing software themselves, not even with winget install --scope machine

I like to read, what you are using and I hope, it can be done without spending money on it. An open source solution is preferred.

I want to transfer 80 TB of data to another locatio . I already have the drives for it. The idea is to copy everything to it, fly it to the target and use or copy the data on/to the server.

What filesystem would you use and would you use a raid configuration? Currently I lean towards 8 single disk filesystems on the 10 TB drives with ext4, because it is simple. I considered ZFS because of the possiblity to scrub at the target destination and/or pool all drives. But ZFS may not be available at the target.

There is btrfs which should be available everywhere because it is in mainline linux and ZFS is not. But from my knowledge btrfs would require lvm to pool disks together like zfs can do natively.

Pooling the drives would also be a problem if one disk gets lost during transit. If I have everything on 8 single disks at least the remaining data can be used at the target and they only have to wait for the missing data.

I like to read about your opinions or practical experience with similar challanges.

Hi, I have an old Dell server I want to repurpose as a storage server and I decided to use zfs. I have read that gradual increases are best done with mirrored vdevs and only if I want to max out, I can choose Z2. What are your experiences in practice?

I like to go the gradual increase route because I cannot afford to maximize from the beginning. I like to start with ~10TB usable space. The server has at least 8 free drive bays. Should I go with a mirror as my first vdev or go 6x3TB with Z2? Ideally I want to be able to add more vdevs to the same pool in the future.

I am also curious whether it is a bad idea to have an uneven pool with mixed vdev types, the first being mirror and the second Z2. Will the vdevs eventually even out over time because zfs is a CoW filesystem?

I have a room with 30 thinclients currently running Windows with the possibility to open RDP sessions.

The current setup using VMWare is slow even for 2D content. That is why I want to replace it with Proxmox. What can I expect? I suspect the current setup is using SAN, I want to go ZFS on local drives.

I experimented in a homelab on KVM to see how fast the VMs can become. With 8 cores Google Earth becomes somewhat usable over RDP. But imagine 30 students using it on the same VM. The VM is Debian 13 btw.

I also experiemented with spice and 3D acceleration, but it works only locally and does not support multiple logins. What other options do I have. Even when I setup the VM to use virgl it uses software rendering over RDP. I thought of replacing the Windows on the thinclients with Linux, but then I would need individual VMs for every student and a secure spice session. Is that even possible? I would need a potent GPU in the server, maybe more than one. Is a 64 core CPU and 512 GB RAM enough for 30 students?

I've read that proxmox uses temporary .vv files for noVNC in the browser. I hope this can be setup permanently to be accessible over the network.

Any advice or new ideas are welcome!

What are your experiences with btrbk? I have found it recently and noticed, the last contributon was 2 years ago. Is it still a healthy project or should one stick to other backup solutions for btrfs snapshots?