Recent content by Dominic

Could you please post your VM configuration? qm config <vmid>

Have you been able to solve this already? Is this the only container for which backups do not work?

But RAID1 requires at least 2 disks? The best place to do so is the Proxmox Bugzilla

Hi, currently such a feature does not exist. There is, however, a request for it in the Proxmox Bugzilla. You can add yourself to its CC list to get notified about changes to its status.

Hi, have you installed the VirtIO Balloon Driver as outlined in the best practices? If so, which version are you using exactly?

There is a bug entry in Bugzilla. You can add yourself to the CC list to get notified about changes to its status. You can also find some unofficial solutions in the forum or rather on GitHub.

The best practices for Windows guests have not been changed by the jump from PVE 6 to 7. Backup & restore is a good & reliable way to transfer your VMs.

Hi, have you been able to solve this already? Usually posting cat /etc/network/interfaces ip a from both Proxmox VE host and within the VM is really helpful. In addition, you can debug really many networking problems using tcpdump, for example: tcpdump -envi vmbr0

You can install rasdaemon in Proxmox VE apt update && apt install rasdaemon

Hi, if you absolutely want TrueNAS then this sounds OK. You can also think about letting Proxmox VE handle the physical disks and create a (maybe ZFS) storage on it install some (other) data storage solution in a VM assign the VM virtual disks residing on the storage from step 1

I would not recommend two-node Ceph cluster. Instead you could look into ZFS & storage replication. In addition, I would avoid the partitioning. You can give PVE one of the slower drives (or an additional 128GB SSD) and use the NVME as cache for ZFS. Note that for both Ceph & ZFS you should use...

Which storage controller do you have in your server? In general ZFS is a good choice, but it does use 50% of the host memory for caching by default. Depending on how much memory you need for your computations and the controller, other storage technologies might be interesting as well.

Why is this unexpected? What VM options have you configured? qm config <vmid>

Yes, as explained by the preconditions in the upgrade guide. A temporary combination of 6.4 and 7 should work. There are some hints in the forum. Octopus is OK before the upgrade. You can upgrade later. See also in the upgrade guide.

Great to hear that you got so far! Unfortunately, that process often requires experiment with a couple of VM configuration options. Did you get the glitches with the noVNC shell? Maybe you can use some remote desktop solution (Chrome Remote Desktop) instead?