Recent content by gurubert

Updates should be made node for node and include a node reboot if necessary (new kernel, systemd etc).

The VMID or the CTID is the first number in the interface name. veth129i0 is the first interface of container 129.

These PGs are lost if all of your OSDs are online again. You will have to remove them. Look in the Ceph documentation under PG Troubleshooting.

You have lost data.

When you only have one fast network connection use that for the Ceph public network and do not configure a cluster network. A cluster network is only useful when you have a separate physical network infrastructure that is at least twice as fast as the one for the public network.

How many MONs does the Ceph cluster have?

If you need more than 32 nodes for compute maybe you have outgrown Proxmox VE and should look at larger systems like OpenStack.

Da ist tatsächlich ein Anwendungsfall für mehrere Pools und Quota. In einem Pool, der in Proxmox als Storage eingebunden wird, kann dann jedes Institut für sich mehrere VMs ablegen. In Proxmox lässt sich das dann auch verrechten.

I do not know if anyone can help you.

Das RBD ist die virtuelle Festplatte für eine VM. Dieses hat immer eine bestimmte Größe. Ein Ceph-Pool kann mehrere RBDs aufnehmen und ist grundsätzlich so groß wie der gesamte Cluster. Außer er wird mit einer Qouta versehen. Mehrere Pools teilen sich die Gesamtkapazität des Clusters. Ich...

You can only jump over one Ceph version when upgrading. So from 14 you can upgrade to 16 and only after that to 18. The upgrade procedures are documented on https://pve.proxmox.com/

Proxmox is a virtualization platform, not a cloud platform.

Why would you want to tunnel storage traffic through a tunnel like VXLAN?

Usually when building a Ceph cluster one starts with the MONs and not the OSDs.

With m=1 you have the same redundancy as with size=2 and min_size=1 or in other words you have a RAID5. You will lose data in this setup. You could run with k=2 and m=2 but will still have to cope with the EC overhead (more CPU and more network communication).