Search results

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).

With 5 nodes you can have k=2 and m=2 which gives you 200% raw usage instead of 300% with size=3 replicated pools. But this is still a very small cluster for erasure coding.

EC with only 4 nodes is not useful. You need at least 8 or 10 nodes to get useful k and m values for erasure coding.

The total capacity of the cluster is defined as the sum of all OSDs. This number only changes when you add or remove disks. Do not confuse that with the maximum available space for pools which depends on replication factor or erasure code settings and currently used capacity.

So you will lose 25% capacity in case of a dead node. Make sure to set the nearfull ratio to 0.75 so that you get a warning when OSDs have less than 25% free space. https://bennetgallein.de/tools/ceph-calculator

Do not run pools with size=2 and min_size=1. You will lose data.

Ich kenne das Problem nur von OCFS2.

This is correct. But depending on how many nodes you have this is not critical.

The rbd_data.* objects seem to be leftovers. As long as there are no rbd_id.* and rbd_header.* objects there are no RBD images in the pool any more. The easiest way (and when you are really sure) would be to just delete the whole pool.

Seit 6.14 ist aio=threads nicht mehr notwendig.

Buy a non-defective enclosure.

You need to configure the old IP addresses from the Ceph public_network on the interfaces before you can do anything with the Ceph cluster.