Search results

For compatibility reasons it is always best to use VMs.

But HW raid is faster,as always.

Yes if you use nvme disks, something like 4-6 per node.

But i'm talking about push,not pull.

Why does source PBS needs to be Internet-accessible in this case?

The problem with Raidz and PBS is really low performance, that is why you always go with raid10.

No,im talking about special device in zfs. It can speed up garbage collection. It is always good to have them, if your zfs is 10tb or 200tb in some cases.

With those backplanes sometimes it works sometimes not. Really you need to test it with zfs. I had some 40disks backplanes work,and some not. But zfs raid10, maybe 2 nvme/ssd for caching and you're good to go.

I have one strange error: Kernel 6.14 ,lxc with mysql on it. /var/log/kern.log:3651:2025-04-10T10:35:40.646702+02:00 sp19 kernel: [236703.525309] audit: type=1400 audit(1744274140.640:2244): apparmor="DENIED" operation="create" class="net" namespace="root//lxc-1193_<-var-lib-lxc>"...

If i'm not mistaken Hetzner box doesnt allow acl?

I tested in around 10 machines,no problems atm.

Yes, if you have on the same machines zfs and ceph, just livemigrate storate from one to another. This is it

Could it be that it is from ubuntu or whatever linux virtual workspaces?

Virtual screens are usually created in the background with tigervnc server or xvfb ,depending on use-case

I meant for 6.14 kernel ofc

I'm already running 2-3 hosts, works okay :)

There is also clonezilla through the network,look at that.

There are two ways: 1. adding more disks to current nodes. 2. adding more nodes with more disks. Second one is always better.

And maybe you are using cephfs to store something

Destroy all osds per node, that would be easiest.