Recent content by telvenes

It is strange, there is no refrences to it trough /etc/ceph/ceph.conf or /etc/pve/ceph.conf cluster: id: id health: HEALTH_WARN 1 pgs not deep-scrubbed in time services: mon: 3 daemons, quorum pve01,pve03,pve02 (age 22h) mgr: pve03(active, since 3w)...

How can i get this mon.mon.pve02 removed? it was wrong created after crash of the pve02

Yea but as you described was to make some custom with simple zone?

I have now tried and cannot get this working, but im using ceph against the looback ip, it it intended to use the bridge for the cluster address?

also other nodes in the cluster?

Have you made it work the other way as well. From VMs to mesh network?

Did you ever find a solution to this? I have finally made it so i can use the mesh network for both sdn and ceph network. VMs can also connect to ceph host ip but not trough the fast meshed network.

Did you ever find a solution to this? Trying to figure this out myself.

Hello I tried to add custom config /etc/network/interfaces ifcae vxlan address 10.0.0.21/24 So now all vms can talk to and vms only located on the host that have configured 10.0.0.21/24 can ping 10.0.0.21, but if i move the vm to another host 10.0.0.21 no longer responds. Is this even...

what about installing frr on on the vm and let it be part of the announcment if we example add a vmbridge into the frr for the hosts?

I have also found this that sounds to be something simular what I ask of but I really do not understand how this solves it: https://forum.proxmox.com/threads/full-mesh-routed-setup-evpn-vm-bridge.148659/

thanks for your reply. I have solved this temproary adding a route ceph cluster ip via management ip. but this limits the whole setup trough the very slow management network. was really hoping someone had a solution for it :)

Hello, Been running full mesh frr network for my ceph and it works great. But im struggeling with getting access from vms, is there someone else here that have been able to solve this?

I have a 3-node cluster, and I followed your instructions, and everything works perfectly. But I have another problem: I have a k3s Kubernetes cluster using the ceph-csi storage plugin. I really want to access the cluster network of Ceph from my VMs. The way I’ve done it now is by adding a...