Starting / Stopping VMs causing all host networking to hang temporarily

[vertigo2049]

Hi All,

Apologies if this is expected behavior just trying to understand.

x4 Node cluster currently on 5.3-8 with Ceph storage.
External Networking is managed by FRR vm talking OSPF on Vlan.
If i start VMs on the node hosting FRR (all using VirtIO interface) then i lose my OSPF routes for 10 - 15s.

Is this expected or should i be changing a something to stop this happening ?

Thanks

Si

 

[wolfgang]

Hi,

this is OSPF related and normally it should be faster.
But this defense from very many factors.

Is this expected or should i be changing a something to stop this happening ?

As Proxmox VE 5.3-8 does not support FRR from Proxmox there is nothing expected.

We have frr packages in or the current version but the development for integration is not final.