Hi!
Our setup for kvm for our company’s internal NFS Server does randomly disconnect from network.
Only shutting down (virsh shutdown, virsh destroy) and stqarting newly (virsh start Server) is working to recover.
Host: Archlinux - current (Nov. 2018, Kernel 4.19.-1, also current qemu/kvm/libvirt packages
extra/qemu-headless 3.0.0-4 [Installiert]
community/libvirt 4.8.0-2 [Installiert]
Network Setup: bridged devices
Guest: Debian GNU Linux stretch/stable, current Kernel
Linux axs 4.9.0-8-amd64 #1 SMP Debian 4.9.110-3+deb9u4 (2018-08-21) x86_64 GNU/Linux
Guest virtual Harddisk files reside on a DRBD8.4 Cluster using ocfes2
The virtual machine is configured via libvirtd Server.xml, excert:
------------------------------------------------------------------
…
kvm64
…
…
…
------------------------------------------------------------------
Mostly all the time the virtual machine runs really fine, works performant …
But sometimes, we did not figured out why, the networking dies a sudden death, the kvm disconnects from the bridged network, no connection is possible anymore.
The hosts’ bridge configuration is setup like this:
[root@h2i ~]# brctl show
bridge name bridge id STP enabled interfaces
br0 8000.2ac3fe217ab3 yes axs0
vnet0
Also trying to delete and readd the “vnet0” device to the host’s bridge does not work.
It seems to be possible, that the disconnect of the virtual host occours during a raising load during a backup session, but this is not certain proved.
Does anyone have a idea?
TIA
J. Sauer
PS: