ifup / ifdown on bridge makes VM unreacheable

[be_tnt]

Hello!

I must forget to do something or I do it wrong but after a modification in the interfaces file on my host, I did:

# ifdown vmbr1
# ifup vmbr1

After that, my VMs linked to this bridge are unreacheable (only via console).
The only way to make it work again is to reboot the host

 

[Stoiko Ivanov]

This is the way things work with ifupdown :/ - once you take a bridge down it gets removed, and along with it the association of all VM's tap-devices to that bridge.

Rebooting is a safe way of doing this, but since a short while we have ifupdown2 (a drop-in replacement for ifupdown) in our repository, which supports reloading an interface - you can give that a try!