You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Right now we use a fake IP addr 10.0.0.15 for every boxen/vrnetlab VM and then do iptables/socat forwarding from container eth0 interface towards nested VM's management interface (which is tap0 inside the container).
Now this fake static IP address causes issues. Not only it is not beautiful to have some internal IP set on all nodes of lab, but that also messes things with management systems when nodes call back and report some fake IP address to the NMS.
I wanted to explore a way to provision VMs with the IP address, that containerlab provides (i.e. real management IP). That way we would have a clean management datapath with no forwarding rules or fake addresses.
This implies a change in the way management interface should be provisioned in vrnetlab/boxen. Instead of using qemu's hostfwd as we do in vrnetlab for cisco/juniper and instead of using br-mgmt and iptables forwarding as we do in SR OS we can use tc mirred redirect (as we do for datapath ports).
It looks like it should be possible, but a few limitations arise:
How to enable tftpboot for SR OS that we use currently for provisioning license and mirror config file?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Right now we use a fake IP addr 10.0.0.15 for every boxen/vrnetlab VM and then do iptables/socat forwarding from container eth0 interface towards nested VM's management interface (which is tap0 inside the container).
Now this fake static IP address causes issues. Not only it is not beautiful to have some internal IP set on all nodes of lab, but that also messes things with management systems when nodes call back and report some fake IP address to the NMS.
I wanted to explore a way to provision VMs with the IP address, that containerlab provides (i.e. real management IP). That way we would have a clean management datapath with no forwarding rules or fake addresses.
This implies a change in the way management interface should be provisioned in vrnetlab/boxen. Instead of using qemu's
hostfwd
as we do in vrnetlab for cisco/juniper and instead of using br-mgmt and iptables forwarding as we do in SR OS we can use tc mirred redirect (as we do for datapath ports).It looks like it should be possible, but a few limitations arise:
Beta Was this translation helpful? Give feedback.
All reactions