network-mode: none - clab_intfs calculation logic #1851
Comments
kafo2205
changed the title
kafo2205
changed the title
kafo2205
changed the title
kafo2205
changed the title
kafo2205
changed the title
|
can you try the beta containerlab build and you will have to build the vsrx again using this branch https://github.com/hellt/vrnetlab/tree/count-eth0 Then you should be able to use eth0 in the vsrx and see it booting. Let me know if something is not clear |
|
hi @hellt |
|
@kafo2205 Here is a fix |
|
Cool, now it boots up :) this is my topology file srx-node0 <-> n9k-switch <-> srx-node1: lab is running: srx succesfully started, ip set to fxp interface (based on the logs) how to connect to srx console? Also tried to set up switch in the srx-fxp subnet but looks like there is no connectivity... |
|
you should be ablet to install telnet on your containerlab host and from there do |
|
actually on the host there isn't ip address assigned for network-mode: none containers so not able to telnet 5000 but I installed telnet in the container itself srx nodes are not able to ping thru the switch, I will try to interconnect them directly, but anyway do you know what is 52:55:0a:00:00:02 10.0.0.2 ?? |
|
it seems that to support Unfortunately at this point I won't have time to look at this. Maybe someone from the community can try that |
|
@hellt thank you, maybe I can try at least some manual w/a if you can share with me some examples how to do it :) |
|
for example how to modify this qemu cmd: to bypass qemu user interface that is used to represent the VM eth0 and allow to interconnect eth0 with tc backend |
|
there is no easy answer to that. It is done in hellt/vrnetlab project via different functions |
|
Unfortunately the change is more involved than I anticipated; We have to also change the provisioning logic for this case of a self-managed eth0 interface. And it drags a lot of changes that are not easy to merge in just now. So I think for now there is no quick way to have eth0 be managed by a user (and not docker) without doing some manual work... I will keep this issue open so that I remember about it when doing some rework in vrnetlab code base |
|
Sure I understand, I hope once it will be available I see several usecases like oob mgmt infra, ztp/dhcp, clustering, |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hello,
I am trying to use network-mode: none with VM-based node, but node does not start:
looks like clab_intfs calculation logic should be taken into account with
network-mode:nonepls also look at this check ethX, where X is >0**, there was an issue:
yaml:
Thank you for your support!
Br,
D
The text was updated successfully, but these errors were encountered: