Automatically create and deploy an L3VPN service (template creation, configuration and validation that the network is in the way its configuration was planned and that there are no floating configurations), the l3vpn service model is loaded into an ansible playbook and based on The same model creates a template of configurations for each type of equipment involved in the service and then through NAPALM-Ansible library deploys the service to multiple network devices, in the example shown in the demonstration video on YouTube (I leave link below) an automated deployment of l3vpn to devices with IOS, XR, JunOS is shown and it is validated that they announce prefixes, BGP is used as PE-CE protocol, but any type of IGP can be used in the same way as PE-CE.
Python 3.x, Ansible 2.7 or higher, it can be run on windows as long as ubuntu as WSL is installed and from there run it
Only one version already tested and validated in emulated enviroment (EVE-NG with multivendor ISP Topology, see the video link below)
Service providers that wish to implement l3vpn services in an automated way, with support for multivendor, generates a standard model for the deployment of the service, in the same way this tool serves to the companies that have this service and require to automate the configuration to many devices saving time in deployment and validation.
Ubuntu 18.04 o any Distribution of Linux with support to Python3 and Ansible2.7 or higher (support with Ansible 2.9)
it can be run on windows as long as ubuntu as WSL is installed and from there run it (recomended)
sudo apt-get install python3 sudo apt-get install ansible sudo apt-get install git sudo apt-get install python3-pip sudo pip3 install virtualenv mkdir l3vpn-automation python3.6 -m venv venv-l3vpn source venv-l3vpn/bin/activate git clone https://github.com/Raul-Flores/L3vpn-Multivendor.git pip install requirements.txt
you have to go to the main directory, some PEs devices are found as an example you can substitute their names and associate them in the file etc/hosts in your linux the name with the IP address you need.
cat /etc/hosts This file is automatically generated by WSL based on the Windows hosts file: (output ommited for brevety) The following lines are desirable for IPv6 capable hosts ::1 ip6-localhost ip6-loopback 192.168.87.120 PE1 #(Change this in your hosts file for your devices) 192.168.87.47 PE3 192.168.87.24 PE4 192.168.87.129 PE2 (venv-l3vpn) rflores@Raul-Flores-Trabajo:~/venv-l3vpn/L3vpn-Multivendor/l3-vpn-svc
(venv-l3vpn) rflores@Raul-Flores-Trabajo:~/venv-l3vpn/L3vpn-Multivendor/l3-vpn-svc $ tree ansible.cfg candidate_config PE1-config.txt PE2-config.txt PE3-config.txt PE4-config.txt deploy-service.yml diff PE1-diff.txt PE2-diff.txt PE3-diff.txt PE4-diff.txt generate-model tasks main.yml templates data-to-device-model.j2 vars main.yaml generate-model.yml generate-template tasks main.yml templates ios core.j2 iosxr core.j2 junos core.j2 vars main.yml generate-template.yml hosts requirements.txt
There are .txt files (you can delete them), the program automatically creates them so delete these files (in the folders of candidate_config and diff)
rm candidate_config/*.txt
rm diff/*.txt
delete the file name main.yml that is in the path: (generate-template/vars/main.yml) This file is generated so that the template generation model can build the equipment configurations, delete it as it will be created automatically from the service model data.
rm generate-model/vars/main.yaml
to deploy the service once you have defined the equipment in the hosts file (here the user and password that will be applied by device type are defined), the names of the network devices correctly (in the file etc/hosts) and the devices have the configuration of SSH, then I would place the example commands to activate ssh and xml agent in XR, (this is important since if it cannot be reached through SSH it will not support it, the other examples of configurations by equipment (IOS, Junos) is beyond the scope of this information, (it is assumed that the engineer can configure this as it is something basic)
! crypto key generate dsa ! xml agent tty iteration off ! ssh server v2 !
(username and password admin by default)
once we have everything ready we will proceed to edit the variables in the file that is in the path: l3-vpn-svc/generate-template/vars/ main.yaml (create it and name it as main.yaml if it is not found), change the data of the variables as you need your L3VPN service:
example of configuration (you need change the values, important to follow the same structure, in this example the minimum required to build an l3vpn service is assumed (does not include the configuration of the session towards the RR, You have currenct active session MP-BGP in the up state with the PE to the RR):
--- common: bgp_asn: 64512 vpn_service: - vpn_name: VPN-A route_distinguisher: '64512:65512' route_target: '64512:65512' neighbor_remote_as: '65512' nodes: - name: PE1 rid: 11.11.11.11 vpn_int_mask: /30 ====> put / format if the device is Juniper vpn_int_ip: 11.0.0.1 ====> put the ip address of the local interface (PE interface connect to the Client) vpn_bpg_neighbor_ip: 11.0.0.2 ====>put the ip address to the customer (its assume do you know what its this ip address) interface: em2 ==> put the name of interface that you use to connect to the client (what interface connect to the CE) - name: PE2 rid: 22.22.22.22 vpn_int_mask: 255.255.255.252 vpn_int_ip: 11.0.0.5 vpn_bpg_neighbor_ip: 11.0.0.6 interface: GigabitEthernet1 - name: PE3 rid: 33.33.33.33 vpn_int_mask: /30 vpn_int_ip: 11.0.0.9 vpn_bpg_neighbor_ip: 11.0.0.10 interface: GigabitEthernet0/0/0/1 - name: PE4 rid: 44.44.44.44 vpn_int_mask: /30 vpn_int_ip: 11.0.0.13 vpn_bpg_neighbor_ip: 11.0.0.14 interface: GigabitEthernet0/0/0/1
run the foloweeds commands:
ansible-playbook -i hosts generate-model.yml
(this command create a main.yml file in the path: l3-vpn-svc/generate-template/vars/) please check if its correctly created.
ansible-playbook -i hosts generate-template.yml
(this command create all txt configs in txt in candidate_config file)
ansible-playbook -i hosts deploy-service.yml
(this command put the configuration using Ansible-NAPALM to the devices.)
you can check the configuration in the devices, or in the RR (see the video example)
If you have questions, concerns, bug reports, etc., please create an issue against this repository, or send me a email
to: raul.flores@ciberc.com
Code Exchange Community
Get help, share code, and collaborate with other developers in the Code Exchange community.View Community