wlan-cloud-ucentral-deploy/docker-compose/owls

OpenWifi OWLS Docker Compose

Deployment with self-signed certificates

To run a load simulation you need to generate a specific Digicert-signed AP certificate which will be used to connect to the gateway. The certificate serial number has to start with the digits 53494d since otherwise the gateway won't allow a load simulation. The rest of the serial number and the specified redirector URL can be chosen randomly. You only need to generate one AP certificate for your simulations. Be aware that since the OWLS deployment partly exposes the same ports on the host as the OpenWifi deployment, it is not intended that both run on the same host.

1. Copy or move your AP load simulation certificate into the docker-compose/certs directory. Don't forget to name the files device-cert.pem and device-key.pem or adapt the path names in the OWLS configuration if you're using different file names.
2. To be able to run load simulation tests against your OpenWifi SDK deployment, you'll have to add the serial number of your generated AP certificate to the gateway configuration. You can do that by either editing owgw.env or doing the changes directly in your OWGW configuration file if it is exposed on your Docker host.
3. Switch into the project directory with cd docker-compose/owls.
4. Add an entry for openwifi-owls.wlan.local in your hosts file which points to 127.0.0.1 or whatever the IP of the host running the OWLS deployment is.
5. Spin up the deployment with docker-compose up -d.
6. Check if the containers are up and running with docker-compose ps.
7. Add SSL certificate exceptions in your browser by visiting https://openwifi-owls.wlan.local:16001 and https://openwifi-owls.wlan.local:16007.
8. If you're using an OpenWifi deployment with self-signed certificates, you'll have to add a custom hosts entry for openwifi.wlan.local on the machine running the OWLS deployment pointing to the remote IP of your OpenWifi host.
9. Login to the UI by visiting https://openwifi-owls.wlan.local and follow the instructions to change your default password.
10. In the Simulation tab, click on the + sign on the right side to add a load simulation.
11. Fill out the required fields. MAC prefix is used for the MAC addresses of the simulated devices, so you can use any six-digit hexadecimal number. Specify the remote address of your OpenWifi gateway in the Gateway field, for example https://openwifi.wlan.local:15002. Adapt the rest of the settings according to your needs.
12. Click on the floppy disk icon to save your load simulation. You can run it by clicking the play symbol in the table view.

Deployment with Let's Encrypt certificates

To run a load simulation you need to generate a specific Digicert-signed AP certificate which will be used to connect to the gateway. The certificate serial number has to start with the digits 53494d since otherwise the gateway won't allow a load simulation. The rest of the serial number and the specified redirector URL can be chosen randomly. You only need to generate one AP certificate for your simulations. Be aware that since the OWLS deployment partly exposes the same ports on the host as the OpenWifi deployment, it is not intended that both run on the same host.

1. Copy or move your AP load simulation certificate into the docker-compose/certs directory. Don't forget to name the files device-cert.pem and device-key.pem or adapt the path names in the OWLS configuration if you're using different file names.
2. To be able to run load simulation tests against your OpenWifi SDK deployment, you'll have to add the serial number of your generated AP certificate to the gateway configuration. You can do that by either editing owgw.env or doing the changes directly in your OWGW configuration file if it is exposed on your Docker host.
3. Switch into the project directory with cd docker-compose/owls.
4. Add an entry for openwifi-owls.wlan.local in your hosts file which points to 127.0.0.1 or whatever the IP of the host running the OWLS deployment is.
5. Create an alias alias dcowls='docker-compose -f docker-compose.lb.letsencrypt.yml.
6. Change SDKHOSTNAME in .env to the desired externally reachable host name. This name must resolve to the IP of the host. ie. SDKHOSTNAME=owls.example.com
7. Change in owls-ui.env: REACT_APP_UCENTRALSEC_URL=https://owls.example.com:16001
8. Change SYSTEM_URI_PUBLIC in owsec.env, owfms.env and owls.env. ie. for owls.env: SYSTEM_URI_PUBLIC=https://owls.example.com:16007 SYSTEM_URI_UI=https://owls.example.com. Make sure you maintain the correct port for each service in SYSTEM_URI_PUBLIC.
9. Make sure the following ports are open: 80, 443, 16001, 16004, 16007 (80 is required by the initial Let's Encrypt challenge.)
10. Spin up the deployment with dcowls up -d.
11. Check if the containers are up and running with dcowls ps.
12. Login to the UI by visiting ie. https://owls.example.com and follow the instructions to change your default password.
13. In the Simulation tab, click on the + sign on the right side to add a load simulation.
14. Fill out the required fields. MAC prefix is used for the MAC addresses of the simulated devices, so you can use any six-digit hexadecimal number. Specify the remote address of your OpenWifi gateway in the Gateway field, for example https://gw.sdk.example.com:15002. Adapt the rest of the settings according to your needs.
15. Click on the floppy disk icon to save your load simulation. You can run it by clicking the play symbol in the table view.