This article provides step-by-step instructions for deploying a new instance of the Network Server Stack on Google Cloud Launcher.
Start the Deployment Process
In the Google Cloud Launcher Console, select OccamSmart Network Server Stack, and click Launch on Compute Engine.
Create a New Deployment
- Rename your deployment.
- Select a machine type. An f1-micro instance should be sufficient for small networks (fewer than 50 devices) with a long transmission interval.
- Configure networking. Note that the Network Server Stack must be have port 1700 open for UDP data from the gateway, port 1883 open for TCP for the MQTT broker, and port 80 open for the web configuration interface. These should be open by default. It is highly recommended to restrict IP ranges to trusted origins, so that others cannot add device registrations to your network.
- Finally, click Deploy.
Your deployment should take just a few minutes to finish instantiating.
To check that all services are running, you can SSH into your newly instatiated vm, e.g.
$ gcloud compute --project YOUR_PROJECT ssh --zone "us-east1-b" YOUR_VM_NAME
and check on all of the server stack services:
systemctl status lora\*
as well as the Mosquitto MQTT broker:
systemctl status mosquitto
All of these should be in the
active (running) state.
Open Source Components
/etc/lora/open-source contains a text document called
open-source-licenses.txt with the licenses of each of the open source components used in the Server Stack, as well as
hashicorp.hcl.tar.gz which contains the source code for Hashicorp’s HCL configuration language, as required under the Mozilla Public License.
Now that your instance has been successfully deployed, you can start configuring it to receive data.