Any particular recommendations on deploying a Connector into GCP?

I am stepping into new territory with TG. I’ve only been a client user. Now I need to venture forth into getting things set up with our new GCP presence and knocking the rust off my GCP skills. I was wondering if there are particular recommendations for working with GCP, like: on what service/VM type a controller should be deployed (GKE? VM?) and best choice for instance size/type. What we need to stand up via TG infra access are a series of VMs and a Cloud Function. The Cloud Function will provide management of the VMs. I also would like to learn what needs to be done to work with automation of provisioning of a VM, specifically regarding registering the VM and getting the domain name (the Cloud Function needs to do this). Thanks for responses and pointers.

Hey Kevin,

We have a bit of documentation on deploying connectors in GCP here

Your “easiest” solution is to stand up a couple Linux VM’s and deploy the connectors via systemd, but deployment via GKE is absolutely possible.

The common gotcha with GCP deployment is that the VM(s) the connector(s) are running on will need internet access in order to download the package initially and of course to facilitate connections to the internal resources.

So as long as your connector(s) can reach both the internet and the internal resources you wish to access, everything should be straightforward.

Let us know if you have any other questions!



Thanks, @Arthur! I was also wondering what would be the best approach to registering new VMs as accessible resources and getting a custom domain name from the code in my Cloud Function.