We currently host Octopus Deploy internally and deploy to around 50 EC2 instances in two different VPCs.
This works because we have VPN between our internal hosting and each VPC. However, the VPCs do not have a VPN between them.
Our OD deployment is on an Windows 2008 server, and we no longer wish to host OD internally, instead we would like to host it on VPC1. However, since VPC1 cannot connect to VPC2, this breaks the connection to the EC2 instances on VPC2.
As far as I can tell, the preferred solution is to host a proxy server on VPC2, but I am concerned about security, since we don’t have any experience with hosting proxy servers in this manner.
Are there any alternatives? I know about polling tentacles, but I’m not too thrilled about those. Is it possible to somehow install OD in both VPCs and have them work in some kind of cluster mode?
Thanks for getting in touch. Sadly, Polling tentacle are the solution that we’d recommend in a situation like this, but it seems like you’re not too keen on that idea - can you elaborate more on why?
Listening tentacles is the “recommended” communication mode and also the one we have a good deal of experience with.
Also, it’s not clear to me how to change a Tentacle to polling mode? In the Tentacle Manager the “Communication Mode” does not seem to be changeable?
To change the tentacle to polling mode, you will need to reconfigure the tentacle, either through the Tentacle manager UI (creating a new instance), or via the CLI with tentacle.exe. Our Automating Tentacle Installation page has some handy scripts for configuring a polling tentacle
I hope this helps, and please don’t hesitate to reach out if you have any further questions.