Installing Tentacle to image for cloud deployments

Legacy Uncategorised
1

Hi all,

We run a cloud based infrastructure with the ability to spin up more servers as load requires it.
We’re looking at using Octopus Deploy for all of our deployment needs and we are wondering if there is a way to have a Tentacle already installed on the image that we use.

To try this; I created a new server, installed the tentacle, took a snapshot of that server and then created a new server from the snapshot. The new machine was then added to Octopus. When running the health check there was a warning message of This machine shares its endpoint..

I’m assuming that this is because there are two tentacles that share the same thumbprint and Octopus believes that they are the same server, just added twice. Is there a way that we can have the tentacle installed on our image but generate new thumb prints when we spin up a new server?

2

Hi,

Tentacles in Octopus all have a “SQUID”, a special unique identifier that shouldn’t be duplicated.

When you provision a machine, you can use the command line/scripting to download, install and configure Tentacle completely. You can see an example here:

This way instead of installing Tentacle and then creating a snapshot, you’d just have a snapshot that you clone, and then install Tentacle and register it with your Octopus.

I hope that helps,

Paul

3

Thanks for the reply Paul.

We’ll look at adding a script that performs the set up and use something like a RunOnce registry key to perform the set up.

Thanks again,
Kristian