Octopus Deploy saying tentacle needs upgraded even after they were upgraded

So we are trying to do clustered servers in octopus, so we have our db01 server and db02 server share the same thumbprint and cert. But we are also adding the virtual server as a tentacle but it really just points to either db01 or db02 at any given time. The issue is that both db01 and db02 were updated, but the virtual server is saying it needs upgraded, when in reality it is actually just one of those servers so it should be upgraded already. what does it look for to say it is upgraded?

I’m guessing this setup is not something that is supported or suggested, but we figured we would try it, and where it does deploy to the active server if you deploy to the virtual server, it’s just doesn’t seem to do well when it comes to upgrading.

Hi Corey,

When using a cluster there’s two ways to use Octopus.

In the first approach, there’s only one “Tentacle” - you install Tentacle twice, use the same SQUID, thumbpprint and cert. The service only runs on one machine at once, and the machine only appears in Octopus once - Octopus uses the cluster IP address to talk to the Tentacle.

The second approach, there are two Tentacles - they have different SQUIDs, thumbprints etc. and both services are running at the same time, and two machines appear in Octopus using different IP addresses/hostnames.

It sounds like you might have a combination of these approaches, perhaps the first might work better?

In either case, whether a machine needs to be upgraded or not is determined during the health check that Octopus runs. You can click the Check Health button on the environment page to force this to happen. After that, the need to upgrade should go away.

Hope this helps,

Paul