Join GitHub today
GitHub is home to over 40 million developers working together to host and review code, manage projects, and build software together.Sign up
Tentacle `poll-server` command fails to successfully register with second and subsequent nodes #5545
When using the
This previously happened during a call to
What I expected to happen
After executing the command, I expect the new node to immediately trust the new thumbprint and start communicating.
Steps to reproduce
Octopus Tentacle: 4.x till fixed
source: #5206 (comment)
If I remember correctly, restarting the nodes didn't seem to have any effect on this.
The other work around, at least in my case, was to use the 3.24.x version of the tentacle installer, and then letting Octopus Server update the tentacle during the first deployment to the machine.
I'm happy nonetheless that you identified this as an issue, hopefully this is something resolved at some point in the near future. We'll just keep using the older tentacle installer in the mean time.
@DerekCampbellOctopus @matt-richardson We started running into this issue after the polling-server cmd was fixed for worker nodes in v4.0.1 of the tentacle (#5206). But I just now ran into this issue again after installing the latest version of the tentacle as a deployment tentacle (not a worker).
If the tentacle communicated with the secondary node it failed. But if it communicated with the primary node it worked. You can test this by draining nodes and running healthchecks on the tentacle. I ran the wrokaround mentioned above and that seemed to have solve the communication problem. Not really sure why or how though....