Octopus Consuming 100% Of Processor


Octopus seems to be maxing out the processors on some of our servers, slowing the servers to a halt and requiring us to restart them.

So far we’ve had this on our Trunk and Staging servers, both of which are 64-bit Windows Server 8 R2 SP1, but not in production yet (luckily). We are currently running Octopus Deploy Is there a way we could upgrade only some of the Tentacles, or will the new Server only talk to Tentacles of the same version?

We’re not entirely sure whether it’s the Octopus Server or Tentacle service which is the problem. Our Trunk server has both services installed and we’ve seen the Server service maxing out the processor. The Staging server only has the Tentacle installed but this server has also become unresponsive.

I can see lots of warnings in the Server logs informing us that loading the dashboard is frequently slow (> 10000ms).

The one strange thing I’ve done in the setup of the system is to have two Octopus Machines which point to the same physical machine. We do get a warning in the health checks informing us that the Machines share an end-point. This is due to two environments sharing the same servers but requiring different applications to be installed, depending on the environment.

I can send the log file, but I’ll do that over email.



Hi Douglas,

Can you try upgrading to the latest Octopus release? We fixed a number of performance related issues in that time which should resolve this.

You can upgrade the Octopus server without necessarily updating all of the Tentacles, though some features added in recent versions of Octopus that depend on the Tentacle may not work.


Hi Paul,

Okay, will upgrade.

Embarrassingly, it might have been one of our apps that was maxing out the
processor, not rock-solid, beautiful octopus. I was misinformed, but how
could I have doubted?!