executorType: machine


It looks like the first step in the Docker CI build is spinning up a new VM:

Build-agent version 0.0.2031-733a8ab (2017-01-03T16:12:43+0000)
Creating a dedicated  VM

Is this a temporary thing or will Circle 2.0 always be like this? I ask because it’s taking about 90 seconds which is a big chunk of our build time. Also sometimes it fails to create the VM.

Getting Started: Docker Engine

It is unclear at this time but we definitely want to speed it up as much as possible. The reliability will absolutely improve.


I appear to be having issues with both methods. The machine method has yet to successfully provision a VM on any build and I’ve had several cases of the allocation of a remote docker engine failing due to time out/greater than 60 attempts to do so.

Is there any way to monitor whether the services that provide those resources are under heavy utilisation or broken?


Hey. We had some issues with VM creation for a while today, it should be resolved now. Could you please try it again?