Failed to update executor docker+machine for 9fb5fe99 No free machines that can process builds
I get the above message whenever I run gitlab-runner. The runner is on an aws ubuntu 14.04 instance with docker-machine version 0.6.0 (also tried it with the previous version, same situation) This is my config:
concurrent = 10
[[runners]]
name = "Coordinator"
url = "my url"
token = "my token"
executor = "docker+machine"
[runners.docker]
image = "ubuntu:14.04"
privileged = false
volumes = ["/cache"]
allowed_images = ["*", "*/*", "*/*/*"]
allowed_services = ["*", "*/*"]
[runners.machine]
IdleCount = 1
IdleTime = 1800
MaxBuilds = 100
MachineDriver = "amazonec2"
MachineName = "auto-scale-%s"
MachineOptions = [
"amazonec2-vpc-id=my vpc id",
"amazonec2-subnet-id=my subnet id",
"amazonec2-zone=b",
"amazonec2-access-key=my key",
"amazonec2-secret-key=my secret",
"amazonec2-region=us-west-2",
"amazonec2-private-address-only",
"engine-registry-mirror=http://172.31.27.193:6000"]
[runners.cache]
Insecure = false
Using gitlab-runner version 1.2.0~beta.99.ga2cf2c6 but I get the same error in the current stable 1.1.3 ( I believe)
Also, docker-machine ls
comes back empty every time, but I do see an ec2 instance spin up.
Update: it turns out docker-machine was being run as the root user, and I needed to su to root to see what was going on. So far, it looks like it’s trying to provision an ami with docker, it just takes forever on a micro. Problem not solved quite yet, I need to see how this resolves itself first.