Installation Questions

I submitted the installation form but got an SSL error reported with my GitHub Enterprise installation…

This means that the GitHub Enterprise server is not returning the intermediate SSL certificates. Check your Github Enterprise instance with - it may report some missing intermediate certs. You can use tools like to get the full certificate chain for your server.

I started a builder box, but builds are not running…

Builder boxes can take a long time to boot - somewhere on the order of 15-25 minutes, depending on the size of the image. When they are ready, you should see some directories in /mnt (e.g. /mnt/box5). If you go for an hour without your build box being available send us /home/ubuntu/circle/circle.log, and we can debug it for you. It should not contain any sensitive information. You can look for errors related to btrfs or container_fs.

Common causes:

  • The box has no attached instance storage
  • The box was started on unsupported instance type

Admin users can view the state of their build fleet inside of the CircleCI Enterprise UI.

Builder boxes are up, I see /mnt/boxNN folders, but builds are not running

Check that your network configuration allows for the services box and builder boxes to communicate with each other on all ports.

You can confirm that this is the issue by looking for non-localhost traffic in /var/log/nginx/access.log. In a healthy installation, you should see some traffic from the services box’s private IP address.

Can I use Terraform? CloudFormation? ARM templates? Custom AMIs with Puppet?

Support for various provisioning tools is in the works. Contact with specific requests.

Can I use AWS Auto Scaling Groups to manage my builder boxes?

Yes! You can configure your ASG with your preferred time-based scaling configuration, and the following instance user data:

curl | \
    SERVICES_PRIVATE_IP=<private ip address of services box> \
    CIRCLE_SECRET_PASSPHRASE=<passphrase entered on system console (services box port 8800) settings> \

My GitHub Enterprise installation doesn’t use SSL or uses a self-signed certificate.

When setting up CircleCI Enterprise, the installation wizard has the option to select the GitHub Enterprise protocol. GitHub Enterprise using HTTP should work out of the box.

If you’re using a self-signed cert, or using a custom CA root, you can select the “HTTPS (with self-signed certificate)” option. You would also need to export CIRCLE_IGNORE_CERT_HOST to be the GHE domain. So the builder initialization command is (replacing below with the host of your GitHub Enterprise instance):

curl | \
    SERVICES_PRIVATE_IP=<private ip address of services box> \
    CIRCLE_SECRET_PASSPHRASE=<passphrase entered on system console (services box port 8800) settings> \ \

Can I use HTTP without SSL for webhooks and other communication with CircleCI Enterprise?

We strongly discourage using HTTP and suggest using HTTPS. You may use self-signed certificates (and the installer wizard auto-creates a self signed as well if you wish.


Can I monitor available build containers?

Yes, basic reporting on the state of your build fleet is available in the Admin UI. We welcome your input on what tools or data would be most useful to you in managing your build resources, so let us know at

How do I provision admin users?

The first user who logs in will automatically be designated an admin user. UI Tools for designating additional admin users can be found under the Users page in the Admin section (normally found at https://[domain-to-your-installation]/admin/users).

How do I back up and restore all of the data in my installation?

Backup utilities are coming soon. We would love to get some feedback on how you like to maintain backups. We are planning to provide a CLI tool to generate a backup dump, which you can then upload to your preferred backup storage. The CLI tool would have a restore option as well. If you have an alternative preferred workflow, let us know at

Starting with releases in 0.1.x, the following manual mechanism is available:


  1. Take an EBS snapshot
  2. Backup /etc/circle independently (this contains sensitive data) in your favorite place


  1. Start a new version from the AMI
  2. Attach the EBS volume
  3. Restore /etc/circle content
  4. Reboot machine

The process works for releases prior to 0.1.0 as well, however, it requires that the new instance reuse the same private ip address of previous installation.

Oops - lost my builder instructions? How can I get the passphrase and private ip address?

SSH into into the services box, and run the following:

$ # To get the passphrase
$ grep PASSPHRASE /etc/circle/generated
$ # To get private ip address
$ ifconfig eth0 | grep "inet addr"
          inet addr:  Bcast:  Mask:

I want to change my settings? Can I do that?

Yes! Take a back up of /etc/circle – edit the values in /etc/circle/user appropriately and run circle-update-settings.

For changing ssl cert, you can place them in /etc/circle/ssl/ssl.key and /etc/circle/ssl/ssl.crt in the services box.

How can I gracefully shutdown an instance

SSH into the builder box you are ready to shutdown and run:

$ echo '(circle.backend.system/graceful-shutdown)` | lein repl :connect 6005

Can I run iOS/OS X builds like on

We currently provide limited, early access for iOS builds from CircleCI Enterprise that run on our own cloud of OS X machines. Contact for more information.

General CircleCI Questions

Do you support “build matrices” (e.g. running a build for each of several different ruby versions)

CircleCI does not support build matrices. There are ways to do this with a fair amount of manual configuration, but such a workaround is incompatible with inferred commands and automatic parallelism support.

Can I automatically create and configure projects based on some standard template?

You can use the API. If you need to export settings from CircleCI, let us know - we have an experimental export tool that we can share with you.