Version 1.43 of the documentation is no longer actively maintained. The site that you are currently viewing is an archived snapshot. For up-to-date documentation, see the latest version.
On-premises setup (macOS)
- Unpack the service package
- Install the service
- Add your license
- Configure the service
- Start the service
- Configure the client
As of release
1.32.0, our Remote Execution software runs on the following
versions of macOS:
- 10.14 (Mojave)
- 10.15 (Catalina)
Need other OS support or want to run in an AWS or MacStadium environment? Contact us.
1. Unpack the service package
It contains the installer for the scheduler and worker services (
It does not contain a valid license file. We send you a license separately.
2. Install the service
On every machine, make sure you have at least Java 11 installed.
Need support for older versions of Java? Contact us.
Select at least one machine that will run the scheduler and install
scheduler.pkgon that machine.
Select at least one machine that will run the worker and install
worker.pkgon that machine.
Note that you typically have significantly more workers than schedulers in a cluster.
Warning: do not copy your source tree on these machines. The build tool uploads files if build actions need them.
3. Add your license
Copy your license onto every machine as
4. Configure the service
See the dedicated articles for details.
Tip: all service instances (schedulers and workers) can use the same
configfile. Schedulers ignore worker-specific options and vice versa.
(Optional) Customize the
configfile we sent you as part of the trial agreement
Copy the file to every machine as
(Optional) Customize the file per-machine:
Tip: for a first time trial setup we recommend using the
configfile we send you as part of the trial agreement. Later (and especially before productionizing) you should customize this config more. Consider:
As of release
localis the only supported execution strategy on macOS.
storage use (
See the Service Options Reference for more info.
5. Start the service
The service is configured to start automatically. All you need to do is to reboot the machine.
If rebooting the machine is not possible or not desirable, you can also start the service by SSH into every machine and running the following commands:
sudo launchctl start com.engflow.re.scheduler
sudo launchctl start com.engflow.re.worker
For first-time setup, or after changing the configuration, we recommend to start the service temporarily in a shell session to verify it is up and running:
6. Configure the client
See Client configuration.