In this lesson you will
- Move your graph to CloverETL Server and run it
- Monitor progress of a running graph
- Schedule its execution
- Send an automated email after the graph executes
- Connect a Server project (sandbox) to Designer
CloverETL Server can run graphs and jobflows unattended, keep logs, send status emails or set up complex workflow processes. The Server is a multi-user platform that can scale up to multiple nodes facilitating parallel transformations and load balancing.
CloverETL Server’s main features
Basics for running and monitoring jobs:
- Monitoring – main Server health overview page
- Executions History – historical log of all jobs executed by the Server
- Sandboxes – file browser for projects (sandboxes) stored on the Server
Advanced orchestration features:
- Launch Services – configuration for “publish transformation as Web Service” feature
- Scheduling – built-in scheduler for tasks like graph execution, email, etc.
- Tasks History – historical log of events like scheduler triggers, event listeners, etc.
- Event Listeners – configuration for triggers such as file appearance, job finish or failure etc.
- Configuration – you don’t need to worry about this for now, unless you’re a power user.
CloverETL Server is accessible from a web browser or directly from Designer (by connecting to sandboxes, read more)
A sandbox is the Server counterpart of a project in Designer. It holds graphs, data, jobflows, metadata, shared connections, etc.
You can create sandboxes using the Server Console
or directly from Designer…
Let’s take what we’ve done so far in CloverETL Designer and deploy it to CloverETL Server
You can manage all assets – data files, graphs, jobflows, shared connections, metadata and so on in Sandboxes.
You can track what’s been running on your Server in Executions History, and also what’s currently running.
Scheduling is where you instruct the Server to run your graphs automatically, based on a schedule.
A schedule can be:
- a one time event
- periodic, running only within given intervals
- periodic, running every second, minute, hour, day, day of week, etc.
For “Start a graph” task, you define the graph (from which sandbox) you want to run.
You can tell the Server to perform various tasks as reactions to events in the system – e.g. a graph finishing successfully or failing, a file appearing in a watched folder, a message landing in a message queue, etc.
Once you have your projects on the Server as sandboxes, you can connect to them from Designer directly. That way, you don’t need to export your work after every change.
Now, the “QuickStart (Server)” project in your Designer is a live connection to the Server sandbox. Any modifications you make are directly reflected on the Server. When you run a graph it will run on the Server as well.
You can view a job’s status (numbers of records processed, log file) in Designer for any historical entry in Executions History.
This is particularly useful when debugging errors for graphs executed automatically in the past.