If this is for the first time that we create a new component on the elastic.io platform, we need to create a new team first. In this particular team, there is currently only one member named Igor Drobiazko.
Next step is to create a new repository for our component. The repository will be assigned to our nearly created team.
Now we need to perform four steps to start the component:
- The first step is to upload the SSH key;
- Then we clone the Hello World component, which is basically the general template for our new component;
- Next we make the necessary changes,
- And lastly, we push the new component back to elastic.io.
Creating the component: Step 1
In order to be able to get identified in the future, you need to copy your public SSH key from your computer and upload it on elastic.io. Due to this, elastic.io will be able to do the proper authorization when you push you code.
Creating the component: Step 2
Then we go to GitHub (we have Java and Node.js), create a clone of our component and copy it on our local computer.
Creating the component: Step 3
On your local computer, you can make any changes to your code that you need. In this video, though, we are skipping this step to save some time.
Creating the component: Step 4
Next, we push the code back to elastic.io. It’s building, downloading everything it needs, running the necessary tests. When it’s ready, we see the message “Build completed successfully”. And basically, this is it.
Testing the component
Now we need to reload the page, and we can already see the first version of our freshly built component. Let’s test this new component with another one by creating a new task. In this task, the new component is the trigger, and the existing component is the action.
We’ll rename the task as Demo Flow. Now we need to perform authorization, i.e. enter credentials; for the sake of time we used here a very simple authorization, just the first name of the team member. In a real-life scenario this would be e.g. an API key.
Next we map the data, save it and start my task. Then we just need to wait a bit until all data records pass from the trigger to the action component.
Checking the logging data
After the start, you can immediately access logs of each single step in case anything goes wrong and check where your data got stuck and why. Or in case you have large data volumes and several components connected to each other, the logs help you check where your data is at a certain moment.
In our case, as you can see we don’t have any errors.