Manage Eventing Functions and Settings

This is a quick introduction to managing the Eventing Service. We will create a default Eventing Handler and subsequently exercising the Eventing Handler’s functionality and operations across its typical lifecycle.

In order to show the basic life cycle of an Eventing Function, we need a source bucket and a metadata bucket. For the source bucket we will use the beer-sample sample document set. You must have a metadata bucket, typically called metadata, to be used solely by Eventing. Refer to Prerequisites

If a cluster is running the Eventing Service, Eventing Functions can be created or accessed under the cluster’s Tools > Eventing Service tab.

The Eventing systems has a prerequisites that there is at least two (2) isolated buckets. Here, for testing purposes we need two buckets, the first a source of the events or mutations and the second is a scratchpad for the Eventing Service (this can be shared across Functions).

For the first bucket or source of the Eventing mutations we will use the sample dataset beer-sample to load this Click the cluster’s Tools > Import tab.

The Import tab will be displayed, scroll down to the section called Beer Sample

Click Import to load the beer-sample dataset.

The second required bucket is an Eventing Service scratch pad area that can be shared between multiple Eventing Functions. We will just name this bucket metadata

Go to the Buckets tab.

Click Add Bucket to create the metadata bucket. Fill in the Bucket Name field with the name metadata, create it with a minimum size of 100MB.

Click Save to create the metadata bucket.

Now Go back to the Tools > Eventing tab.

If there are any Functions saved on the cluster, they are displayed in the Functions summary in the cluster’s Tools > Eventing tab.

The Functions summary lists all Eventing Functions regardless of the functions status. This list displays the following information:

When you create or import an Eventing Function, it is initially saved in an undeployed state. The Functions summary shows all functions on the cluster, along with their respective deployment status.

This operation activates an Eventing Function (or handler). Several validations are performed, and only valid handlers can be deployed. The JavaScript source code of an activated handler cannot be edited. Unless a handler is in deployed state, it will not receive or process any events (or mutations). Deployment creates necessary metadata, spawns worker processes, calculates initial partitions, and initiates check pointing of processed stream data.

Select Deploy from the Function’s pop-up menu

To deploy an existing Function, start by finding the specific Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Deploy to deploy the Function.

Click Confirm to deploy Confirm Deploy Function fly-out menu.

The Function will now enter the Deploying state and in several seconds will transition to the Deployed state and start processing events (or mutations).

The Functions summary now lists the function as deployed.

The deployed Eventing Function is now processing mutations.

This operation deactivates an Eventing Function (or handler). The JavaScript source code of a deactivated handler can be edited.

Select Undeploy from the Function’s pop-up menu

To undeploy an existing Function, start by finding the specific Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Undeploy to undeploy the Function.

Click Confirm to undeploy Confirm Undeploy Function fly-out menu.

The Function will now enter the Undeploying state and in several seconds will transition to the Undeployed state and cease processing events (or mutations).

This operation pauses and checkpoints an Eventing Function (or handler). The JavaScript source code of a deactivated handler can be edited. The checkpoint saves the point in the processing stream such that when the Function is later resumed no mutations will be missed. Handlers in Paused state can be either Resumed or Undeployed.

Select Pause from the Function’s pop-up menu

To pause an existing Function, start by finding the specific Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Pause to pause the Function.

Click Confirm to deploy Confirm Pause Function fly-out menu.

The Function will now enter the Pausing state and in several seconds will transition to the Paused state and cease processing events (or mutations).

This operation is similar to Deploy but reactivates an Eventing Function (or handler) and continues processing of a Function that was previously Paused. Several validations are performed, and only valid handlers can be resumed. The JavaScript source code of an activated handler cannot be edited. Unlike Deploy this operation will start from a previously saved checkpoint from the pause operation to ensure that no mutations are missed. On a resume the backlog of mutations that occurred when the Function was in the paused state will now be processed. In addition the backlog of timers that came due while the Function was paused will now also fire as soon as possible (ensuring that no timers are lost). Depending on the system activity and how long the Function was paused, clearing the backlog of mutations may take some time before Handler moves on to current mutations and timers.

Select Resume from the Function’s pop-up menu

To resume an existing Function, start by finding the specific Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Resume to resume the Function.

Click Confirm to deploy Confirm Resume Function fly-out menu.

The Function will now enter the Deploying state (there is no Resuming state) and in several seconds will transition to the Deployed state and start processing events (or mutations).

Eventing Statistics, for each deployed Function, can be viewed in the Couchbase Cloud UI, the available deployment statistics are:

To display the statistics for a deployed Function, start by finding the specific Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Deployment Statistics.

The resulting Deployment Statistics fly-out will show the current deployment statistics for the Function.

Each deployed Function has its own application log. Application logs allow you to identify and capture various business logic related activities and errors via user-defined messages specific to each Function.

The information that goes to these log files is solely dependent on the logic of the Function via log(…​) statements put inside the individual Function handler’s JavaScript code. As such, there is no logging level for application logs.

Application logs are primarily used for the development and debugging of business logic.

To display the application for a deployed Function, start by finding the specific Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Logs.

The resulting Logs fly-out will show the current application log for the Function.

To modify a specific Function’s settings, start by finding the Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Edit Settings.

Only a subset of the Function’s settings are available for modification after the Function is created.

To modify any of the above settings that require the Function to be in an undeployed or paused state, you will first need to undeploy or pause the Function by invoking the appropriate action from that pop-up menu at the right of the Function’s row.

Once you’re satisfied with the modifications you’ve made to the Function’s settings, make sure to click Save to commit the changes.

To modify a specific Function’s JavaScript code, start by finding the Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Edit JavaScript.

You may now use fly-out editor to modify and change your business logic.

Once you’re satisfied with the modifications you’ve made to the Function’s JavaScript code, make sure to click Save to commit the changes.

To delete a specific Function’s settings, start by finding the Function in the Functions summary list, and then invoking the pop-up menu at the right of the Function’s row select the menu item labeled Delete.

Before you can delete a specific Function, you must first ensure that it is undeployed. Only Functions that are in an undeployed state can be deleted.

Click Confirm to delete Confirm Delete Function fly-out menu.