Export Job Automation

Connector jobs can be automated in multiple ways. This article explores the most common methods in detail.

Automation via Contentserv Active Job

The SAWSConnector suite contains several Contentserv Active Jobs, one such is the “SAWS – Start connector job” job.

This job allows you to start a connector job. Note that this Active Job only acts as a “kick-off”, it only starts the connector job and does not monitor or wait for it to finish.

Using the automation tab of the Contentserv Active Jobs, the automation can be timed. If an incremental/delta connector export is used, most user let it run hourly, bigger exports daily or weekly.

Incremental export/delta export meaning: An export that only exports the changes since the last export e.g. changed products since the last run.

Automation via Contentserv Workflow Action

Connector jobs can be started when a workflow action like for example Set Productive is executed. To configure this head into the action and select the action component Run SAWSConnector job.

You are then able to select the job that shall be started.

Automation via REST API

Connector jobs can be started via HTTP requests to the REST API. Read more here.

Chaining Connector Jobs

In many instances connector jobs shall be chained. For example the media shall be uploaded, then the categories shall be sent, followed by the products. These jobs shall not run in parallel since the categories and products need the existing media, and the products need the existing categories to link to them. Hence these jobs need to be chained.

To chain jobs the job-related task Run follow-up SAWSConnector job must be selected.

Since the connector version 1.19.0, you are now able to see the chained job from the outer view of the job center.

Do not use the Active Job for chaining!

These jobs would start these 3 jobs simultaneously instead of after one another. If you want to automate your chain, only select the first job of it in the Active Job.

 

Common Automation Use Case

A very common use of the SAWSConnector is to export attributes, categories and products to the target system. For cases like these we recommend the following the automation setup. The thought process will be explained as well, so you can apply it to your automation.

Attributes

The list of sent attributes defines which attributes a product in the target system can have. As this list rarely changes, there is no need for it to be updated in real time.

Categories

Most projects have fewer than 200 categories. Categories are usually small data packages, with the category image being the most significant factor. This means that a delta export is not usually necessary.

Products

Centric PXM systems can contain millions of products, each with many images and large amounts of attribute data. A delta job is useful here to avoid sending huge amounts of redundant data as well as increasing export speeds.

The screenshot shows delta option in the job. The most commonly used options are Load all data sets since the last run and Load all data sets since the last successful run. The second option requires daily oversight to check for export errors, since the delta (all exported differences) will grow larger as long as the error occurs while the job is running, because products are being changed yet the delta comparison date is not being set while the job is running erroneously.

Another popular option is to select Load all changes within the past 2 hours and automate the export hourly. This means that every change is exported twice, providing a sort of soft security.

Note: The connector delta compares the Centric product change date with the job’s delta date. However, the Centric change date is not always the best indicator of changes. For example, an image or an inherited value in the parent can be changed without the product receiving a new change date. The connector is designed to ignore these other changes and only use the change date, because otherwise there would be no delta and every export would be a full export.

Therefore, it is recommended that you also perform a full export of all products from time to time, to ensure that any changes missed by the change date are exported.

Conclusion

Your use case may differ; for example, you may not need to send attributes or a large number of categories, so adjust your export automation accordingly.

For this use case, we recommend creating an hourly product delta export with an Active Job that runs hourly and starts a separate product delta job. This exports all data sets since the last run.

Then, on Mondays at 1 am, schedule a full export of all attributes, categories, and products by creating an Active Job that starts the attribute job, which is then chained to the category job, and then the product job. This ensures that new attributes or categories exist before products are sent because they link the previous two. Product changes that the delta export may have missed are also exported. Sunday or Monday is ideal for a full export because your shop team can respond to any errors that may occur.

Was this article helpful?

1 0

How can we improve this article?

Please submit the reason for your vote so that we can improve the article.

Name

Email

Submit