Skip to main content
  1. Home
  2. Search on GOV.UK
Last updated: 18 Jul 2025

How to resync content in Vertex

To resync content items in Vertex search, publishing-api needs to push the documents to search-api-v2. There are various rake tasks to achieve this.

Resyncing all content

A rake task requeue_all_the_ever_published_things exists to requeue all items for search. This task “requeues” documents rather than “republishs” them to avoid triggering updates in other parts of the system, i.e. email alerts.

As this rake task doesn’t have any arguments it was possible to config a cron job that can be run from Argo CD.

This job will need to be run when the datastore schema is updated.

To find the job in Argo:

  1. Login to Argo
  2. Navigate to the publishing-api app
  3. Find the publishing-api-search-api-v2-bulk-import cron job
  4. Click on the kebab menu next to the job name and click “Create Job”

Image of the bulk import job in Argo CD

Resyncing one content item

To resync a single content item one of the represent_downstream tasks can be used. As the task takes an argument it needs to be run using a kubectl command.

k exec deploy/publishing-api -- \
  rake 'represent_downstream:content_id[57a1253c-68d3-4a93-bb47-b67b9b4f6b9a]'

More in the Search on GOV.UK section

Learn

How to...