Triggering & Scheduling
Last updated
Last updated
Running a sync manually is not that useful on its own. The real power of Census is having your syncs run automatically. Once you've got your sync up and running, you can configure your sync to run automatically in several ways:
After another sync with a Sequence
Schedules let you specify a time and frequency that Census can use to run your sync automatically. You can choose options from weekly all the way to Continuous, which means Census checks your source roughly every minute for new changes.
To remove a schedule from a sync, click the edit icon and select Manual from the drop down list.
The last scheduling option is Cron. Cron lets you schedule syncs on arbitrary schedules such as every 3 hours, or only week days. Census accepts standard Cron definitions up to minute granularity (second-level granularity cron definitions are not supported). A Cron schedule is specified by a series of five values, separated by spaces. In order, the values are Minute-of-hour, hours-of-day, Days-of-month, Months-of-year, and Day-of-week. The timezone is UTC.
Here's a few examples of common Cron schedules:
Once an hour, on the hour, every four hours: 0 */4 * * *
Hourly during weekdays: 0 * * * 1,2,3,4,5
Minutely on the 5th, 6th, and 7th hours of the day in UTC timezone:0 5,6,7 * * * , executes sync jobs at 9pm, 10pm and 11pm PST or 12am, 1am and 2am EST
Please note: If including a range in your CRON schedule you'll want to make sure the values in the range are in a comma separated list.
Example:
Hourly during weekdays should be: 0 * * * 1,2,3,4,5 instead of 0 * * * 1-5
If you're using dbt Cloud to compile your dbt project, Census can trigger syncs whenever one of your dbt Cloud project runs have completed. Simply select your dbt Cloud project's job to monitor and Census will automatically trigger a sync when it completes.
Using dbt Cloud to trigger syncs works great with dbt Models but is not required. Both dbt integrations can be used independently.
To connect Census to your dbt Cloud, you'll first need a dbt Cloud API key. We use this to subscribe to webhooks for each job you specify.
dbt strongly recommends you use a Service Account token, though User API keys with the correct permissions are accepted. Service Account tokens can be created by users with Account Admin (Enterprise plan) or Owner (Team plan) permissions.
The Service Token requires at least the developer permission (for Enterprise) or Member permission (for Team) in order to configure new webhooks. See dbt Cloud docs for more details.
You'll also need to know your dbt Cloud account's region URL. This may be one of the shared regions, or a custom URL specific to your organization. It should be of the form something.getdbt.com (you can skip the https://).
With your token and your region URL in hand, you can now connect dbt Cloud to your dbt project.
Provide your region URL and copy your dbt Cloud API key
Verify your credentials are correct and have the proper permissions, then Save your integration.
Now, you'll be able to use a dbt Cloud job to trigger syncs. Visit the Configuration tab of any of your syncs.
If you use Fivetran to load data into your data warehouse, or make use of their data transformations, you can trigger Census syncs to run once that work has completed.
To connect Census to Fivetran, you'll first need an API key and secret associated with an Account Admin user.
Copy and paste your Fivetran API Key and Secret. Press Verify to confirm they were copied correctly and have the correct permissions. Save your settings.
Now you'll be able to select a Fivetran Connector or Transformation to trigger syncs. Visit the Configuration tab of any of your syncs.
If your syncs have dependencies and you'd like to organize them to run in order, you can use a Sequence. A Sequence runs a dependent sync whenever its specified parent sync completes successfully. Sequences can be found on the sync configurations page:
Sequences do not currently support specifying multiple parent syncs. If you are interested in multi-parent functionality, please email support@getcensus.com.
Each sync can also be triggered via API. On the sync configuration page, you can access the trigger API endpoint for the sync.
An empty HTTP POST call to this endpoint will trigger the sync (no need to provide any data in the body). You can use this API to automatically trigger Census syncs as part of your data pipeline, running syncs once the models they depend on have been rebuilt.
POST /syncs/[ID]/triggerstatus
success or error indicating whether the sync was triggered.
data
Present if successful. An object containing the sync_run_id
message
Present if error. Contains message describing the error.
GET /sync_runs/[ID]You can use the sync_run_id returned when successfully triggering a sync execution and get status on its progress or determine when it has completed.
status
success if sync_run was found
data
Present if successful. Contains the following properties:
status
working if the sync is currently executing
completed if the sync finished successfully
failed if the sync failed during execution
records_processed
Number of new or updated records retrieved from the source
records_updated
Number of records successfully sent to the destination
records_invalid
Number of records skipped by Census because of data quality issues.
records_failed
Number of records rejected by the destination.
Heads up: Unlike Airflow 2, Airflow 1 doesn't show any non-"core" providers (i.e. Census!) in the connections UI. If you're using Airflow 1, Census should be configured as an "HTTP" Conn Type, as documented here.
Whether you're using Astronomer or self-hosting your own instance, you can use Census's Airflow Provider to trigger and monitor Census syncs.
Visit the Census Airflow Provider GitHub repository for more details on how to use it for your project.
The Dagster team maintains our Dagster Census integration for triggering syncs. You can read more in their documentation.
Similarly, Prefect maintains our Prefect Census integration for triggering syncs. You can read more about how it works in our blog post.
