Introduction
In this article, we'll demonstrate creating a simple streaming data pipeline using a small micro-batching tool and CockroachDB's CDC Webhook sink.
Setup
Prerequisites:
CockroachDB, with a trial enterprise license.
Pipeline, an open-source Java tool built on top of Spring Batch
Java 17+ Runtime
Linux / macOS
CockroachDB Setup
Initially, create a local cluster of three nodes (or one node, not important):
cockroach start
cockroach start
cockroach start
cockroach init
Next, setup a source database called tpcc and a target database called tpcc_copy:
cockroach sql
cockroach sql
Lastly, load a small TPC-C workload fixture (schema and data) to the source database:
cockroach workload fixtures import tpcc
The objective is to have a few of these tables copied and mirrored in the target database.
Pipeline Setup
Initially, clone the repo and build it locally:
git clone git@github.com:kai-niemi/roach-pipeline.git pipeline
cd pipeline
chmod +x mvnw
./mvnw clean install
The executable jar is now available under the target folder.
Try it out with:
java -jar target/pipeline.jar --help
We are now ready to create cdc2sql jobs for each TPC-C table we want to be streamed from the source to the target database.
We'll use the REST API of Pipeline. If you remember, REST is really about following links and POSTing forms (rather than concatenating URIs). The tool provides a real hypermedia-driven API, but we'll not going to build any smart client app for it, just use cURL instead.
First off, we get a form template for each table. The form is going to be pre-populated with SQL statements for the table. We are only using a subset of the TPC-C workload tables, but the process is the same for all tables *).
*) There's also an option to get a form bundle with all tables in one go, but it's WIP.
curl -X GET http://localhost:8090/cdc2sql/form?table=warehouse > warehouse-cdc2sql.json
curl -X GET http://localhost:8090/cdc2sql/form?table=district > district-cdc2sql.json
curl -X GET http://localhost:8090/cdc2sql/form?table=customer > customer-cdc2sql.json
Feel free to inspect the JSON files which should give you an idea of how the batch jobs are configured and run. At this point, we haven't started anything yet. The JSON files typically don't need much editing if the template settings are properly set (everything defaults to using localhost).
Submit Batch Jobs
The next step is to POST the forms back, which will register the jobs and start them up. The jobs need to be registered in the sorted topology order of the foreign key constraints (warehouse <- district <- customer) since we'll be creating tables on-the-fly.
curl -d "@warehouse-cdc2sql.json" -H "Content-Type:application/json" -X POST http://localhost:8090/cdc2sql
curl -d "@district-cdc2sql.json" -H "Content-Type:application/json" -X POST http://localhost:8090/cdc2sql
curl -d "@customer-cdc2sql.json" -H "Content-Type:application/json" -X POST http://localhost:8090/cdc2sql
Take note of the CREATE CHANGEFEED statements in the responses. We will need them for the next step, which is to configure the change feeds.
Connect to the source database and execute (after changing URIs):
CREATE CHANGEFEED FOR TABLE warehouse
INTO 'webhook-https://localhost:8443/cdc2sql/5803c5a2-707a-4fb1-8faf-615d95896664?insecure_tls_skip_verify=true'
WITH updated, resolved='15s',
CREATE CHANGEFEED FOR TABLE district
INTO 'webhook-https://localhost:8443/cdc2sql/5803c5a2-707a-4fb1-8faf-615d95896664?insecure_tls_skip_verify=true'
WITH updated, resolved='15s',
CREATE CHANGEFEED FOR TABLE customer
INTO 'webhook-https://localhost:8443/cdc2sql/5803c5a2-707a-4fb1-8faf-615d95896664?insecure_tls_skip_verify=true'
WITH updated, resolved='15s',
You should now see the target database starting to fill up and eventually reach the same state as the source database. If you would also run the TPC-C workload, you will see any changes reflected also in the target.
Conclusion
In this article, we looked at creating a simple streaming data pipeline at table level between two separate CockroachDB databases using the CDC webhook sink.
