← All platforms

Apache Spark error tracking

A failed Spark job deserves a closer look. Connect the Sentry PySpark integration to Telebugs and keep its Python error reports on your server.

~/hoofnotes  pip install sentry-sdk
~/hoofnotes  nvim pipeline.py
import sentry_sdk
from sentry_sdk.integrations.spark import SparkIntegration
sentry_sdk.init(
# Copy the complete Telebugs DSN, key and path included.
dsn="YOUR_TELEBUGS_DSN",
integrations=[SparkIntegration()],
)
from hoofnotes.pipeline import run
run()
~/hoofnotes  spark-submit pipeline.py

Hang on, I already use Sentry

Keep sentry-sdk and change its destination to your Telebugs DSN. Your error reports stay on your server. You don’t need a Sentry account.

Python or JVM errors?

This connection is for PySpark’s Python errors. JVM exceptions use the Java SDK. Neither connection turns Telebugs into a cluster monitoring service.

Does the driver cover every worker?

No. Workers have a separate SDK integration with runtime restrictions. Sentry marks its Spark integration as experimental; Telebugs accepts the application-error reports it sends.