RQ (Redis Queue)
The RQ integration adds support for the RQ Job Queue System.
sentry-sdk from PyPI with the
pip install --upgrade 'sentry-sdk[rq]'
If you have the
rq package in your dependencies, the RQ integration will be enabled automatically when you initialize the Sentry SDK.
Create a file called
mysettings.py with the following content:
# mysettings.py import sentry_sdk sentry_sdk.init( # Set traces_sample_rate to 1.0 to capture 100% # of transactions for performance monitoring. traces_sample_rate=1.0, dsn="https://examplePublicKey@o0.ingest.sentry.io/0", )
Start your worker with:
rq worker \ -c mysettings \ # module name of mysettings.py --sentry-dsn="https://examplePublicKey@o0.ingest.sentry.io/0" # only necessary for RQ < 1.0
The integration will automatically report errors from all RQ jobs.
Generally, make sure that the call to
init is loaded on worker startup, and not only in the module where your jobs are defined. Otherwise, the initialization happens too late and events might end up not being reported.
In addition, make sure that
init is called only once in your app. For example, if you have a
Flask app and a worker that depends on the app, we recommend only initializing Sentry once. Note that because the Flask integration is enabled automatically, you don't need to change the configuration shown above.
# app.py import sentry_sdk sentry_sdk.init( dsn=https://examplePublicKey@o0.ingest.sentry.io/0, # Set traces_sample_rate to 1.0 to capture 100% # of transactions for performance monitoring. traces_sample_rate=1.0, )
The worker configuration
mysettings.py then becomes:
# mysettings.py # This import causes the Sentry SDK to be initialized import app
To verify, create a
main.py script that enqueues a function in RQ, then start an RQ worker to run the function:
# jobs.py def hello(name): 1/0 # raises an error return "Hello %s!" % name
# mysettings.py import sentry_sdk # Sentry configuration for RQ worker processes sentry_sdk.init( dsn=https://examplePublicKey@o0.ingest.sentry.io/0, # Set traces_sample_rate to 1.0 to capture 100% # of transactions for performance monitoring. traces_sample_rate=1.0, )
# main.py from redis import Redis from rq import Queue from jobs import hello import sentry_sdk # Sentry configuration for main.py process sentry_sdk.init( dsn=https://examplePublicKey@o0.ingest.sentry.io/0, # Set traces_sample_rate to 1.0 to capture 100% # of transactions for performance monitoring. traces_sample_rate=1.0, ) q = Queue(connection=Redis()) with sentry_sdk.start_transaction(name="testing_sentry"): result = q.enqueue(hello, "World")
When you run
python main.py a transaction named
testing_sentry will be created in the Performance section of sentry.io and spans for the enqueueing will be created.
If you run the RQ worker with
rq worker -c mysettings, a transaction for the execution of
hello() will be created. Additionally, an error event will be sent to sentry.io and will be connected to the transaction.
It takes a couple of moments for the data to appear in sentry.io.
--sentry-dsn="" to RQ forcibly disables RQ's shortcut for using Sentry. For RQ versions before 1.0 this is necessary to avoid conflicts, because back then RQ would attempt to use the
raven package instead of this SDK. Since RQ 1.0 it's possible to use this CLI option and the associated RQ settings for initializing the SDK.
We still recommend against using those shortcuts because it would be harder to provide options to the SDK at a later point. See the GitHub issue about RQ's Sentry integration for discussion.
- RQ: 0.6+
- Python: 2.7+ (RQ 0.6+), 3.5+ (RQ 1.4+)
Our documentation is open source and available on GitHub. Your contributions are welcome, whether fixing a typo (drat!) to suggesting an update ("yeah, this would be better").