How To Create Alerts in GoodData

If you wish to set alerts for specific metrics or visualizations in GoodData, this tutorial will guide you through the process of using the GoodData Python SDK. Alerts enhance your monitoring capabilities for critical metrics and automate the notification process, which can assist you in preventing serious issues. With the flexibility of the GoodData Python SDK, you aren't limited to just one predefined solution, so you can customize it to your heart's content.

You can check the whole repository here.

Prerequisites:

• GoodData Cloud and GoodData API token
• S3 bucket
• SMTP server to send emails (for example, Gmail)
  • What Is Gmail SMTP and How to Use Gmail With My Domain?
• We reccomend you to use Python virtual environment

Disclaimer:

• It works only with visualization type Headline

Step 1: Install dependencies

# install all dependencies defined in requirements.txt 
$ pip install -r requirements.txt

Step 2: Setup all environment variables

export GOODDATA_HOST='<gooddata-host>'
export GOODDATA_TOKEN='<gooddata-token>'
export GOODDATA_WORKSPACE_ID='<gooddata-workspace-id>'
export S3_ACCESS_KEY_ID='<s3-access-key-id>'
export S3_SECRET_ACCESS_KEY='<s3-secret-access-key'
export S3_BUCKET_NAME='<s3-bucket-name>'
export EMAIL_PASSWORD='<email-password>'

Step 3: Define thresholds

visualizations:
    - id: <visualization-id> # id of visualization
      column: <entity-id> # the number to watch, it is the name of column returned from GoodData pandas dataframe
      threshold: <threshold> # threshold number, for example 300
      threshold_type: uptrends # threshold type can be 'uptrends' or 'downtrends'
    - id: <visualization-id> # id of visualization
      column: <entity-id> # the number to watch, it is the name of column returned from GoodData pandas dataframe
      threshold: <threshold> # threshold number, for example 4
      threshold_type: downtrends # threshold type can be 'uptrends' or 'downtrends'

Step 4: Define time to run alerting

In the gooddata_alerting.py, you can adjust your preferred time to run alerting.

schedule.every().day.at("08:00").do(main)

while True:
    schedule.run_pending()
    time.sleep(1)

Step 5: Run it! 🚀

$ python gooddata_alerting.py

How does it work?

The best demonstration of the whole functionality is the following script:

for new_id, new_item in new_dict.items():
    old_item = old_dict.get(new_id)

    if old_item:
        if new_item.notify and not old_item.notified_before:
            notification.send_email("Alert!", f"Visualization with id {new_item.id} has reached threshold!")
            new_item.notified_before = True

        if old_item.notified_before and new_item.notify:
            new_item.notified_before = True

        if old_item.notified_before and not new_item.notify:
            new_item.notified_before = False
    else:
        if new_item.notify:
            notification.send_email("Alert!", f"Visualization with id {new_item.id} has reached threshold!")
            new_item.notified_before = True

• new_dict contains all the information about the current 'numbers' from GoodData.
• old_item contains all the information about past 'numbers' from GoodData, for example, from yesterday.
• If there are no past 'numbers,' the script sends a notification when a visualization reaches a defined threshold.
• If there are past 'numbers,' the script checks whether a visualization has reached a defined threshold; if so, it will send a notification.
  • If a notification has already been sent, the script does nothing.
  • If a visualization has dropped below a defined threshold, the script resets the notification mechanism (to send a notification once the threshold is reached again).

Conclusion

If you have any questions or require assistance, don't hesitate to reach us on the GoodData Slack channel. Willing to try GoodData? Explore the possibilities by signing up for our free trial today.