Data Watch is a powerful tool to automate your alerts. Use this page to help identify how you want to monitor your different types of data feeds!
Critical Data Feed Use Case
You have a steady data stream that needs to be coming in all the time. The amount of data doesn’t fluctuate too much based on time or day. You want to ensure this data keeps coming in multiple times a day.
- Percent Drop in Hosts:
- 50%
- Present Drop in Events:
- 50%
- Schedule:
- Run on Chron Schedule:
- 7 */6 * * *
- Time Window:
- Past Day
- Time Span:
- Hour
Summary: Every 6 Hours on the 7th Minute, we will inspect to see if the volume of data or number of sources reporting data has been cut in half (or worse!) based on averages. This will alert if the data behaves oddly within the day.
Heart Beat Check
You have a data stream that you need to ensure is coming in for your weekly reports. While important data, it’s not high in volume, and isn’t too predictable.
- Percent Drop in Hosts:
- 100%
- Present Drop in Events:
- 100%
- Schedule:
- Run Every Week
- Monday at 7 AM
- Time Window:
- Past Week
- Time Span:
- Day
Summary: Every Monday morning, we will check to ensure that data has been ingested from the past week. If this alert triggers, we know that the data stream has been turned off or cut, and that any reports utilizing it will be inaccurate.
Morning To-Do List
Your Data Stream is populating important dashboards that are used throughout the day, so you want to make sure you are the first person to know if there’s any strangeness in this predictable data set.
- Percent Drop in Hosts:
- 30%
- Present Drop in Events:
- 30%
- Schedule:
- Run on Cron Schedule
- Monday at 7 AM
- Time Window:
- Past Day
- Time Span:
- Hour
Summary: Every Morning, we are validating that data is coming in as expected from an important data source. If the amount of data, or number of hosts reporting data, ever drop below 30% of baseline, then we will be notified at the start of the day so we can fix this before it becomes an issue down stream.
