Tara Calihman - January 23, 2014
The VictorOps Alert Ingestion API provides a direct connection into the timeline and the ability to create informational, warning and critical alerts in our platform. Once an alert has been sent, it has to be acknowledged to stop notifying you. When you use our API, you know you’re going to get an alert. VictorOps will also help you to sort through all of the different types of alerts by displaying them using different templates.
Dan Hopkins knows a lot about our Alert Ingestion API. As he should, since he built it. That being the case, he likes to see how people are using our API as well as how they could use it. This lead us to a creative discussion about our dream user cases…some real, some not yet built, all possible.
Below are some current examples of how we currently rock our Alert Ingestion API:
Bitcoin - Another enterprising hacker decided he wanted to monitor the price of Bitcoin. Who doesn’t, right? When Bitcoin goes up 10% over the 7-day average, he gets an alert created in his VictorOps timeline and our system sends him a notification.
Salesforce - You can read more specific details about the integration work Nate did on this integration, but basically anytime a support issue comes into Salesforce, it creates a critical alert in our company’s timeline, allowing the entire business to know when something has come up on the support front.
New Relic - This integration is under construction. We’re using simple alert proxy concept to take SAAS alerts and send them to our SAAS platform. By converting the NR alerts to VictorOps alerts with a server deployed on Heroku, we were able to quickly get running. Now if only we had time to build out a UI so other people could use it!
Twilio - We currently monitor Twilio using our API to see whether or not we’re sending SMS notifications. Since this is obviously a big part of our service, this integration is important to insure we’re taking care of our customers.
Stack Overflow - Our support team, in another stroke of monitoring genius, created an integration that sends an alert if VictorOps gets mentioned on Stack Overflow. We’re calling it “predictive customer service”.
Sensu - Built by a dev in the community (holler Javier!), this integration takes data from another alerting system and plugs it into VictorOps. Javier used standard Ruby libraries to translate the JSON that Sensu creates and create the fields displayed in the timeline. You don’t need a lot of data in order for an alert to be actionable.
And now, for a few dream examples:
Jenkins - When you break a build, it sends alerts into the VO timeline and it’s gonna notify you (and your boss).
Github - Managing Open Source communities is hard work. Imagine if you got an alert when someone submitted a pull request to Github and then you’re informed of that (and can ACK it) via the VictorOps timeline.
Managing Communications - What if you could use VictorOps to assist with different groups talking to one another and interteam engagement? By setting up different rotations, you could limit the number of people who get distracted and not harass the entire team.
Food / Drink Rotations - People get paged when it’s their turn to pick up lunch or make coffee for the office.
This is just the tip of the API iceberg. Now get cracking and let us know about any other creative uses we might have missed!