What to do with an operations alert can sometimes be a daunting task. Some of our customers opt to get started with email alerts, the easiest and quickest way to consume alerts. But with email alerting, if not granularly customized, can often result in email fatigue. On the other hand, some customers will put significant investment in ingesting alerts and integrating test data into their correlation engines using webhooks or the ThousandEyes API.
Pull | Push | Native
Beyond email, ThousandEyes customers have three main ways to raise awareness about an operational problem. Customers can leverage the ThousandEyes API to query (pull) active alerts that ThousandEyes picks up as anomalies. With this approach, the anomaly detection is done by ThousandEyes and surfaced as an active alert, which is queried by the customer. Under the pull approach, customers may also choose to query raw test data in form of performance metrics and apply logic outside of ThousandEyes to govern what constitutes a problem.
Under the push approach, customers will utilize our webhooks that send alert metadata in the form of a JSON response to a HTTP callback server that processes the alert and POSTs the alert to another application. For instance, customers have used webhooks in ThousandEyes for generating a ticket in ServiceNow or JIRA, sending test data into Datadog for event correlation, and sending alerts into a IM chatroom.
We’ve also worked towards building native integrations with commonly used third-party applications to help quickly operationalize ThousandEyes.
How RingCentral Cloud Ops Use ThousandEyes
Last year, we announced RingCentral, a leading provider of global enterprise cloud communications and collaboration solutions, selected ThousandEyes as part of its strategic business initiative to enhance digital experiences for its customers. From the outset, we worked with the RingCentral team to operationalize ThousandEyes. This involved using the API to query active alerts that fed into Kibana, the visualization plugin for elasticsearch, and also their BMC based NOC monitoring console.
Integrating with Glip
The operational plan also included integrating alert messages into Glip, a team messaging and collaboration application that’s part of the RingCentral UC suite. Glip allows users to manage information in one place allowing multiple teams to efficiently deliver on collaborative efforts. Troubleshooting a production issue may be a multi-team effort and having a shared chatroom for ThousandEyes allows team members to quickly receive notification and provides a seamless way to launch into a RingCentral meeting for further collaboration.
Glip makes it easy for any external application to integrate with the messaging platform. Glip admins can create a webhook URL which is used as the HTTP POST target by the ThousandEyes alerts engine. When ThousandEyes picks up a production problem based on the configured alert rules, it sends an alert in the form of an HTTP POST that gets translated by a webhook server and sent to a Glip webhook URL.
The combined power of ThousandEyes webhooks and easy to use Glip integrations allows production teams to be more efficient at handling escalations, but more importantly, gives cloud operations teams time back to use on projects that move the needle towards scaling the business.