November 2018 Summaries

The Amazon CloudWatch provides standard metrics for serverless applications on AWS Lambda, but custom metrics can provide additional visibility into unique business use cases and application performance. The Datadog Lambda extension enables developers to collect and forward custom metrics, traces, and logs directly to Datadog, allowing for enhanced monitoring and troubleshooting capabilities. With the extension, users can instrument their functions to send custom metrics, generate metrics from logs and traces, and rapidly identify and troubleshoot business-impacting issues. By leveraging custom metrics, developers can create useful dashboards, monitors, and SLOs to visualize and alert on data specific to their business, ultimately gaining full visibility into their serverless applications and infrastructure.

Envoy is an open-source proxy developed to simplify microservices communication by serving as the data plane in a service mesh, facilitating data routing, load balancing, health checking, and service discovery. It can be integrated with Datadog, along with AWS App Mesh, a managed control plane that dynamically configures these proxies, enabling comprehensive monitoring of service mesh performance. Datadog integration allows for the collection and visualization of various metrics, including request counts and error rates, and supports distributed tracing by adding unique identifiers to service calls, providing detailed flame graphs and service maps to trace request flows. This integration enhances observability and management of microservices architectures, offering insights through visualizations and alerts on service performance, Envoy activity, and application metrics, with the option to create custom filters for additional data collection.

Datadog has announced the availability of Synthetic Monitoring, a new layer of visibility on its platform that allows users to monitor applications and API endpoints via simulated user requests. This feature helps ensure uptime, identify regional issues, track application performance, and manage SLAs and SLOs by unifying it with metrics, traces, and logs. Synthetic Monitoring provides proactive user experience monitoring through Datadog Synthetic API tests and browser tests, which can simulate different types of traffic such as HTTP, DNS, TCP, SSL, and ICMP. The feature also offers end-to-end visibility for validating that users can perform critical business transactions, and minimizes alert fatigue by allowing finely tuned alerts based on synthetic test data. With the integration with Datadog APM and collaboration tools like Slack and PagerDuty, Synthetic Monitoring provides a simulated user's view of applications, enabling high-level performance insights to quickly jump-start investigations.

The Datadog platform has announced the general availability of its serverless monitoring feature, allowing users to search, filter, and explore all their AWS Lambda functions in one place. This feature provides detailed performance data for each function, including metrics, logs, and distributed traces. With this addition, customers can troubleshoot performance issues with distributed traces, collect custom metrics in real-time from serverless functions, and visualize the bigger picture in the Service Map, which visualizes Lambda functions along with their dependencies. The platform also enables tracing of Lambda functions with X-Ray, allowing users to zero in on the source of errors or slowdowns and see how performance impacts user experience. Additionally, customers can track business goals with real-time custom metrics and instrument Lambda functions for tracing without requiring code changes.

AWS re:Invent is an annual event in November that attracts numerous AWS staff, partners, and users to Las Vegas for a week-long immersion in AWS-related activities. Datadog, a prominent participant, invites attendees to explore new features and engage with its engineers for monitoring insights. The event's core lies in its extensive sessions, which cover a range of topics such as leveraging AWS for gaming with Activision Blizzard, chaos engineering, machine learning with Amazon SageMaker, and serverless architecture. Datadog co-presents sessions highlighting its collaborations with companies like Coinbase and Trek10, focusing on reliability engineering and distributed tracing. Additionally, Datadog participates in the AWS re:Invent GameDay, offering teams a platform to tackle real-world challenges using AWS tools. Attendees are encouraged to explore the diverse session schedule and visit Datadog's booths for further engagement.

In this guest post by Priya Matpadi, Principal Engineer at Lookout, we learn about implementing monitoring for a Spark streaming application running on Amazon EMR using Datadog. The author explains why they chose to use Datadog and how they integrated it with EMR. They also provide scripts for installing the Datadog Agent on EMR nodes and configuring the Spark check. Finally, the author discusses monitoring custom application metrics in Datadog and creating a comprehensive dashboard for visibility into the health of their Spark streaming application.

The authors of the text describe how they implemented monitoring for a Spark streaming application running on Amazon EMR. They chose to use Datadog, which provides built-in integrations with both EMR and Spark. The authors set up the Datadog integration with EMR by linking their AWS account and ensuring the relevant permissions were in place. They then installed the Datadog Agent on each node of the EMR cluster using a bootstrap action script, configured the Agent to run the Spark check at regular intervals, and published Spark metrics to Datadog. To collect custom application metrics, they instrumented their application code to publish metrics as events are processed by the application. The authors believe that monitoring their Spark streaming application from all sides will provide them with visibility into its health and performance.

Datadog uses Amazon Simple Email Service (SES) to send password reset emails and other important messages. To ensure high trust in these communications, they use a managed service from their cloud provider and leverage Datadog's Log Management product for observability. They create an SNS topic that triggers a Lambda function whenever an email event is generated on Amazon SES. The Lambda function transfers the email event to Datadog Log Management. This setup allows easy access to email-related metrics, which helps support teams troubleshoot login issues and provide better assistance to customers.

Datadog uses a managed service from one of its cloud providers, Amazon Simple Email Service (SES), to send password reset emails. To accommodate the delivery and analytics for this volume of outbound emails, Datadog uses a third-party mail provider and relies on Datadog's Log Management product to make it easy to search password reset emails via predefined facets. The solution involves creating an Amazon SES configuration set that publishes email events to an SNS topic, which triggers a Lambda function that transfers the event to Datadog Log Management. This setup provides simple, low-maintenance infrastructure for password reset emails with increased trust and built-in observability for the support team.