June 2014 Summaries

The addition of Agent check status information to the Datadog Infrastructure page allows users to monitor all parts of their infrastructure, ensuring no gaps in performance monitoring. This feature provides at-a-glance visibility into integration issues with hosts monitored by Datadog, enabling users to identify configuration problems and resolve them quickly. With this new feature, users can now enable integrations without worrying about missing important metrics, providing detailed logging to pinpoint the issue. The feature is available for hosts with Agents v4.3+ installed, allowing users to take advantage of improved monitoring capabilities.

Amazon Web Services (AWS) introduced CloudTrail, a log of key events and configuration changes for AWS services, during re:Invent 2013. This log can serve as an audit trail to enforce compliance or as a data source to understand application and infrastructure performance. Datadog has integrated with AWS CloudTrail to simplify the process of extracting and correlating these events with other performance metrics. The integration collects a subset of CloudTrail logs from S3, indexes them, and tags them with metadata inherited from AWS for easier identification. Users can search for specific events using sophisticated queries and overlay AWS CloudTrail events on any graph to look for correlation between changes in the AWS environment and changes in application performance. Datadog also provides a platform for collaboration through its Events Stream, allowing users to start conversations or broadcast notifications via messaging systems like PagerDuty and OpsGenie or chat platforms like HipChat and Slack.

Amazon Web Services released CloudTrail, a log of key events and configuration changes of AWS services, which can be used as an audit trail to enforce compliance and as a data source to understand application and infrastructure performance. CloudTrail logs are stored in S3 and need extra work to extract these events and correlate them with other performance metrics. Datadog provides a new integration with AWS CloudTrail, making this process easy and enabling highly visual correlation analysis of CloudTrail events with performance metrics. The integration collects a subset of CloudTrail logs from S3, which are then available for teams to view and comment on in Datadog's Events Stream, indexed and tagged with metadata inherited from AWS, allowing for search and overlaying of AWS CloudTrail events on any graph for any metric to look for correlation between changes in the AWS environment and application performance. The platform also provides a collaboration feature through its Events Stream, enabling teams to start conversations by mentioning another user's name or broadcasting notifications through messaging systems like PagerDuty and OpsGenie or chat platforms like HipChat and Slack.

The text discusses the integration of Docker and Datadog, highlighting how Docker's philosophy of isolating applications with containers can benefit monitoring. The traditional Datadog setup involves installing the Agent on a server, while the Docker-ized approach uses containers to isolate the Agent from the rest of the system. This allows for efficient monitoring of containerized applications without requiring a full virtualization setup. The integration enables users to monitor custom metrics and send data to Datadog, with options for alerting and visualizing Docker metrics through a free trial.

Docker is an emerging platform that allows developers to build and deploy software using lightweight, pared-down virtual machines known as containers. The platform provides easy-to-provision recipes for developers and bit-for-bit compatibility between environments, making it a popular solution for continuous delivery in modern infrastructure. Datadog has integrated with Docker, allowing users to monitor containers by running the Datadog Agent on the host machine or using a "dockerized" version of the Agent that runs all software in containers. The integration provides tags-based monitoring, which enables users to track specific metrics for many containers in aggregate and create graphs to visualize this data. Additionally, Datadog's alerts feature allows users to define multi-alerts that span clusters of containers, making it easier to monitor large numbers of containers efficiently. Users can also explore Docker metrics using the Metrics Explorer and apply search filters to find specific events or metrics.

Adrian Vaschalde highlights the utility of Datadog's integration with Bitbucket, which aids development teams in quickly identifying issues caused by code changes by correlating commits with performance metrics. This integration enables the visualization and monitoring of code changes, allowing teams to filter and overlay Bitbucket events on timeboards to spot performance issues or verify the effectiveness of code fixes. It also facilitates collaboration on commits, issues, and pull requests directly within Datadog’s event stream, with the ability to tag and link back to Bitbucket for detailed examination. Notifications and comments can be shared across platforms like PagerDuty, OpsGenie, HipChat, and Slack, ensuring seamless communication among team members. To utilize this functionality, users must set up the integration by specifying repositories and actions to monitor, and adding a webhook URL to Bitbucket. For those interested in assessing the impact of code changes, Datadog offers a free trial to get started.

Alerting is crucial for monitoring applications and infrastructure. Datadog's alerts notify users when metrics cross predefined thresholds, allowing them to respond promptly to issues. Two new features have been introduced: re-notifications and muted notifications during maintenance windows. Re-notifications enable users to receive periodic reminders about unresolved issues until they are fixed. Muted notifications allow users to silence alerts temporarily during scheduled maintenance periods, resuming normal notification settings once the maintenance window has passed.

Datadog is introducing two new features to improve alert notifications: re-notifications, which allow users to receive periodic reminders until an issue is resolved, and muted notifications during maintenance windows, which reduce alerting noise by silencing alerts for preset durations. These features aim to provide better control over alert notifications, enabling users to anticipate issues before they develop or respond quickly to isolated incidents.