February 2021 Summaries

Isaac Kim, an Enterprise Account Executive at Grafana Labs, balances his passion for sales with a deep appreciation for technology and customer interaction. Based in Austin, Texas, Isaac transitioned from marketing technology sales to Grafana Labs, seeking a more technically challenging and enterprise-focused environment. He enjoys the rewarding experience of helping diverse customers enhance their observability strategies without the need for complete system overhauls, embracing Grafana's inclusive approach to integrating tools like Elastic and Datadog. In his free time, Isaac indulges in cooking, supporting Liverpool FC, and exploring photography with his Fujifilm Instax Mini 11. Despite the pandemic curtailing his travel plans, he looks forward to visiting the Amalfi Coast and appreciates Korean cuisine, particularly his family's traditional stews. Isaac's hidden talents include playing multiple instruments and having performed at the Notre Dame Cathedral, reflecting his multifaceted interests beyond his professional life.

The GitLab data source plugin for Grafana provides a tool for users with a Grafana Cloud account or a Grafana Enterprise license to visualize and transform GitLab activity data into meaningful insights. By leveraging the GitLab API, users can track detailed stats, create annotations to correlate events with other metrics, and utilize template variables to parameterize dashboards, enhancing data analysis and troubleshooting capabilities. Despite some limitations such as API rate limits and the absence of a cache mechanism, the plugin allows for data manipulation through Grafana's transformation features, offering potential advancements in alerting and aggregations. The plugin's beta release supports a subset of GitLab API resources, with plans for future enhancements, making it a valuable tool for integrating GitLab activity with broader observability practices.

Introduced in Grafana v7.0, transformations allow users to manipulate data directly in the browser before visualization, offering capabilities such as joining separate time series, performing mathematical operations across queries, and manipulating both time series and static relational data. Key transformations include the join transformation, which consolidates data for improved readability and downstream calculations, and the group by transformation, which distills patterns by collapsing data into manageable subsets based on column values. The organize fields transformation enables users to reorder, rename, and hide fields to enhance data visualization, while the sort by transformation allows for sorting data in ascending order. Users can layer multiple transformations, though the order of application affects the final output, and Grafana provides tools like the transformation debug view to facilitate understanding. New transformations in version 7.4 are available, and Grafana Cloud offers a streamlined way to start with various data tools, featuring a new free plan and upgraded paid plans.

Grafana Enterprise Logs, recently introduced as part of the Grafana Enterprise Stack, is designed for large organizations to manage observability at scale on their own infrastructure. Leveraging the Grafana Loki open-source project, it offers a distinctive approach to log indexing, storage, and administrative control, enabling secure and scalable operations. A brief demo highlights the ease of creating and managing instances and access policies using its management interface. For further exploration, a webinar is scheduled for March 18, alongside documentation for those interested in trying out Grafana Enterprise Logs.

The Splunk Infrastructure Monitoring plugin, which integrates the monitoring capabilities of what was formerly known as SignalFx, is now available for Grafana users with a Cloud account or Enterprise license, allowing them to visualize Splunk and other data sources side-by-side through Grafana dashboards. This plugin is particularly useful for businesses aiming to maintain high availability targets, like a "four nines" SLA, as it provides real-time monitoring and metrics for cloud infrastructure, microservices, and applications, offering high-resolution metrics and real-time analytics. Users can configure the plugin by downloading it from Grafana's plugins page, setting up their data source with a realm name and SignalFx access token, and using SignalFlow queries to populate data. The integration simplifies monitoring Kubernetes clusters and other infrastructures by enabling the use of pre-existing SignalFlow queries and adding variables for filtering data, which helps in creating comprehensive dashboards that can include metrics from Prometheus and SignalFx side-by-side.

During the pandemic, David Kaltschmidt, a director at Grafana Labs, shares his experiences managing engineering teams while maintaining work-life balance. Grafana Labs, with its Scandinavian roots, emphasizes employee well-being, which has been crucial during the pandemic's challenges. The company has introduced initiatives like health/work-life balance OKRs, company-wide shutdown days, and mental health support to help employees manage stress. Kaltschmidt highlights the importance of flexible working hours, advocating for reduced stress through time management hacks, such as committing to a minimum number of work hours and adjusting meeting schedules to prioritize focus time. He describes the challenges of remote onboarding and the efforts to integrate new employees through structured plans and virtual interactions. Emphasizing quality output over hours, Kaltschmidt encourages managers to be mindful of their team's mental health and to adapt their management style to the exceptional circumstances presented by the pandemic.

Grafana Cloud offers a comprehensive observability solution that integrates Prometheus for metrics, Loki for logs, and Tempo for tracing, all managed by Grafana Labs, providing an easy setup for users to create detailed graphs, logs, and alerts quickly. Despite requiring users to configure Prometheus or the Grafana Agent to collect data, Grafana Cloud delivers significant benefits, including scalable querying and storage with sub-second latency, a unified global data view, and high reliability even during cluster outages. This service alleviates the complexities associated with managing Prometheus, such as lifecycle management and data aggregation across multiple servers. While some users still maintain a local Prometheus for backup and cost-saving reasons, Grafana Cloud's robust infrastructure, enhanced by its use of Cortex, makes it an attractive option for those seeking efficient and reliable monitoring without the burden of managing on-premises resources. Additionally, new free and enhanced paid plans are available, making it accessible for a wide range of users.

Grafana Enterprise Logs is a new product launched by Grafana Labs as part of their integrated Grafana Enterprise Stack, designed to provide scalable, cost-effective log management for enterprises using the open-source Loki project. This solution addresses the challenges of managing large volumes of logs in cloud-native and microservice-based architectures by indexing metadata rather than log lines, ensuring fast and scalable log aggregation. It allows seamless integration with Prometheus for consistent labeling and easy switching between metrics and logs, preserving context and saving time. Developed with enterprises in mind, Grafana Enterprise Logs offers enhanced security features, access controls, and support guarantees, making it suitable for organizations with strict security policies or regulatory requirements. It also supports direct integration with existing logging tools and provides a familiar interface for users experienced with Prometheus, facilitating easy adoption without extensive retraining. Moreover, Grafana Labs remains committed to the open-source Loki project, ensuring compatibility and continued development alongside its enterprise offerings.

The article provides insights into configuring Loki, a log aggregation system, to prevent log loss and ensure reliability in its write path. It highlights the importance of using a replication factor, typically set to three, to store multiple copies of logs in the ingester component, ensuring data redundancy and minimizing the risk of data loss if an ingester fails. Key components such as the distributor, ring, and ingester are discussed, emphasizing their roles in log ingestion and storage. Additionally, the article recommends essential configuration settings, such as heartbeat timeout, chunk target size, and chunk encoding, to optimize performance and durability. The introduction of the Write Ahead Log (WAL) in Loki version 2.2 is also covered, offering an additional layer of protection by writing incoming data to disk before acknowledging it, with suggestions on isolating WAL storage to enhance reliability. The article concludes by promoting Grafana Cloud as an accessible platform for deploying Loki alongside other tools like Grafana, Prometheus, and Tempo, offering new free and upgraded paid plans for users.

Sam Coren, the Community Manager at Grafana Labs, plays a pivotal role in engaging with and supporting the vibrant Grafana community, which includes over 600,000 active installations worldwide. Based in Boston, Coren's responsibilities involve facilitating communication through forums, Slack, and social media, and collaborating with the marketing team to highlight community projects. With a passion for music production and teaching, Coren balances professional life with personal interests, such as visiting New England beaches with a dog and indulging in electronic music and drumming. Coren's past experiences at DigitalOcean underscore the importance of Grafana in monitoring data centers, while the dedication to community-driven development remains a core aspect of Grafana Labs' operations. The diverse applications of Grafana, from IT infrastructure to beekeeping, are a source of motivation and pride, reflecting the unique ways users leverage the platform.

Marty Disibio's guide explores the process of instrumenting a .NET 5.0 web API using OpenTelemetry, Tempo, and Grafana Cloud to enhance observability by standardizing logs, metrics, and traces across platforms. By setting up a web API project and integrating OpenTelemetry nuget packages, users can trace HTTP requests, with traces initially logged to stdout and then uploaded to Grafana Cloud using a Grafana Agent configured via docker-compose. The guide details sending application logs to Grafana Cloud using the Loki log driver, recommending Promtail for production workloads in Kubernetes, and illustrates how to connect logs and traces using Loki's derived field functionality. This setup allows users to view logs and traces side by side in Grafana Cloud, offering a comprehensive solution for monitoring and observability, adaptable for .NET 3.1 and supported by Grafana's new free and paid plans.

Grafana 7.4 introduced the Time series panel, an evolution of the graph panel architecture from Grafana 7.0, offering significant performance enhancements and new visualization features. This panel, still in beta, integrates with field options, overrides, and transformations while transitioning from the jQuery Flot library to the high-performance uPlot library. It allows for seamless migration of existing graph functionalities and introduces new features such as smooth line interpolation, auto point display based on data density, gradient fill modes, and multiple y-axes, addressing popular community requests. The Time series panel's architecture lays the groundwork for future high-performance visualizations like bar charts and scatter plots, with plans for further functionality in Grafana 8.0. Users are encouraged to explore these features through a demo dashboard and can easily start using Grafana Cloud, which now offers a free plan and upgraded paid plans.

Grafana is offering a limited-edition Observability Sticker Pack as a way to engage with its community, particularly in light of the absence of in-person events. To obtain these stickers, participants need to express their appreciation for Grafana on Twitter or LinkedIn using the hashtag #GrafanaLove and submit an entry form by February 15 at 8:00 UTC. This initiative aims to strengthen community ties and share some love during Valentine's Day, but the sticker supply is limited, and interested individuals are encouraged to act quickly to avoid missing out.

Matthew Helmke's blog post details his experience of monitoring an OpenWrt router using Grafana Cloud and Prometheus. By running OpenWrt on his Linksys WRT1900AC router, Helmke gains enhanced stability and customization options not available with stock firmware. To monitor the router's performance, he utilizes Prometheus node exporter packages available through OpenWrt's package manager, installing them via the command line to expose metrics. These metrics are made accessible on his local network and are then scraped by Prometheus running on a separate Linux workstation. The data is subsequently forwarded to Grafana Cloud using Prometheus's remote write feature, allowing Helmke to visualize the data through a dashboard. Despite some limitations with storage and memory on consumer-level devices, he notes potential improvements for those with more robust hardware, such as using full Prometheus node exporters. This setup, though time-consuming due to initial configuration challenges, ultimately provides a comprehensive monitoring solution for his network.

The Splunk plugin for Grafana has been updated to version 2.1.0, introducing significant enhancements such as support for data links, which allow users to efficiently navigate from logs to traces within Grafana, thereby streamlining data analysis and troubleshooting processes. This new feature is compatible with tracing backends like Tempo, Jaeger, and Zipkin, enabling seamless integration and easier access to trace data without leaving the Grafana interface. Data links facilitate the exploration of relevant traces by parsing trace IDs directly from log messages, which enhances the capability of Grafana to provide a unified view of data across various sources, democratizing data access and improving issue resolution efficiency. Additionally, the update extends the capability of Grafana's existing integration with other data sources like Loki and Elasticsearch, making it a cost-effective solution for storing and accessing 100% of traces, particularly when using Grafana Tempo as a backend. This development allows users to maintain their existing application instrumentation while optimizing storage costs and efficiency, offering a robust solution for modern observability needs.

Pratik Thanki, a software engineer, explores how he uses Grafana Cloud for monitoring his NFL play-by-play API, FourthDown, which allows fans to access a wide range of NFL data. Developed to be language-agnostic and built on ASP.NET Core, the API leverages Prometheus and Jaeger for observability, utilizing logging, tracing, and metrics to monitor and improve performance. Following the launch of a new free plan, Thanki implemented these tools, identifying and resolving performance bottlenecks, leading to significant response time improvements. He expresses enthusiasm for the open-source observability ecosystem and plans to expand monitoring capabilities with Prometheus-based alerting and Loki for logging, inviting others to explore the API through provided documentation and samples.

Grafana 7.4 introduces several notable updates, including a next-generation graph panel named "time series," which supports high-performance visualization and is in beta. This panel utilizes a new architecture and aims for full feature coverage by Grafana 8.0, offering enhanced interpolation modes, axis control, and live streaming at 30 fps. The release also adds Prometheus exemplar support, enabling quicker trace discovery, along with a variable inspector to aid in understanding dashboard variable usage. Other features include a new node graph panel for visualizing complex networks, new data transformations for sorting and filtering, trace-to-log navigation, server-side expressions for data manipulation, and enhanced variable support in alert notifications. Additionally, Elasticsearch plugin updates improve alerting and aggregation, and Grafana Enterprise users can now export usage insights to Loki for detailed analysis. The release encourages users to explore these features through demos, documentation, and a new free plan on Grafana Cloud.

Auto-instrumenting a Java Spring Boot application with OpenTelemetry and Grafana Tempo involves using OpenTelemetry auto-instrumentation libraries to manage traces and logs effectively, despite the author's limited experience with Java. By referencing a specific jar file and setting appropriate parameters, traces are offloaded to Tempo, while logs are sent to Loki, allowing for efficient trace discovery and management. The process also includes configuring logging to include trace IDs, which aids in correlating logs with traces. Although the current setup allows for basic trace and log handling, there is potential for further enhancements, such as implementing a custom request logger to improve log querying capabilities. The article also encourages readers to explore the provided repository and learn more about Grafana Tempo through available resources.

Paolo Patierno, a Principal Software Engineer at Red Hat, explores the integration of various open-source technologies to process and monitor Formula 1 telemetry data in real-time using a cloud-native environment. Leveraging tools like Apache Kafka, Strimzi, Grafana, and InfluxDB, the article outlines how simulated F1 telemetry data from the F1 2020 game can be ingested, processed, and visualized. The system uses Apache Camel to route data from UDP to Kafka topics, and the data is then stored in InfluxDB for visualization in Grafana. This setup allows for the real-time analysis of various telemetry metrics such as car speed, engine performance, and driver behavior, providing valuable insights for engineers to enhance vehicle performance. The project underscores the potential of open-source technologies in creating complex, scalable analytics pipelines and is documented in a GitHub repository for public access.

Grafana Labs has announced the transition from its original monitoring tool, worldPing, to a more advanced feature called synthetic monitoring within Grafana Cloud. This new application, built on the Prometheus blackbox exporter, offers enhanced capabilities such as advanced alerting, expanded monitoring options like SSL certificate expiration and network checks, and integration of metrics and logs for comprehensive troubleshooting. While worldPing will be deprecated on April 1, 2021, synthetic monitoring provides a seamless upgrade with its free tier offering, allowing users to monitor services without cost barriers. Additionally, the new monitoring tool emphasizes user experience by simplifying the creation of checks and alerts and offers a pricing model based on active series rather than a flat rate per check. Grafana Labs encourages users to migrate to synthetic monitoring to benefit from its robust features and ensure continued service monitoring.