August 2024 Summaries

PromCon EU 2024 in Berlin will showcase the major release of Prometheus 3.0, featuring new advancements such as OpenTelemetry compatibility, native histograms, and remote write 2.0, with Grafana Labs playing a significant role as a diamond sponsor. The conference will offer various sessions led by experts from Grafana Labs, covering topics like OpenTelemetry integration, UTF-8 support, PromQL query mechanics, metadata management, scalable anomaly detection, and native histograms. Attendees can engage with Grafana Labs' representatives at the event, explore a 3D printer demonstration, and watch live streams if unable to attend in person, with a 5% registration discount available using a specific code.

CockroachDB has been introduced as an Enterprise data source for Grafana, allowing users to securely access the distributed SQL database while utilizing Grafana’s visualization capabilities. Available in private preview, this addition joins over 30 other Enterprise data sources in Grafana's catalog, accessible in both Grafana Enterprise and all tiers of Grafana Cloud, including a free tier. CockroachDB is a cloud-native, distributed SQL database designed for high availability, horizontal scalability, and strong consistency, with features like ACID transactions, geo-distribution, and enterprise-grade security. Users can configure the CockroachDB data source in Grafana using a host URL, default database, and one of three authentication methods, enabling the creation of dashboards to visualize and monitor data effectively. As development continues, Grafana seeks community feedback to shape the future of the CockroachDB plugin, encouraging users to engage via their GitHub repository, and offering a forever-free tier for easy access to the Enterprise data sources.

The Grafana 11.2 release introduces a range of updates across various functionalities, including new data sources, visualization enhancements, and advanced transformation features. Noteworthy additions include the support for new data sources like Yugabyte and Amazon Managed Service for Prometheus, expanded Enterprise data sources, and a public roadmap for data source plugins. Visualization updates include improvements to the Canvas panel tooltips and state timeline pagination, while transformation enhancements bring new functions like data transposition and template variable support. Additionally, Grafana Alerting now offers a centralized alert history view, and Grafana's authentication capabilities have been bolstered with organization mapping integration for OAuth providers and improved SAML integration for Azure AD. The release also introduces a Grafana Cloud Migration Assistant, streamlining the transition from on-premises to cloud with a user-friendly interface. These updates, alongside the active involvement of the Grafana community, continue to advance Grafana's capabilities in data analysis and visualization.

Grafana Cloud provides a user-friendly, cloud-hosted observability platform designed for both novice and experienced users to efficiently collect and visualize telemetry data. It offers over 100 integrations, known as solutions, which include pre-configured Grafana dashboards and best-practice alerting defaults for monitoring various data sources such as Linux hosts, databases, and Kubernetes clusters. Central to these solutions is Grafana Alloy, an open-source distribution of the OpenTelemetry Collector, which collects and forwards telemetry data to Grafana Cloud backends. The platform enables users to quickly set up a Prometheus and Grafana-based observability stack, offering insights through customizable dashboards and alerts for common failure modes. Grafana Cloud's opinionated approach simplifies decision-making, allowing users to focus on gaining valuable insights into the health and performance of their infrastructure, with the flexibility to tailor alerts and metrics to specific needs.

Grafana Cloud Frontend Observability, powered by the open-source Grafana Faro Web SDK, is a managed real user monitoring solution designed to provide comprehensive insights into end-user experiences of web applications. It allows observability teams to monitor web vitals like page load times, interactivity, and visual stability, ensuring optimal user experiences and quick troubleshooting through error aggregation and prioritization. Recent updates include enhanced integration with Grafana Cloud Application Observability for streamlined correlation between frontend and backend signals, capabilities to upload private source maps for secure error context, and cost control features via adjustable telemetry sampling. This integration aids in faster identification of performance issues, such as excessive time-to-first-byte (TTFB) delays, by tracing them back to their root causes within backend services. Additionally, the service provides a unified view of both website and mobile app services, facilitating comprehensive monitoring and resolution of errors while maintaining code privacy through private source map uploads. Grafana Cloud offers easy setup for Frontend Observability with a forever-free tier supporting 50,000 user sessions per month.

Season 2 of the "Grafana's Big Tent" podcast has launched, continuing its exploration of observability by spotlighting key figures and innovative projects in the field. Originating from Grafana Labs' "big tent" philosophy, the podcast aims to foster open discussions on accessing and utilizing diverse data sources and tools. The new season kicks off with an episode titled "Cache Rules Everything Around Me," featuring Mat Ryer, Ed Welch, Danny Kopping, and Alan "dormando" Kasindorf, who delve into the intricacies of caching, covering technical considerations and common challenges in optimizing the performance of distributed systems. They discuss the fundamental aspects of caching, such as its role in reducing redundancy and latency, and explore best practices and potential pitfalls, including cache invalidation and the appropriate conditions for implementing caching solutions. The episode also highlights recent advancements in Memcached, an open-source caching system, emphasizing its improved efficiency and adaptability for managing large-scale data tasks.

Cloud Provider Observability, announced at ObservabilityCON 2024 and now generally available, offers a streamlined solution for monitoring AWS, Microsoft Azure, and Google Cloud services within Grafana Cloud, addressing the complexity of managing multi-cloud environments with multiple tools. This application, an extension of the AWS Observability app, provides easy setup and scaling through comprehensive scraping instructions, enabling users to start ingesting cloud provider telemetry seamlessly. Its automatic service detection feature ensures immediate and continuous monitoring of supported services, while detailed log views enhance troubleshooting by reducing mean time to resolution (MTTR). For Google Cloud and Azure, the app leverages Grafana Alloy to offer actionable insights and aims to achieve feature parity with AWS offerings soon. The solution simplifies multi-cloud monitoring by offering pre-configured dashboards and a unified user experience, empowering teams to manage complex environments with increased confidence and efficiency.

Adaptive Metrics in Grafana Cloud is a transformative feature that aggregates unused and partially used metrics into lower cardinality versions, significantly reducing metrics costs for over 1,200 organizations. The introduction of Exemptions within Adaptive Metrics addresses concerns about potential data loss by allowing teams to identify and preserve critical metrics, ensuring they are excluded from automated aggregations. Exemptions provide four options for data preservation, including keeping entire metrics intact, preserving specific labels across all metrics, retaining labels on particular metrics, and disabling recommendations for specific metrics entirely. This capability empowers users to maintain essential data integrity while benefiting from cost optimization and reduced cardinality. Adaptive Metrics continues to provide daily aggregation recommendations, adapting to usage patterns and allowing teams to leverage comprehensive data sets for various applications, from new projects to incident reviews. Grafana Cloud offers a user-friendly platform for managing metrics, with a forever-free tier and scalable plans for diverse needs.

Grafana Cloud has introduced several updates and new features aimed at enhancing user experience and expanding its capabilities. This includes new data visualization options like pagination in state timelines and enhanced transformations, allowing more dynamic and insightful data manipulations. Grafana Cloud k6 has seen improvements in performance testing, with features like exporting browser test results to dashboards, integrating into CI/CD pipelines with new GitHub actions, and implementing role-based access control for more precise user permissions. A centralized view of alert history now allows users to track alert events comprehensively, aiding in trend analysis and debugging. In Kubernetes monitoring, users can now view energy usage metrics, enhancing insights into infrastructure efficiency. The introduction of native histograms in Prometheus offers higher resolution and simpler instrumentation for data analysis. Grafana Cloud has also updated its integration lineup, adding support for platforms like VMware vSphere and Alloy Health, while expanding its catalog of Enterprise data source plugins. A new public roadmap for data source plugins is now available, encouraging community engagement and feedback on Grafana's development plans.

Grafana Labs has established itself as a leader in the observability space, being recognized as a Leader in the 2024 Gartner® Magic Quadrant™ for Observability Platforms and ranked among the top vendors across all use cases in the Gartner Critical Capabilities for Observability Platforms report. With a robust user base exceeding 20 million and over 5,000 paying customers, the company's products, including the Grafana LGTM Stack and Grafana Cloud, enable organizations to unify disparate data sources, enhance incident response, and improve system reliability. The company's origins trace back to Torkel Ödegaard's side project in 2013, which has since evolved into a key player in modern software development management, earning accolades such as a spot in the Silicon Valley Defense Group's NATSEC100 and recognition on the Forbes Cloud 100 list. User reviews on Gartner Peer Insights™ highlight Grafana's ease of use, cost-effectiveness, and comprehensive support, contributing to an overall rating of 4.6 out of 5, with 96% of users recommending its products.

Grafana k6, an open-source load testing tool, has introduced several new features aimed at enhancing user experience and aligning with modern development practices. The recent updates include native TypeScript support in k6 v0.52, allowing tests to run without JavaScript bundlers, and the inclusion of ES6+ features such as optional chaining and object spread. The browser APIs have been updated to support asynchronous operations, aligning with the broader JavaScript ecosystem, though this introduces breaking changes that require migration. The gRPC module has become stable, supporting non-blocking async operations, and a new Stream module allows efficient consumption of large datasets by reading files in chunks. Timer methods like setInterval and setTimeout are now globally available, and there are new Web Crypto methods for cryptographic operations. The k6 v0.53 release also adds support for storing results in OpenTelemetry backends, enhancing interoperability with industry standards. These developments reflect ongoing efforts to cater to community requests and ensure compatibility with widely used JavaScript libraries and frameworks.

Juraci Paixão Kröhling, a principal engineer at Grafana Labs and a member of the OpenTelemetry Governance Committee (GC), provides an insight into his multifaceted role, which includes writing code, maintaining components, participating in security audits, and fostering community collaboration for the success of OpenTelemetry. Elected initially in 2021 and re-elected in 2023, Kröhling highlights the GC's strategic role in defining the project's roadmap and ensuring vendor-neutrality, contrasting it with the Technical Committee's (TC) focus on technical issues. His responsibilities as a GC member involve representing the project at various events, reviewing and updating project policies, sponsoring Special Interest Groups (SIGs), and conducting monthly check-ins with SIG maintainers to ensure their needs are addressed. He emphasizes the importance of wearing different "hats" depending on the context, whether representing Grafana Labs, the wider open-source community, or the GC, to maintain clarity and alignment. Kröhling also discusses the challenges of conflict mediation within the community and reflects on the importance of mentoring future leaders to ensure sustainable growth. He underscores the need for GC members to balance their responsibilities with engineering tasks and encourages potential candidates to consider what unique contributions they can make to the GC.

Span events, as unique points within a span's duration, provide additional granularity in distributed tracing, offering valuable insights for debugging and performance optimization without significantly increasing trace sizes. Unlike span attributes, span events can store timestamps, allowing precise performance monitoring and structured logging. They are especially useful for error tracking, as they integrate with spans to form comprehensive traces that capture the sequence of actions leading to errors. TraceQL, a query language for Grafana Cloud Traces, now includes support for querying span events, enabling users to search for specific event names and attributes, thus enhancing monitoring and troubleshooting capabilities. The recent enhancements to TraceQL offer richer sampling capabilities, allowing selective data capture and improved application performance monitoring.

Grafana has released versions 11.1.4, 11.0.3, and 10.4.7 to address a medium severity security vulnerability identified as CVE-2024-6837, which affects the Swagger API documentation, allowing for potential XSS exploits. The vulnerability, impacting Grafana OSS and Enterprise versions from 10.4.0 to 11.1.3, enables attackers to inject arbitrary HTML content that could steal session cookies or interact with the API on behalf of users. Grafana Cloud offerings have received the necessary patches, and cloud providers like Amazon and Azure have been notified to ensure security. Users are advised to upgrade to patched versions and implement a Content Security Policy to mitigate risks. The vulnerability was introduced in January 2024 and discovered in July through Grafana's Security Bounty program, with a fix merged and publicly released by mid-August. Grafana Labs encourages reporting security issues via their dedicated page and maintains a blog for security updates and announcements.

ObservabilityCON on the Road is an extension of the flagship open-source observability conference, ObservabilityCON, which will take place in New York City on September 24-25, 2024. It will include sessions on root cause analysis, synthetic monitoring, and AI/ML, with community talks from organizations like Novo Nordisk and IG Group. Following the main event, ObservabilityCON on the Road will travel to Berlin, London, São Paulo, and Paris, offering one-day events featuring technical deep dives, live demos, and expert panels. These events will also include presentations in the local languages and opportunities for local observability professionals to share their success stories. Registration is open for Berlin and London, with notifications available for Paris and São Paulo as additional dates and locations in North America and Australia are anticipated to be announced later in the year.

Adaptive Metrics, a feature within Grafana Cloud, effectively addresses the challenges of managing large volumes of metrics data by aggregating unused and partially used metrics into lower cardinality versions, leading to significant cost savings and efficiency improvements. This tool has helped over 1,200 organizations reduce metrics costs by an average of 35%, allowing companies like SailPoint, TeleTracking, Dell Technologies, and Mux to reinvest in their observability stacks and improve their operational efficiency. SailPoint successfully reduced their metrics volume by 33%, while TeleTracking cut their spending on Grafana Cloud Metrics by 50%. Dell Technologies benefited from the elimination of unused metrics, enhancing decision-making processes, and Mux achieved a 60% reduction in metrics volume, enabling them to expand metrics retention and adopt a more proactive incident management approach. These companies have leveraged the insights provided by Adaptive Metrics to focus on high-value metrics, streamline operations, and optimize resource allocation, ultimately enhancing their overall observability capabilities and business outcomes.

Schwarz IT has strategically implemented Grafana Enterprise to enhance observability and data visualization across its multinational operations, benefiting both technical and non-technical users. Initially adopting Grafana OSS in 2014 and transitioning to Grafana Enterprise in 2022, the company has effectively utilized the platform to unify its monitoring strategy while empowering diverse users to create personalized data views. The versatility of Grafana has facilitated widespread adoption, with over 6,000 dashboards and 1,500 active users within Schwarz IT, which supports the IT needs of the Schwarz Group's various divisions, including Lidl and Kaufland. The creation of an active Grafana community has been pivotal, driven by initiatives such as hackathons and internal training sessions that encourage knowledge sharing and innovation. Felix Spitzer, a Monitoring Specialist at Schwarz IT, highlights the platform's adaptability as a key factor in building a robust community, noting that Grafana serves as a central visualization layer for multiple data sources. Future plans include expanding Grafana's usage through additional hackathons and developing standard dashboards for broader business process monitoring across different countries within the group.

Integrating Okta logs with Grafana Loki enhances security information and event management (SIEM) capabilities by allowing seamless retrieval, analysis, and alerting on event logs from Okta’s System Log API. The process involves setting up an Okta logs collector, which automates the fetching of logs and sends them to STDOUT to be forwarded to observability platforms like Loki using agents such as Alloy or Promtail. This integration simplifies the traditionally complex task of log retrieval by eliminating the need for custom code, leveraging Docker containers to efficiently manage and process log data. The integration also supports advanced log analysis with Grafana and Loki, enabling users to utilize Sigma rules for detecting critical log lines and setting up alerts, thereby improving system security and operational capabilities. Future developments aim to further streamline this process with enhanced features and deeper integration within the Alloy framework.

In the context of managing Kubernetes infrastructure, the use of Kubernetes Monitoring in Grafana Cloud significantly enhances the ability of site reliability engineers (SREs) to troubleshoot, manage resources, and plan for future incidents by providing historical visibility of deleted components. Unlike the manual and error-prone processes associated with using kubectl for historical data, Grafana Cloud offers an integrated solution that allows engineers to access timelines and logs of pods and nodes that have been deleted, thereby reducing downtime and improving system resilience. This capability is particularly beneficial in scenarios such as debugging post-deployment issues, optimizing resource usage, and analyzing past incidents to prevent future outages. By offering a comprehensive view of the system’s past states, Grafana Cloud aids in identifying root causes and implementing preventive measures, making it an invaluable tool for SREs, cloud infrastructure admins, and developers.

Grafana Alloy 1.3 introduces live debugging capabilities, enhancing the real-time monitoring and debugging of data pipelines within its open-source distribution of the OpenTelemetry Collector. This feature allows users to visualize data transformations, identify errors, and analyze pipeline behavior efficiently, though it is disabled by default to protect sensitive telemetry data and requires enabling via configuration. Designed to streamline development and optimization, live debugging supports components such as loki.process, loki.relabel, otelcol.processor, otelcol.receiver, and prometheus.relabel, allowing for incremental pipeline construction and verification of data formats. Grafana encourages users to explore these new features through their documentation and community resources, offering a free tier to facilitate easy access to metrics, logs, and dashboards on Grafana Cloud.

Japan's space agency, JAXA, achieved a historic milestone on January 19, 2024, by becoming the fifth country to land a spacecraft on the moon, with its SLIM (Smart Lander for Investigating Moon) successfully achieving a pinpoint landing within 100 meters of its target. This operation was closely monitored in real-time using Grafana dashboards, which tracked critical functions such as navigation, heater status, and landing maneuvers. At GrafanaCON 2024, Satoshi Nakahira, an associate senior researcher at JAXA's Institute of Space and Astronautical Science (ISAS), highlighted the significant role Grafana played in providing enhanced observability and visualization capabilities. The agency used InfluxDB as a data source and an API for image calculations, which were integrated into more than 10 Grafana dashboards to monitor SLIM's various subsystems during different phases of the mission. Grafana's features, such as variable capabilities, allowed JAXA to efficiently manage and display extensive telemetry data, facilitating a successful lunar landing. The dashboards created were not only instrumental in the control room but were also live-streamed on YouTube for public viewing, showcasing their real-time application during the mission.