November 2025 Summaries

Amazon Bedrock AgentCore is a managed AI agent service by AWS that enables developers to build and deploy generative AI agents without managing infrastructure, facilitating complex task automation and data interaction. Monitoring these agents can be challenging due to the absence of deep insights, making debugging and performance tuning difficult. Grafana Cloud offers a solution by providing a unified platform for comprehensive observability across the entire AWS ecosystem, including Bedrock AgentCore agents. By leveraging OpenTelemetry, Amazon CloudWatch, and Grafana Cloud, users can achieve end-to-end visibility, track essential performance metrics, and streamline troubleshooting. Grafana Cloud's Cloud Provider Observability application provides out-of-the-box dashboards and alerts tailored to AWS resources, eliminating the need for manual configuration and enabling immediate access to key performance indicators. This centralized monitoring approach simplifies complex challenges, ensures consistent operational standards, and enhances the efficiency of generative AI agents.

AI agents have become essential in engineering workflows, necessitating robust observability to manage their complexities. This tutorial guides users on deploying AI agents on Amazon Bedrock AgentCore with comprehensive observability using OpenTelemetry and Grafana Cloud. Amazon Bedrock AgentCore provides a serverless runtime for AI agents, handling infrastructure and scaling, while OpenTelemetry offers standardized instrumentation to monitor distributed applications. The tutorial emphasizes deploying a CrewAI agent with Meta's Llama 3 on AgentCore and using OpenLit for automatic observability, allowing users to track performance, debug issues, and optimize costs. It also details configuring Grafana Cloud to visualize telemetry data and provides insights into AI agent operations through pre-built dashboards. With AWS Bedrock AgentCore, OpenTelemetry, and Grafana Cloud, users can achieve enterprise-grade observability for their AI applications.

GrafanaCON 2026 is set to take place from April 20-22 in Barcelona, serving as Grafana Labs' largest community event of the year, where participants can enjoy technical sessions, hands-on learning, and user stories. This event will spotlight the new features in Grafana 13 and include over 20 sessions related to open-source projects like Prometheus, OpenTelemetry, Loki, Mimir, and Tempo. It offers a platform for peers to connect and share innovative uses of Grafana, with presentations ranging from beginner to advanced levels. The call for presentations is open until January 19, 2026, inviting contributions from diverse backgrounds, including first-time speakers. Registration opens in February, with early access to tickets and special pricing available for those who sign up for notifications.

Temporal Technologies has announced a new integration between its managed service, Temporal Cloud, and Grafana Cloud, aiming to enhance the observability of Temporal Workflows. This partnership allows users to seamlessly visualize, monitor, and receive alerts on the health of their workflows directly within Grafana Cloud dashboards. The integration simplifies the process by providing expanded metrics including new dimensions like temporal_task_queue and temporal_workflow_type, improving accuracy and offering enhanced operational insights. Users benefit from effortless observability with a prebuilt dashboard that requires no custom configuration, and setup is streamlined to just adding a Temporal Cloud API key. This new feature eliminates the need for complex configurations previously required, offering a real-time, unified view of workflow and service performance across various environments and teams.

Grafana enables seamless cross-store correlations, allowing users to pivot effortlessly between charts, logs, and traces, even when the data resides outside its native backends like Grafana Loki, Grafana Mimir, and Grafana Tempo. This functionality supports Grafana's "big tent" philosophy, accommodating diverse data needs without requiring users to store data within Grafana. By utilizing a feature called correlations, users can click on a chart spike and directly access the relevant logs, maintaining efficiency and minimizing disruption during investigations. Although the blog exemplifies this process using VictoriaMetrics, the approach can be adapted to other data sources. Setting up correlations involves minimal configuration, such as establishing a Grafana Cloud account and using compatible data sources, yet significantly enhances the speed and effectiveness of troubleshooting. This cross-platform capability underscores Grafana's commitment to providing flexible and user-friendly data solutions, making it an accessible tool for managing metrics, logs, and traces without extensive reconfiguration.

ObservabilityCON on the Road is a series of one-day events taking place in multiple cities in 2026, following the success of ObservabilityCON 2025 held in London. Designed for observability enthusiasts, the events will feature technical deep dives, live demos, and an "Ask the Experts" booth, along with opportunities to explore the latest AI features and Grafana Cloud solutions aimed at enhancing user experiences and optimizing costs. Attendees will benefit from technical sessions offering actionable advice, real-world success stories from regional leaders, and hands-on demos, as well as networking opportunities. Registration is currently open with early bird discounts available, and more locations and dates are expected to be announced.

Grafana 12.3 introduces a host of new features designed to enhance user experience and data analysis capabilities, including interactive learning resources and improved log visualization tools. Users can now access context-aware tutorials and documentation directly within the platform, facilitating quicker learning without the need to switch contexts. The release also includes a redesign of the logs panel, offering features such as color highlighting and client-side search to improve log analysis. Additional updates include the ability to export dashboards as images, apply CSS styling to table cells, and enhanced data source integrations, such as new capabilities with SolarWinds, Amazon CloudWatch, and Databricks. Grafana 12.3 also addresses a critical security vulnerability and invites users to engage with the community to share experiences and feedback, which are vital to the platform's ongoing development.

Grafana Labs announced a critical security update for Grafana Enterprise versions 12.0.0 to 12.2.1, addressing a vulnerability identified as CVE-2025-41115 in the SCIM (System for Cross-domain Identity Management) provisioning feature. This issue, which could lead to privilege escalation or user impersonation, arises when specific configurations such as the enableSCIM feature flag and user_sync_enabled option are set to true, allowing a compromised SCIM client to override internal user IDs. With a CVSS score of 10.0, indicating critical severity, Grafana Labs has released patched versions of Grafana Enterprise and applied appropriate fixes to Grafana Cloud, ensuring that services like Amazon Managed Grafana and Azure Managed Grafana are secure. This vulnerability does not affect Grafana OSS users. The company has urged users of affected versions to upgrade immediately and has provided detailed incident timelines and guidelines for reporting security issues, emphasizing non-disclosure until fixes are implemented and announced.

Grafana Logs Drilldown, when paired with Grafana Loki, offers an enhanced logging experience by allowing users to quickly filter and analyze large volumes of log data without writing queries. The tool supports features like automatic stream sharding to expedite data retrieval, customizable views for organizing logs by various dimensions, regex support for complex searches, and advanced sorting and numeric filtering options. New functionalities include automatic visualizations, JSON log visualization, and machine learning algorithms for sorting, all aimed at improving troubleshooting efficiency and reducing mean time to resolution (MTTR). Grafana Cloud users can benefit from these updates in logging as well as explore other telemetry types such as metrics with Grafana Mimir, traces with Grafana Tempo, and profiles with Grafana Pyroscope, making it a versatile platform for comprehensive infrastructure and application monitoring.

Azure Monitor has announced the general availability of its integration with Grafana, allowing users to access Grafana dashboards directly within the Azure Portal without additional costs or administrative complexity. Initially launched in a public preview in May 2025 and now fully available, this service offers prebuilt dashboards for over a dozen Azure services, including Azure Monitor metrics, logs, and Application Insights. Users benefit from compatibility with Grafana OSS, enabling them to import a wide range of community-created dashboards and share them across different Grafana instances. The latest features include support for Grafana Explore, enhanced Azure Kubernetes Service monitoring, and new prebuilt dashboards for Azure Container Apps and High Performance Computing, among others. This integration aims to enhance real-time operational monitoring and visualization for Azure's extensive user base, further embedding Grafana as a key tool for observability data visualization.

Grafana Play, a free and publicly accessible sandbox environment for exploring Grafana without setup or sign-in, has unveiled a redesigned homepage to better reflect its dynamic, community-driven nature. Hosted on Grafana Cloud, Grafana Play has become a popular destination for learning and experimentation with thousands of public dashboards available. The new homepage design aims to be more welcoming and playful, celebrating community contributions by highlighting creative dashboards and providing easy access to community links, forums, documentation, and events. Notable features include a new header panel for community connections, a Featured Grafana Contributor section, a Dashboard of the Month spotlight, and a Play Launchpad that offers a hub for exploring Grafana's visualization and data source examples. Additionally, the Grafana Arcade introduces interactive games for learning, and the Featured Dashboards section offers inspiration for users. The redesign also emphasizes the importance of community feedback and contributions, especially through the Grafana Community Slack, and plans to expand with a demo of the Grafana LGTM Stack to enhance the user experience further.

Grafana Labs recently demonstrated the effectiveness of their AI-powered tool, Grafana Assistant Investigations, which found the root cause of an incident 3.5 times faster than their on-call engineering team. This AI tool, now available in public preview for Grafana Cloud users, analyzes observability data such as metrics, logs, traces, and profiles to identify anomalies and provide actionable recommendations for incident resolution. During an incident caused by an AI-generated SQL query that led to database saturation, the Assistant Investigations tool ran parallel investigations with specialized agents, accurately pinpointing the problem and suggesting remediation before the human team completed their analysis. This incident highlights the growing role of AI in accelerating incident response and understanding complex systems, with AI tools complementing rather than replacing human engineers by providing data-backed insights and reducing response time. Grafana's approach emphasizes human-in-the-loop interactions, ensuring that AI assists in decision-making while maintaining transparency and control, ultimately enhancing productivity and system comprehension in a rapidly evolving tech landscape.

Continuous profiling has emerged as a crucial component of observability in software development, complementing logs, metrics, and tracing by providing insights into CPU usage and performance inefficiencies in native codebases. Initially a practice mainly for managed languages, recent advancements, particularly in eBPF technology, have made continuous profiling applicable to native code as well. This approach allows developers to detect performance issues proactively, conduct differential performance analysis, achieve cost savings in cloud deployments, and identify inefficiencies without preconceived notions, thereby addressing "unknown unknowns" in application performance. The blog post illustrates the practical application of continuous profiling using tools like Grafana Alloy and Grafana Cloud Profiles in benchmarking open-source applications like DuckDB and PostgreSQL, demonstrating minimal performance impact and providing actionable insights for CPU optimization. The availability of off-the-shelf components has significantly simplified the process of gaining detailed performance insights, making it accessible to development teams without the need for specialized engineering skills.

Ronald McCollam's blog post details the creation of a Star Wars-themed dashboard using Grafana, launched on Life Day, a lesser-known holiday within the Star Wars universe. The dashboard features various creative visualizations, including a map of the Kessel Run and details on the Millennium Falcon and Death Star. McCollam shares insights on using Grafana's tools, such as substituting a stat panel for a text panel to achieve custom styling and utilizing the TestData plugin to simulate dynamic data. He also discusses using the canvas panel for creating intricate navigation paths and employing transparent backgrounds to integrate images and text seamlessly. The blog emphasizes the value of experimentation and customization in crafting engaging and informative dashboards, encouraging readers to explore these techniques in their own projects.

Grafana Cloud Database Observability is a new feature designed to aid application developers, SREs, and DBAs in understanding, diagnosing, and optimizing SQL queries, thereby addressing a common source of application performance issues: inefficient database queries. This tool allows users to correlate data across their entire stack, from application to infrastructure, and provides insights into query performance that traditional application-level observability tools lack. It offers a comprehensive overview of database operations through metrics such as rate, error, and duration (RED), and allows for detailed query analysis, including performance trends and schema details, all within the Grafana Cloud environment. Additionally, the tool incorporates AI to suggest optimization strategies, includes capabilities for examining query execution plans, and facilitates collaboration across development and operations teams by eliminating the need for specialized DBA resources. It aims to simplify the troubleshooting process by providing granular insights into the database layer, which is often a black box in performance incidents. Users are encouraged to set up Database Observability by enabling telemetry collection on their databases and using Grafana Alloy for data collection, making it a powerful addition to Grafana Cloud’s suite of observability tools.

Periods of high customer activity, such as Black Friday and major sales, can strain software and infrastructure systems, leading companies like Just Eat Takeaway.com to utilize Grafana Cloud k6 for performance testing to ensure system reliability under load. This testing framework helps organizations prepare for peak demands by prioritizing critical user journeys and employing proactive monitoring to detect performance regressions early. Grafana Cloud k6 provides flexible testing options, including synthetic monitoring and customizable dashboards, allowing for continuous feedback on application performance and the ability to identify patterns in testing data. By testing early and often, companies can better manage peak loads and optimize user experiences, as highlighted by Grafana Labs' recognition as a leader in the 2025 GigaOm Radar Report for Cloud Performance Testing.

Grafana Mimir 3.0 introduces significant enhancements to this open-source, horizontally scalable time series database, primarily focusing on improved reliability, performance, and cost efficiency. The new version features a decoupled architecture that separates read and write operations, employing Apache Kafka as an asynchronous buffer to enable independent scaling paths, which results in faster queries and smoother ingestion processes. The Mimir Query Engine, a key component of the update, offers a more memory-efficient approach by streaming query results rather than loading entire datasets, reducing peak memory usage by up to 92% and ensuring compatibility with Prometheus' PromQL. These architectural advancements, along with reduced resource consumption by up to 15%, make Mimir 3.0 a robust solution for managing large-scale data metrics, and it is available both as an open-source project and through the Grafana Cloud Metrics service. The release is a collaborative effort, driven by community feedback and contributions, showcasing Grafana Labs' commitment to advancing open observability solutions.