What’s new in Grafana Cloud

On any Pod or workload detail page, you can navigate directly to Application Observability to see more data about the application. (Release 2.0.3)

If you’ve wanted to visualize your Grafana Cloud k6 test results in a dashboard you’ve been limited to displaying data as a time series. But sometimes a single number is more digestible and can help you make an assessment of your test results quicker.

Now you can aggregate your data not only over time, but as a single aggregated value. This is especially useful in combination with the Stat panel. Just change the type of your query to “Aggregate” in the new options section and data will now be aggregated for the entire time series.

The new alert rule detail view has a new look and feel with helpful metadata at the top. The namespace and group are shown in the breadcrumb navigation. This is interactive and can be used to filter rules by namespace or group. The rest of the alert detail content is split up into tabs:

Query and conditions

View the details of the query that is used for the alert rule, including the expressions and intermediate values for each step of the expression pipeline. A graph view is included for range queries and data sources that return time series-like data frames.

Instances

Explore each alert instance, its status, labels and various other metadata for multi-dimensional alert rules.

History

Explore the recorded history for an alert rule.

Details

Debug or audit using the alert rule metadata and view the alert rule annotations.

Use Alerting insights to monitor your alerting data, discover key trends about your organization’s alert management performance, and find patterns in why things go wrong.

AngularJS support in Grafana was deprecated in v9 and will be turned off by default in Grafana v11. When this happens, any plugin which depended on AngularJS will not load, and dashboard panels will be unable to show data.

To help you understand where you may be impacted, Grafana now displays a warning banner in any dashboard with a dependency on an AngularJS plugin. Additionally, warning icons are present in any panel where the panel plugin or underlying data source plugin has an AngularJS dependency.

This complements the existing warnings already present on the Plugins page under the administration menu.

In addition, you can use our detect-angular-dashboards open source tool, which can be run against any Grafana instance to generate a report listing all dashboards that have a dependency on an AngularJS plugin, as well as which plugins are in use. This tool also supports the detection of private plugins that are dependent on AngularJS, however this particular feature requires Grafana v10.1.0 or higher.

Use the aforementioned tooling and warnings to plan migrations to React based visualizations and data sources included in Grafana or from the Grafana plugins catalog.

To learn more, refer to the Angular support deprecation, which includes recommended alternative plugins.

We are announcing a license change to the anonymous access feature in Grafana 11. As you may already be aware, anonymous access allows users access to Grafana without login credentials. Anonymous access was an early feature of Grafana to share dashboards; however, we recently introduced Public Dashboards which allows you to share dashboards in a more secure manner. We also noticed that anonymous access inadvertently resulted in user licensing issues. After careful consideration, we have decided to charge for the continued use of anonymous access starting in Grafana 11.

Affected Grafana versions

Anonymous authentication is disabled by default in Grafana Cloud. This licensing change only affects Grafana Enterprise (self-managed) edition. Anonymous users will be charged as active users in Grafana Enterprise.

The Datadog data source supports blocking API requests based on upstream rate limits (for metric queries). With this update, you can set a rate limit percentage at which the plugin stops sending queries.

To learn more, refer to Datadog data source settings, as well as the following video demo.

You can now apply data transformations to annotation data. For example, you can now configure how exemplar data is displayed in tooltips.

AWS Logs with Firehose is currently in public preview. Grafana Labs offers limited support, and breaking changes might occur prior to the feature being made generally available.

Logs with Firehose leverages Amazon Kinesis Data Firehose and a minimal infrastructure to deliver logs to the ingestion pipeline within Grafana Cloud.

You can now configure the Azure Monitor data source to authenticate as the logged-in Grafana user when making query and resource requests if you also use Azure Entra to sign your users into Grafana.

Current User authentication allows you to enforce Azure RBAC restrictions on your Grafana users by removing the need to provide broad service credentials. Once a data source is configured with Current User authentication a user will only have access to resources they can access directly in Azure.

Additionally, data sources configured to use Current User authentication are less likely to be impacted by throttling issues due to the individual level of access.

Current User authentication does not inherently support backend features such as alerting. To account for this, data sources configured with Current User authentication can optionally specify service credentials that will be utilized for backend features when no signed-in user is available.

To get started with Current User authentication, refer to the Azure Monitor data source documentation.

From a Pod, Cluster, namespace, or workload view, you can begin an incident investigation by clicking Run Sift investigation.

The Metrics status tab, which shows the configuration status of Kubernetes Monitoring components, is updated for better usability and quicker comprehension.

We released generative AI support in both OSS and Cloud last year, supporting features like Incident auto-summary and dashboard creation assistance. One downside was that it was tedious to set up – you had to install & enable the LLM plugin, and separately sign up with OpenAI or Azure to configure your own API keys.

On Cloud, this just got a lot easier!

In your Grafana Cloud account, you’ll find the LLM plugin already there and ready to go. To turn on generative AI features for your instance, just choose “Use OpenAI provided by Grafana”, accept the terms, and click “Save & test”.

Then try one of the Gen AI powered features:

• Explain Flamegraphs & offer suggestions to fix issues
• Suggest titles & descriptions for dashboards & dashboard panels
• Explain error log lines in Sift
• Auto-summarize an incident while resolving
• Generate KQL queries in the Azure Data Explorer plugin

To learn more, check our online documentation.

Canvas visualizations now support panning and zooming. This allows you to both create and navigate more complex designs.

To enable this feature, you must first enable the canvasPanelPanZoom feature toggle.

Use the in-app Cardinality page to discover on a Cluster-by-Cluster basis where all your active series are coming from for:

• Troubleshooting after initial configuration
• Determining potential duplicate metrics
• Finding other potential cardinality issues

From the main menu, click Configuration and then the Cardinality tab.

You can respond to and troubleshoot alerts that are firing about your Kubernetes infrastructure and the applications running within it, without leaving the context of Grafana Kubernetes Monitoring. You can start your troubleshooting either through the home page or the Alerts page.

At the Pods in trouble section on the home page, you can view the alert associated with each Pod in the list.

The Alerts page displays all alerts related solely to your Kubernetes infrastructure and any applications within your infrastructure. Here you can filter and drill into issues for quick resolution.

Release v2.0.0

It’s hard to diagnose and resolve issues when your observability data is dispersed across many systems, which can lead to longer times to troubleshoot. Grafana’s AWS Observability provides a centralized location to work with critical observability data so you can fully understand the state of your systems. You can:

• Configure simply and securely
• Access, query, alert on, and interact in one place

AWS Observability allows you to:

• Monitor AWS EC2 services
• Connect and pull CloudWatch metrics
• Connect and send logs, including access logs

The integration for ClickHouse is available for use with Kubernetes Monitoring.

Release v2.0.0

We are happy to announce Cloud Logs Export is now Generally Available to all Grafana Cloud users!

Cloud Logs Export lets users ship logs to their own object storage (AWS, GCP, Azure) for low-cost, long term retention. If you need to retain logs for long periods of time, without a need to frequently query them, Cloud Logs is the most cost effective way to do that.

To learn more, read our public preview blog post announcement, or try it out yourself (From the Grafana menu, “Plugins” -> “Cloud Logs Exporter” to get started!)

Cost monitoring and investigation and troubleshooting of CPU and memory usage are now easier for both Clusters and Nodes. Go to the optimization panels for the Cluster and Node detail pages to view the CPU and memory utilization, as well as cost information.

These graphs show a history based on the time range you select.

Grafana 11 adds the ability to color full table rows using the Colored background cell type of the table visualization. When you configure fields in a table to use this cell type, an option to apply the color of the cell to the entire row becomes available.

This feature is useful for a wide variety of use cases including mapping status fields to colors (for example, info, debug, warning) and allowing rows to be colored based on threshold values. This is one of the first steps in making formatting tables more seamless, and allows for quick scanning of data using the table visualization.

To learn more, refer to the documentation for the **Colored background ** cell type.

Plugins can now add custom actions into Grafana’s command palette. Actions can link directly to a plugin page, or open a modal to allow users to interact with a plugin without losing their current page.

In Grafana Cloud you can use the new ‘Run investigation’ action to start a Sift investigation from anywhere in Grafana. Keep an eye out for new actions appearing in your command palette in the near future!

When configuring Kubernetes Monitoring with the streamlined configuration wizard that uses Grafana Kubernetes Helm Chart, you can select the platform you are using.

You can choose to configure manually, or use a more streamlined configuration process with CloudFormation or Terraform. To send CloudWatch metrics to Grafana Cloud, you:

• Connect to your AWS account.
• Configure the connection between Grafana Cloud and your AWS account.
• Continue configuration with either CloudFormation or Terraform.
• Choose what service to monitor, what metrics to gather, the scrape interval, and what statistics to gather.
• Add any custom namespaces you want to monitor.

You can also edit or delete scrape jobs.

To send CloudWatch or ALB logs to Grafana Cloud, choose CloudFormation or Terraform to configure for setting up a Lambda function.

When you use the easy configuration process with Grafana Kubernetes Monitoring Helm chart, you can switch on the use of OTLP and Zipkin receivers for containerized application monitoring.

(Release 2.0.3)

With Grafana v9.3, we introduced a feature toggle called accessTokenExpirationCheck. It improves the security of Grafana by checking the expiration of the access token and automatically refreshing the expired access token when a user is logged in using one of the OAuth providers.

With the current release, we’ve introduced a new configuration option for each OAuth provider called use_refresh_token that allows you to configure whether the particular OAuth integration should use refresh tokens to automatically refresh access tokens when they expire. In addition, to further improve security and provide secure defaults, use_refresh_token is enabled by default for providers that support either refreshing tokens automatically or client-controlled fetching of refresh tokens. It’s enabled by default for the following OAuth providers: AzureAD, GitLab, Google.

For more information on how to set up refresh token handling, please refer to the documentation of the particular OAuth provider.

The accessTokenExpirationCheck feature toggle will be removed in Grafana v10.3.

The Contact points list view has been redesigned and split into two tabs: Contact Points and Notification Templates, making it easier to view all contact point information at a glance. You can now search for name and type of contact points and integrations, view how many notification policies each contact point is being used for, and navigate directly to the linked notification policies.

Introducing Content Outline in Grafana Explore. We recognized the challenges of complex mixed queries, as well as, lengthy logs and traces results, leading to time-consuming navigation and the loss of context. Content outline is our first step towards seamless navigation from log lines to traces and back to queries ensuring quicker searches while preserving context. Experience efficient, contextual investigations with this update in Grafana Explore. To learn more, refer to the Content outline documentation, as well as the following video demo.

Copying and pasting time range in the time range picker is now available. For example, you can copy a time range in Explore and paste it into Dashboards and vice versa. You can also copy and paste a time range using the new keyboard shortcuts t+c and t+v, respectively.

Throughout Kubernetes Monitoring, hover to reveal the copy icon. You can click to copy any text so that you can:

• Use it in your own query.
• Paste it in the command line.
• Send it to others.

Creating correlations has just become easier. Try out our new correlations editor in Explore by selecting the + Add > Add correlation option from the top bar or from the command palette. The editor shows all possible places where you can place data links and guides you through building and testing target queries. For more information, refer to the documentation.

Explore cost information at any level within your Kubernetes infrastructure. In lists of Clusters, Nodes, namespaces, and workloads, you can switch from a usage view to a cost view to see the cost of each item in the list.

A continuous path of troubleshooting is available as well. Each detail page offers adjustment recommendations to help lower costs.

Release 2.0.0

Plugin developers and users typically want their plugins to be compatible with a range of Grafana versions. However, ensuring this is true can be a challenge. The environment, APIs, and UI components may differ from one version to another. Manual testing is tedious and error-prone. End-to-End testing across multiple versions is prohibitively complicated given the version-to-version changes in Grafana itself - until now.

We would like to introduce @grafana/plugin-e2e - a new end-to-end testing library that is designed specifically for Grafana plugin developers. It extends @playwright/test capabilities with relevant fixtures, models, and expect matchers; enabling comprehensive end-to-end testing of Grafana plugins across multiple versions of Grafana. This package simplifies the testing process, ensuring your plugin is robust and compatible with various Grafana environments.

The first major version of @grafana/plugin-e2e has just been released. This is the library we will offer from now on as part of our tooling, the grafana/e2e package is now deprecated.

For everyone who is using the latest version of our create-plugin tool: All new plugins you scaffold already have this library integrated and set up - all you need to do is start writing tests.

And for those who already have plugins they are maintaining: You can keep using cypress as you did before and bit by bit, in parallel, migrate to using this library instead.

Create alerts from dashboard panels. You can reuse the panel queries and create alerts based on them.

You can now create correlations using either the Administration page or provisioning, regardless of whether a data source was provisioned or not. In previous versions of Grafana, if a data source was provisioned, the only way to add correlations to it was also with provisioning. Now, that’s no longer the case, and you can easily create new correlations mixing both methods—using the Administration page or provisioning.

To enable this feature, contact Grafana Support.

You can now add buttons to your canvas visualizations. Buttons can be configured to call an API endpoint. This pushes Grafana’s capabilities to new heights, allowing you to create interactive dashboards that can be used to control external systems.

To learn more, refer to our Canvas button element documentation.

You can now create subtables out of your data using the new Group to nested tables transformation. To use this feature, enable the groupToNestedTableTransformation feature toggle.

In addition to Query and Edit access, you can now grant users, teams, or basic roles Admin access to data sources. Users with Admin access to a data source can grant and revoke permissions to the data source, as well as to manage query caching settings for the data source. Users are automatically granted Admin access to data sources that they create.

You can change the data source you are viewing on the Cost and Efficiency views. (Release 1.2.1)

We’ve made a number of smaller improvements to the data visualization experience in Grafana.

Geomap marker symbol alignment options

You can now offset geomap marker symbols from the underlying data point.

Gauge visualization overflow support

You can now visualize gauges in vertical and horizontal orientations with overflow. This resolves an issue where the design would break when the number of gauges exceeded the available space.

Bar chart axes improvements

You can now center bar chart axes on zero and configure axes border and color settings.

We’ve made a number of small improvements to the data visualization experience in Grafana.

Geomap geojson layer now supports styling

You can now visualize geojson styles such as polygons, point color/size, and line strings. To learn more, refer to the documentation.

Canvas elements now support snapping and aligning

You can precisely place elements in a canvas with ease as elements now snap into place and align with one another.

View data links inline in table visualizations

You can now view your data links inline to help you keep your tables visually streamlined.

New for the Databricks datasource plugin is the schema aware SQL query builder to help you build faster queries. For general documentation on querying data sources in Grafana, see Query and transform data.

You can run the built query by pressing the Run query button in the top right corner of the editor.

The response from Databricks can be formatted as either a table or as a time series. To use the time series format one of the columns must be named time.

The builder supports aggregations, filtering, grouping and previewing of the query.

Query Editor

To make advanced queries, switch to the code editor by clicking code in the top right corner of the editor. The code editor support autocompletion of tables, columns, SQL keywords, standard sql functions, Grafana template variables, Grafana Macros and SQLDS Macros Columns cannot be completed before a table has been specified.

You can expand the code editor by pressing the chevron pointing downwards in the lower right corner of the code editor.

CTRL/CMD + Return works as a keyboard shortcut to run the query.

Demo video:

We’ve added initial support to detect situations in which various transformations won’t work appropriately based on current data. Previously, selecting the appropriate transformation and configuring it correctly required a process of trial and error or already knowing how a given transformation worked. Now, transformations that we’ve detected can’t be used are shaded in the interface to indicate this, along with a helpful message explaining why.

You can now use the Grafana Cloud Traces integration with Grafana Cloud k6 to quickly debug failed performance tests and proactively improve application reliability.

Distributed tracing in Grafana Cloud k6 only requires two things:

• An application instrumented for tracing with Grafana Cloud Traces.
• Adding a few lines of code to your existing k6 scripts.

The integration works by having k6 inject tracing metadata into the requests it sends to your backend services when you run a test. The tracing data is then correlated with k6 test run data, so you can understand how your services and operations behaved during the whole test run. The collected tracing data is aggregated to generate real-time metrics—such as frequency of calls, error rates, and percentile latencies—that can help you narrow your search space and quickly spot anomalies.

To learn more, refer to the Integration with Grafana Cloud Traces documentation and Distributed Tracing in Grafana Cloud k6 blog post.

With Kubernetes Monitoring, you can effectively identify, prioritize, and handle efficiency-related issues. Efficiency data is available globally, and your troubleshooting path enables you to create an efficiency feedback loop.

Release v2.0.0

AWS Observability provides preconfigured dashboards embedded within the app.

You can easily access dashboards to monitor AWS costs and cloud services. For example, the following shows the billing dashboard.

On any panel of the Cost page in Kubernetes Monitoring, click the Explore button.

This launches the query details, where you can debug or create your own dashboard.

Click the Explore icon to view the query and additional tools available for you to further explore and query your data.

You can explore a query from the Metrics status tab.

You can also explore:

• A Cluster

• A list of Clusters

• A Pod

(Release 2.0.3 - 2.0.4)

Explore Metrics is a query-less experience for browsing Prometheus-compatible metrics. Search for or filter to find a metric. Quickly find related metrics - all in just a few clicks. You do not need to learn PromQL! With Explore Metrics, you can:

• easily slice and dice metrics based on their labels, so you can see anomalies right away
• See the right visualization for your metric based on its type (e.g. gauge vs. counter) without building it yourself
• surface other metrics relevant to the current metric
• “explore in a drawer” - expand a drawer over a dashboard with more content, so you don’t lose your place
• view a history of user steps when navigating through metrics and their filters
• easily pivot to other related telemetry - IE, logs or traces 

… all without writing any queries!

To learn more, refer to Explore Metrics, as well as the following video demo:

After you run a performance test in Grafana Cloud k6, the next step is to analyze the results and share any findings with your team. Sometimes, you don’t need your team members to view all the resulting metrics, or you might want to generate a PDF in an accessible, easy-to-read format to share with external stakeholders.

The new export summary functionality in Grafana Cloud k6 allows you to export your test results as a Grafana dashboard. This dashboard includes panels and charts with key metrics from your test, such as max throughput, HTTP failures, average response time, and more. You can then share it with other team members, edit it, or export it as a PDF.

To learn more, refer to the Export test results documentation.

Export your alerting resources, such as alert rules, contact points, and notification policies as Terraform resources. A new “Modify export” mode for alert rules enables you to edit provisioned alert rules and export a modified version.

Filter by Pod type on the Workloads list page to find static Pods and bare/unmanaged Pods.

Flamegraph AI in Grafana Cloud Profiles is now GA

Flamegraph AI helps you understand your profiling data by using a large-language model (LLM) to assist with data interpretation.

A flamegraph provides a convenient way to visualize performance data. These graphs provide a clear, intuitive understanding of resource allocation and bottlenecks within an application.

Flamegraphs and profiling data in general can help you understand:

1. Performance bottlenecks: What’s causing the slowdown?
2. Root causes: Why is it happening?
3. Recommended fixes: How would you resolve it?

Flamegraph AI answers these questions when explaining your flamegraph.

To learn more, refer to the Flamegraph AI documentation.

Sometimes profile stacks contain lots of levels with similar repeating items, for example long stacks of framework code that usually isn’t of interest but takes up a lot of visual real estate. With this feature, instead of rendering all of the similar items we render only one and allow to expand those collapsed items on demand.

To try it out, enable the ‘traceToProfiles’ feature toggle. To enable it in your Grafana Cloud stack, contact Grafana Support.

Are you a k8s user? Do your requirements involve testing internal services or generating load from inside your network? Private Load Zones are a new feature of Grafana Cloud k6 that allow you to run k6 load generators from your k8s cluster and push metrics to Grafana Cloud and visualize them in the Grafana Cloud k6 app plugin.

This feature is built on top of our OSS k6-operator and lets you start tests with the k6 cloud command the same way you would do if you used Grafana Cloud’s load zones.

For more information on how to set up your own Private Load Zone and execute a test, please view our documentation.

Navigate to the Alert rules page from a list item on a Cluster, namespace, Node, Pod, or container list. To do so, click the underlined alert number for next to the list item.

Users looking to migrate to the new Grafana Alerting product can do so with confidence with the Grafana Alerting migration preview tool. The migration preview tool allows users to view, edit, and delete migrated rules prior cutting over, with the option to roll back to Legacy Alerting.

We’ve identified a need for users who enable anonymous authentication to monitor the anonymous devices connected to their Grafana instance. This feature is part of our ongoing efforts to enhance control and transparency regarding anonymous usage.

Anonymous access now allows users, including those in open-source and enterprise self-managed environments, to view and monitor their anonymous access. They can also set a device limit, configuring a specific number of anonymous devices to connect to their instance.

Once this limit is reached, any new devices attempting to connect will be denied access until existing devices disconnect.

The anonymous devices feature improves the management and monitoring of anonymous access within your Grafana instance.

Anonymous Device:

When anonymous access has been enabled, any device which accesses Grafana in the last 30 days (without logging in) is considered an active anonymous device. Users can now view anonymous devices on the users page, anonymous usage statistics, including the count of devices and users over this period.

Grafana UI:

• Navigate to Administration -> Users to access the anonymous devices tab.

• A new statistic has been added to the Usage & Stats page, displaying active anonymous devices from the last 30 days.

We’re excited to announce a big update to our Synthetic Monitoring product!

Until now, Synthetic Monitoring has used the Prometheus blackbox exporter to test at the protocol level: HTTP, DNS, TCP, gRPC, and ICMP (for ping and traceroute). This worked well for health and uptime monitoring, but it didn’t cover the full range of synthetic monitoring use cases. With modern applications, it’s important to test not only single endpoints but also complex transactions and critical user journeys.

Today we’re launching a revamped Grafana Cloud Synthetic Monitoring, which integrates k6 OSS as part of the engine. This unlocks two new check types to expand your test coverage and increase confidence in your applications:

• MultiHTTP checks (generally available): A form-based approach for defining multi-step tests. Under the hood, the form generates a k6 script, so you don’t have to write code.
• k6 Scripted checks (public preview): Gives you the power and flexibility to define your workflow tests in JavaScript, using the k6 API to efficiently author tests.

With both new check types, you chain steps together to simulate a workflow, passing data from one step to the next. Like all other checks, you can set up Prometheus-style alerts to track SLOs and notify your team of any problems. You can also leverage the built-in dashboards to visualize results, analyze trends, and quickly resolve performance issues.

To learn more about the new check types, refer to the Check types documentation, announcement blog post, and the following video demo:

Use the Grafana Alerting - Grafana OnCall integration to effortlessly connect alerts generated by Grafana Alerting with Grafana OnCall. From there, you can route them according to defined escalation chains and schedules.

To learn more, refer to the Grafana OnCall integration for Alerting documentation.

When using the Grafana Kubernetes Monitoring Helm chart to configure with the configuration wizard, you can select IBM Cloud as a platform.

(Release 1.6.1)

New in Explore, more options for sharing your query results! The time picker in Grafana has always had two categories - relative (for example, now to 2 hours ago) and absolute (8am to 10am). Explore has allowed people to copy links and create short links with exactly what their time picker had set, but now additionally you have the option to get a link or create a short link with the time range made absolute, even if you are viewing relative time. This means whoever looks at your link will see exactly the data you are seeing, no matter when they open the link. This is available as a dropdown next to the existing shortlink button in Explore.

We’ve introduced enhanced tooltips as part of our standardization initiative, unifying the tooltip architecture for a consistent user experience across panels. Packed with features like color indicators, time uniformity, and improved support for long labels, these tooltips go beyond a cosmetic redesign, bringing fundamental changes to elevate your data visualization experience. Stay tuned for more updates!

To try out the new tooltips, enable the newVizTooltips feature toggle. Enhanced tooltips have been implemented for the following visualizations:

• Time series
• Trend
• Heatmap
• Status history
• Candlestick
• State timeline
• XY Chart
• and more coming soon!

Note: As this is an ongoing project, the dashboard shared cursor and annotations features are not yet fully supported.

The TraceQL query editor has been improved to facilitate the creation of TraceQL queries. In particular, it now features improved autocompletion, syntax highlighting, and error reporting.

We’ve made a number of improvements to the canvas visualization.

Enhanced flowcharting functionality

With this release, we’ve updated the canvas visualization to include much-requested flowcharting features. These improvements are:

• Addition of widely-used elements: cloud, parallelogram, and triangle.
• Addition of midpoint controls so that the connectors no longer have to be straight lines.
• Addition of more connector styles including dashed lines as well as corner radius and direction control.
• Horizontal and vertical snapping for connectors.
• Addition of rounded corner styling for elements.
• Ability to rotate elements in the canvas.

Universal data link support

We’ve updated data links so that you can add them to almost all elements or element properties that are tied to data. Previously, you could only add data links to text elements or elements that used the TextConfig object. This update removes that limitation.

With the newly added Infinite panning editor option, you can now view and navigate very large canvases. This option is displayed when the Pan and zoom switch is enabled.

To try out this feature, you must first enable the canvasPanelPanZoom feature toggle.

InfluxDB introduced a new version, 3.0, in April. With this new version, InfluxDB has put Flux in maintenance mode. But with the new version we have a new querying language, Native SQL. With v10.3.0, Grafana has built-in support for SQL query language in InfluxDB.

All you need to do is set up your InfluxDB Cloud Account and create your InfluxDB data source on Grafana with the query language “SQL” selected.

Administrators are now able to directly install, update and uninstall plugins from within Grafana itself, removing the need to switch context and navigate to Grafana.com. This makes it easier than ever to manage plugins in your Grafana Cloud instance and reduces the risk of attempting to install plugins which are incompatible with your Grafana configuration.

This feature will gradually roll out to all Grafana Cloud users with no action required.

To learn more, refer to the following video:

To provide quicker identification and troubleshooting, the home page contains a snapshot of issues that reach the following thresholds:

• Pods that have been in a non-running state for more than 15 minutes
• Nodes with CPU usage above 90% for more than five minutes
• Nodes using more than 90% of memory for more than five minutes
• Persistent Volumes with capacity above 90%
• Node disks with capacity above 90%

(Release 1.3.1)

The Kafka integration is available for use in Kubernetes Monitoring. (Release 1.3.6)

(Re-)introducing “Keep Last State” to Grafana managed alert rules.

You can now choose to keep the last evaluated state of an alert rule when that rule produces “No Data” or “Error” results. Simply choose the “Keep Last State” option for no data or error handling when editing a rule. Refer to the Alerting documentation on state and health of alert rules for more information.

The Datadog data source now supports log aggregation. This feature helps aggregate logs/events into buckets and compute metrics and time series. For more information, refer to Datadog log aggregation.

The Log Volume Explorer has been promoted to General Availability! First announced at ObsCon London in Nov 2023, the Log Volume Explorer adds to our set of features that help you manage your observability spend. Identify the source of logs ingested into Grafana Cloud and answer questions like “Which of my teams, environments, Kubernetes clusters, and/or applications are generating the most logs?” Find the Log Volume Explorer under Administration > Cost Management > Logs.

Table view was created to help facilitate ease of use in a point and click UI, as opposed to data source specific query language formatting options, like loki’s line_format.

Tables can be configured and shared with team members via explore URLs or by adding the table to a dashboard panel.

Logs Table view is currently rolling out across Grafana Cloud. Try it out by switching on the `logsExploreTableVisualisation` feature toggle.

Introducing several improvements to the Loki data source.

Line filter “or” syntax

Loki’s line filter syntax is great to find specific substrings of your log lines. If users want to find multiple different substrings it was cumbersome to use the regex =~ operator. With this change it is possible to chain multiple strings with the existing filter operators.

Example:

Copy

{app="foo"} |= "foo" or "bar" != "baz" or "qux"

Filter based on label type

Grafana users can use the action buttons in the log details to filter for specific labels. Those would be always added as a LabelFilter expression regardless of the type of the label. Now, filtered labels will be added either to the stream selector if the label is an indexed label, or as a LabelFilter expression if the label is a parsed label or part of structured metadata.

Derived fields based on labels

Derived fields or data links are a concept to add correlations based on your log lines. Previously it was only possible to add derived fields based on a regular expression of your log line and doing it based on labels was not possible. With this change derived fields can be added either based on a regex of a log line or based on a label, parsed label or structured metadata.

The following example would add the derived field traceID regex based on a regular expression and another app label field based on the app label.

Grafana AWS Observability offers an out-of-the-box, embedded experience for you to efficiently explore and analyze your Amazon EC2 data.

The list of EC2 instances is available from the Overview tab.

In addition to sorting and filtering capabilities, you can:

• Choose the timeframe you want to view by using the time range selector.

• View the details of each instance.

• Switch to the Regions tab to view how instances scale by region.

• Determine how to allocate resources based on utilization and scaling demands you discover in the Rightsizing tab.

Grafana AWS observability introduces monitoring for your EC2 instances, all within an embedded experience, and without the need for managing or configuring local agents or building dashboards.

From the list of scrape jobs, you can navigate to your list of EC2 instances shown on the Overview tab, and drill into instance details.

The Regions tab shows the instance count within a selected time frame, and the Rightsizing tab shows disk write and read bytes as well as CPU utilization.

You can use these EC2 pages to:

• Filter, sort, and add or remove tags for more focus and granularity.
• Uncover anomalies in your data.
• Identify instances that do not contain expected tags.
• Discover potential issues, such as instances with higher CPU than normal or instances that are about to burst or hit IO limits.
• View how your instances scale over a region.
• Choose a tag to see how the instance autoscales.
• Learn which instances are not fully utilized and should be repacked, and which should be scaled down due to lack of need.

We’ve added two new features to transformations in Grafana, to help you make sense of your data.

Moving average

Sometimes your data is too noisy to quickly grasp what’s going on. A common way to address this issue is to calculate the moving mean, or moving average, to filter out some of that noise. Luckily, many data sources already support calculating the moving mean, but when the support is lacking or you’re not well versed in the query language, until now, you were stuck with the noise.

By selecting the Window functions mode and using Mean as the calculation for the Add field from calculation transformation, Grafana adds a field with the moving mean for your selected field.

The Window functions mode also supports moving variance and moving standard deviation calculations if you need to analyze the volatility of your metric.

Trend lines

We’re also adding some basic statistical analysis features as a way to help you visualize trends in your data. The Regression analysis transformation will fit a mathematical function to your data and display it as predicted data points in a separate data frame.

The transformation currently supports linear regression and polynomial regression to the fifth-degree.

We hope these two features will help you better visualize trends in your data.

Introducing Multi-Stack Data Sources

It’s complicated to query across multiple data sources if you store your metrics or logs across separate tenants.

Multi-stack data sources lets you set up one data source that includes multiple tenants, ensuring simpler faster queries.

The SLO App generates dashboards to help user pinpoint where they are burning their error budget, in which clusters, for example, by supporting multidimensional SLOs (SLOs that preserve one or more dimensions / label-values). The dashboards that help identify in which dimension the SLI is underperforming were previously only available for ratio-type SLOs or SLO expressions that were fairly simple.

ex:

Copy

sum by (cluster) (rate(http_requests_total{code!~"5.."}[$__rate_interval]))
/ sum by (cluster) (rate(http_requests_total[$__rate_interval]))

But with new changes in the parsing logic in the SLO app, we now support these dashboard features for complex SLOs, as long as they are ratios and have the same “group by” dimensions aggregating up in both the numerator and the denominator.

ex:

Copy

(sum by (cluster, namespace) (rate(request_duration_seconds_bucket{status_code!~"5..", le="1.0", route="opentelemetry_proto_collector_trace_v1"}[$__rate_interval]))
-
(
  sum by (cluster, namespace) (rate(envoy_cluster_grpc_proto_collector_trace_v1_TraceService_1[$__rate_interval]))
  or
  sum by (cluster, namespace) (build_info) * 0
))
/ (sum by (cluster, namespace) (rate(request_duration_seconds_count{route="opentelemetry_proto_collector_trace_v1"}[$__rate_interval])))

The TraceQL query editor has been improved to facilitate the grouping of multiple spans per trace in TraceQL queries. For example, when the following by(resource.service.name) is added to your TraceQL query, it will group the spans in each trace by resource.service.name.

By leveraging structured metadata under the hood, Cloud Logs now provides a simplified syntax and better query performance for logs sent through our Cloud OTLP endpoint.

Directly from Kubernetes Monitoring, you can click into the Asserts RCA Workbench to see what is occurring between components and their relationships.

(Release 2.0.4)

The improved navigation menu gives you a better overview by showing all levels of navigation items in a more compact design. We also implemented a better dock and imprved scrolling behavior. Furthermore, we improved the structure of the nav menu and added several new items.

The new browse dashboards interface features a more compact design, making it easier to navigate, search for, and manage for your folders and dashboards. The new interface also has many performance improvements, especially for instances with a large number of folders and dashboards.

To make using folders easier and more consistent, there is no longer a special General folder. Dashboards without a folder, or dashboards previously in General, are now shown at the root level.

To learn more, refer to the video demo:

View the details of any container, pod, workload & namespace with these improved pages, which include CPU and memory utilization graphs and cost details.

The container page also shows adjacent restart and termination panels so you can visually correlate more quickly the restart and termination events, along with the reason for termination.

(Release 1.5.0)

AWS data sources have migrated to new form styling in Configuration and Query editors in order to comply with the Grafana design system guidelines (found here) . This feature enables Public Preview of the migrated forms for AWS data sources:

• Cloudwatch
• X-Ray
• Athena
• Redshift
• Sitewise
• Twinmaker

It includes the following changes:

• revamped data source configuration forms, including block input fields, descriptions and separated sections
• a new look and feel for data source query editors

The data sources must be updated to their latest versions in order to make use of the feature.

We’ve revamped the Transformations user interface to make it cleaner, more user-friendly, and overall better for visualizing, selecting, and comprehending transformation options for your data.

Improved UI

In the past, transformations were applied through a dropdown menu, indicated solely by text names like Merge, Sort, JoinByLabels, etc. Now, we’ve introduced a much more user-friendly interface. A convenient drawer allows seamless access to all transformation options, each accompanied by visual/graphical representations and a brief description. These enhancements are designed to enhance the user’s comprehension of their data transformation choices.

In-App documentation

We’ve also streamlined the user experience by integrating documentation directly into the core Grafana application. Gone are the days of navigating to a separate browser page for Transformation documentation. Now, users can conveniently access documentation within the app interface, providing a more seamless and efficient way to understand and utilize various features. This enhancement aims to save time and enhance user convenience, ensuring that valuable information is readily available at their fingertips.

We’re excited to introduce the “No basic role,” a new basic role with no permissions. A basic role in Grafana dictates the set of actions a user or entity can perform, known as permissions. This new role is especially beneficial if you’re aiming for tailored, customized RBAC permissions for your service accounts or users. You can set this as a basic role through the API or UI.

Previously, permissions were granted based on predefined sets of capabilities. Now, with the “No basic role,” you have the flexibility to be even more granular.

For more details on basic roles and permissions, refer to the documentation.

To assign the “No basic role” in your Grafana Cloud stack, contact Grafana Support and ask them to enable the skip_org_role_sync feature toggle. You’ll be able to change basic roles that are synced using GCom.

PagerDuty enterprise data source plugin for Grafana allows you to query incidents data or visualize incidents using annotations.

Plugin is currently in a preview phase.

You can find more information and how to configure the plugin in the documentation.

Screenshots:

Introducing a major performance improvement for the PDF export feature.

Are you tired of waiting for your PDF to be generated or your report to be sent? We’re working on a major update of the dashboard-to-PDF feature to make it faster for large dashboards. The generation time will no longer be proportional to the number of panels in your dashboard. As an example, an SLO dashboard containing around 200 panels has gone from taking more than seven minutes to be generated to only eleven seconds.

This update also fixes all caveats related to rendering a report with panels or rows set to repeat by a variable, like rendering repeating panels inside collapsed rows.

To try out this update, enable the newPDFRendering feature toggle.

With the current release, we enabled RBAC permission validation (rbac.permission_validation_enabled setting) by default. This means that the permissions provided in the request during custom role creation or update are validated against the list of available permissions and their scopes. If the request contains a permission that is not available or the scope of the permission is not valid, the request is rejected with an error message.

You can now plot enum values in your time series and state timeline visualizations. This feature is useful when you want to visualize the state of a system, such as the status of a service or the health of a device. For example, you can use this feature to visualize the status of a service as ON, STANDBY, or OFF. To display enum values you can use the convert field transform.

The Grafana plugins catalog contains over 200 data sources, panels and applications to help you make the most of your data. However, finding the right plugin can sometimes be a challenge. 

Our initial search capability required an exact match on the name of the given plugin. 

With our latest update, the new fuzzy search algorithm lets you combine ID, name, plugin type, organization name, and keywords in one input, making it more flexible and accurate. It can handle partial matches and minor mistakes, like a missing letter.

To learn more, refer to the following video:

Click Predict Memory usage to predict namespace memory usage on the namespace detail page. (Release 1.2.1)

Public dashboards allow you to share your visualizations and insights to a broader audience without the requirement of a login. You can effortlessly use our current sharing model and create a public dashboard URL to share with anyone using the generated public URL link. To learn more, refer to the Public dashboards documentation, as well as the following video demo:

The Tempo data source now supports query-type template variables. With this update, you can create variables for which the values are a list of attribute names or attribute values seen on spans received by Tempo.

To learn more, refer to the following video demo, as well as the Grafana Variables documentation.

The Alerting Provisioning HTTP API has been updated to enforce Role-Based Access Control (RBAC).

• For Grafana OSS, users with the Editor role can now use the API. 
• For Grafana Enterprise and Grafana Cloud, users with the role Rules Writer and Set Provisioning status can access the API and limit access to alert rules that use a particular data source. 
• Other roles related to provisioning, for example Access to alert rules provisioning API still work.

With recorded queries, a single recorded query can now record multiple metrics.

To reduce the noise of flapping alerts, you can set a recovery threshold different to the alert threshold.

Flapping alerts occur when a metric hovers around the alert threshold condition and may lead to frequent state changes, resulting in too many notifications being generated.

Introducing query caching for async queries in the Athena and Redshift data source plugins. We previously introduced async queries for the Athena and Redshift plugins, and this feature adds support for caching those queries. To use this, you must have query caching enabled for the Athena or Redshift data source you wish to cache. This feature was previously available behind a feature toggle and is now generally available and enabled by default.

Removal of old Tempo Search tab

In Grafana v10.1, we added a Tempo search editor powered by TraceQL (search tab). We also recommended using this new editor over the older non-TraceQL powered editor.

The older non-TraceQL powered editor has been removed. Any existing queries using the older editor will be automatically migrated to the new TraceQL-powered editor.

The new TraceQL-powered editor makes it much easier to build your query by way of static filters, better input/selection validation, copy query to the TraceQL tab, query preview, dedicated status filter, and the ability to run aggregate by (metrics summary) queries.

Refer to Query tracing data to learn more.

Removal of Loki Search tab in Tempo

The Loki Search tab has been around since before we could natively query Tempo for traces. This search is used by a low number of users in comparison to the TraceQL-powered editor (Search tab) or the TraceQL tab itself.

If you would like to see what logs are linked to a specific trace or service, you can use the Trace to logs feature, which provides an easy way to create a custom link and set an appropriate time range if necessary.

When you’re browsing Grafana - for example, exploring the dashboard and metrics related to an alert - it’s easy to end up far from where you started and hard get back to where you came from. The ‘Return to previous’ button is an easy way to go back to the previous context, like the alert rule that kicked off your exploration. This first release works for Alerts, and we plan to expand to other apps and features in Grafana in future releases to make it easier to navigate around.

Return to Previous is rolling out across Grafana Cloud now. To try Return to Previous in self-managed Grafana, turn on the returnToPrevious feature toggle in Grafana v10.4 or newer.

NOTE
The term context refers to applications in Grafana like Incident and OnCall, as well as core features like Explore and Dashboards.

To notice a change in your context, look at Grafana’s breadcrumbs. If you go from Home > Dashboards to Home > Explore, you’ve changed context. If you go from Home > Dashboards > Playlist > Edit playlist to Home > Dashboards > Reporting > Settings, you are in the same context.

You can now map Google groups to Grafana organizational roles when using Google OIDC. This is useful if you want to limit the access users have to your Grafana instance.

We’ve also added support for controlling allowed groups when using Google OIDC.

Refer to the Google Authentication documentation to learn how to use these new options.

Grafana Alerting previously evaluated rules at the start of the evaluation interval. This created a sudden spike of resource utilization, impacting data sources. Rule evaluation is now spread over the entire interval for smoother performance utilization of data sources.

Dashboards, when accessed by users with the Viewer role, are now using the Scenes library. Those users shouldn’t see any difference in the dashboards apart from two small changes to the user interface (UI): the variables UI has slightly changed and the time picker is now part of the dashboard container.

Dashboards aren’t affected for users in other roles.

This is the first step towards a more robust and dynamic dashboarding system that we’ll be releasing in the upcoming months.

For the past few months we’ve been working on a major update of our Dashboards architecture and migrated it to the Scenes library. This migration provides us with more stable, dynamic, and flexible dashboards as well as setting the foundation for what we envision the future of Grafana dashboards will be. Here are two of the improvements that are being introduced as part of this work:

Edit mode

It can be difficult to efficiently navigate through the visually cluttered options during the dashboard editing process. With the introduction of the edit mode, we aim to provide an easier way to discover and interact with the dashboard edit experience.

Fixed positioning of template variables and time picker

We moved the time picker into the dashboard canvas and now, together with template variables, it will stick to the top as you scroll through your dashboard. This has historically been a very requested feature that we’re very happy to be able to finally roll out!

Known limitations

• The variable dependency graph is not yet available.
• It’s no longer possible to switch a regular panel to a library panel from the edit view.

If you want to learn more, in detail, about all the improvements we’ve made, don’t miss our blog post.

Using Static IPs with k6 allows you to know which IP addresses will be used by the cloud instances generating load during your test runs. This allows you to whitelist these IP addresses only. Knowing the IP addresses that are used can make it easier to identify logs and traffic created specifically by the k6 load tests.

Static IPs are only available to Grafana Cloud Pro, Grafana Cloud Advanced and Enterprise customers. Accounts are limited to 5 static IP addresses but if more IPs are needed customers can contact customer support.

We’ve added the option to manage library panel permissions through role-based access control (RBAC). With this feature, you can choose who can create, edit, and read library panels. RBAC provides a standardized way of granting, changing, and revoking access when it comes to viewing and modifying Grafana resources, such as dashboards, reports, and administrative settings.

You now have the ability to customize specific colors for individual thresholds when using the Config from query results transformer. Previously, when you added multiple thresholds, they all defaulted to the same color, red. With this addition, you gain the flexibility to assign distinct colors to each threshold.

This feature addresses a common pain point highlighted by users. With customizable threshold colors, you now have greater control over your data representation, fostering more insightful and impactful analyses across diverse datasets.

This feature will be rolled out over the next few weeks.

You can add an expiration date when you create a Grafana Access Policy token during the easy configuration process of Kubernetes Monitoring. The permission scope of the token also displays as you configure.

Release v1.7.0

This feature simplifies your options for configuring where your notifications are sent when an alert rule fires. Choose an existing contact point directly from within the alert rule creation form without the need to label match notification policies.  You can also set optional muting, grouping, and timing settings directly in the alert rule.

Simplified routing inherits the alert rule RBAC, increasing control over notification routing while preventing accidental notification policy updates, ensuring critical notifications make it to their intended contact point destination.

To try out Simplified Alert Notification Routing enable the alertingSimplifiedRouting feature toggle.

The SLO Performance page provides a filterable view as to how your team and service SLOs are performing. This is critical for customers with a large amount of SLOs who would like to view a logical subset. To get started - tag your SLOs with the team_name and/or service_name labels.

Configuring OAuth providers was a bit cumbersome in Grafana: Grafana Cloud users had to reach out to Grafana Support, self-hosted users had to manually edit the configuration file, set up environment variables, and then they had to restart Grafana. On Cloud, the Advanced Auth page is there to configure some of the providers, but configuring Generic OAuth hasn’t been available until now and there was no way to manage the settings through the Grafana UI, nor was there a way to manage the settings through Terraform or the Grafana API.

Our goal is to make setting up SSO for your Grafana instance simple and fast.

To get there, we are introducing easier self-serve configuration options for OAuth in Grafana. All of the currently supported OAuth providers are now available for configuration through the Grafana UI, Terraform and via the API. From the UI, you can also now manage all of the settings for the Generic OAuth provider. 

We are working on adding complete support for configuring all other supported OAuth providers as well, such as GitHub, GitLab, Google, Microsoft Azure AD and Okta. You can already manage some of these settings via the new self-serve configuration options, and we’re working on adding more at the moment.

Configure with Grafana Kubernetes Monitoring Helm chart using a streamlined process. With this method, you can set on/off switches to gather metrics, logs, events, traces, and cost metrics. (Release 1.2.0)

Structured metadata is a feature in Loki and Cloud Logs that allows customers to store metadata that is too high cardinality for log lines, without needing to embed that information in log lines themselves.

It is a great home for metadata which is not easily embeddable in a log line, but is too high cardinality to be used effectively as a label.

For more information on how to configure and query structured metadata, please view our documentation.

Subfolders are here at last!

Some of you want subfolders in order to keep things tidier. It’s easy for dashboard sprawl to get out of control, and setting up folders in a nested hierarchy helps with that. 

Others of you want subfolders in order to create nested layers of permissions, where teams have access at different levels that reflect their organization’s hierarchy. 

We are thrilled to bring this long-awaited functionality to our community of users! Subfolders are currently being rolled out to Grafana Cloud instances and will be generally available to all Grafana users for the Grafana 11 release.

Just a quick note: the upgrade to enable subfolders can cause some issues with alerts in certain cases. We think these cases are pretty rare, but just in case, you’ll want to check for this: 

If you’ve previously set up a folder that uses a forward slash in its name, and you have an alert rule in that folder, and the notification policy is set to match that folder’s name, notifications will be sent to the default receiver instead of the configured receiver.

To correct this, take the following steps:

• Create a copy of the affected routes
• Rewrite the matchers for the new copy. For example, if the original matcher was grafanafolder=folder_with/in_title, then the new route matcher will be grafana_folder=folder_with/_in_title
• After rewriting the matchers, you can delete the old routes.

If you use file provisioning, you can upgrade and update the routes at the same time.

This update to the Filter data by values transformation simplifies data filtering by enabling partial string matching on field values thanks to two new matchers: Contains substring and Does not contain substring. With the substring matcher built into the Filter data by values transformation, you can efficiently filter large datasets, displaying relevant information with speed and precision. Whether you’re searching for keywords, product names, or user IDs, this feature streamlines the process, saving time and effort while ensuring accurate data output.

In the Filter data by values transformation, simply add a condition, choose a field, choose your matcher, and then input the string to match against.

This update will be rolled out to customers over the next few weeks.

Sumo Logic enterprise data source plugin for Grafana lets users query metrics and logs from their Sumo Logic instance.

Plugin is already live in v 1.0

You can find more information and how to configure the plugin in the documentation.

Screenshots:

The Opsgenie contact point has been extended to allow users to optionally fill out responder information for their integration. Responders tell Opsgenie who an alert should notify according to their escalation policies and routing rules.

Previously, the only transformation that supported dashboard variables was the Add field from calculation transformation. We’ve now extended the support for variables to the Filter by value, Create heatmap, Histogram, Sort by, Limit, Filter by name, and Join by field transformations.

We’ve also made it easier to find the correct dashboard variable by displaying available variables in the fields that support them, either in the drop-down or as a suggestion when you type $ or press Ctrl + Space:

A SurrealDB data source has been added to the Plugin Catalog, enabling the integration of SurrealDB, a real-time, multi-model database, with Grafana’s visualization capabilities. This datasource allows users to directly query and visualize data from SurrealDB within Grafana, using SurrealDB’s query language.

The SurrealDB data source launches with just the basics today. You can write queries in SurrealQL using the built-in query editor, although many Grafana features like macros are not supported for now. 

You can find more information and how to configure the plugin on Github.

Grafana Synthetics now supports longer durations for two options:

Timeout -> Increased maximum to 60 seconds.
Frequency -> Increased maximum to 60 minutes.

We’ve improved the reporting experience with options to make all of your table data accessible in PDFs. Previously, if your dashboard included large table visualizations, you couldn’t see all of the table data in your PDF report. Unlike in Grafana, you couldn’t scroll in the PDF table visualization or click on the page numbers. With this new feature, you now have the option to see all the data directly in your PDF without losing your dashboard layout.

We’ve added two format options to the report creation form:

• Include table data as PDF appendix - Adds an appendix to your dashboard PDF.
• Attach a separate PDF of table data - Generates a separate PDF file for your table panel data.

To try out this feature, enable the pdfTables feature toggle or contact Grafana Support to have it enabled in on your Grafana Cloud stack.

The table visualization for logs, announced in public preview for Grafana 10.3, is generally available in Cloud (all editions) and with Grafana 10.4.

New to the table visualization with 10.4:

• the ability to sort columns
• data type autodetection of fields
• autodetection and clean formatting of json fields

Try it out today!

Quickly switch between the Cluster, namespace, workload, and Node views on the Cluster Navigation page using tabs. (Release 1.3.0)

It is hard for teams to collaborate on dashboards because they have to use different datasources. Grafana instances become cluttered and confusing with 100s of datasources.

Team LBAC (Label Based Access Control) is our first step towards seamless management of Grafana Teams access for Loki logs. Each team views the same data source filtered by their team’s label permissions.

Feature highlights

• Teams can view queries to the same data source with different LBAC rules applied
• Configurable via API and the UI
• Simplified LBAC notation is automatically converted in correct custom headers

Best practices

• We recommend you only add query permissions for teams that will have LBAC rules, and remove default Viewer and Editor query permissions.
• As an initial setup, we recommend defining as few rules as possible for each team, making sure that they are additive rather than negating one another.
• For validating rules, we recommend testing each rule in Loki Explore view. This allows you to see the logs that would be returned for the specific rule.

We’ve added an Aggregate By option to the TraceQL query editor to leverage Grafana Cloud Traces’ metrics summary API. You can calculate RED metrics (total span count, percent erroring spans, and latency information) for spans of kind=server received in the last hour that match your filter criteria, grouped by whatever attributes you specify.

This feature is disabled by default. To enable it, file contact Grafana Support.

For more information, refer to the documentation.

The Grafana Assume Role authentication provider lets Grafana Cloud users of the CloudWatch data source authenticate with AWS without having to create and maintain long term AWS Users. Using the new assume role authentication method, you no longer have to rotate access and secret keys in your CloudWatch data source. Instead, Grafana Cloud users can create an identity access and management (IAM) role that has a trust relationship with Grafana’s AWS account; Grafana’s AWS account will then use AWS Secure Token Service (STS) to create temporary credentials to access the user’s AWS data.

To learn more, refer to the CloudWatch authentication documentation.

The SAP HANA® data source now supports tenant databases connections by using the database name and/or instance number. For more information, refer to SAP HANA® configuration.

To aid in understanding historical data, a time range selector is available on every page of Kubernetes Monitoring where this function is appropriate. This selector is also available on the Workloads detail page, for detecting outlier CPU usage in the Pods. (Release 1.4.0)

We’ve made a number of small improvements to the way tooltips work in Grafana. To try out the new tooltips, enable the newVizTooltips feature toggle.

Copy on click support

You can now copy the content from within a tooltip by clicking on the text.

Scrollable content

You can now scroll the content of a tooltip, which allows you to view long lists. This is currently supported in the time series, candlestick, and trend visualizations. We’ll add more improvements to the scrolling functionality in a future version.

Added tooltip options for candlestick visualization

The default tooltip options are now also visible in candlestick visualizations.

Hover proximity option in time series

We’ve added a tooltip hover proximity limit option (in pixels), which makes it possible to reduce the number of hovered-over data points under the cursor when two datasets are not aligned in time.

Using Trace to profiles, you can use Grafana’s ability to correlate different signals by adding the functionality to link between traces and profiles.

Trace to profiles lets you link your Grafana Pyroscope data source to tracing data. When configured, this connection lets you run queries from a trace span into the profile data.

There are two ways to configure the trace to profiles feature:

• Use a simplified configuration with default query, or
• Configure a custom query where you can use a template language to interpolate variables from the trace or span.

To try out Trace to profiles, enable the ’traceToProfiles’ feature toggle.

If you would also like to try out the Embedded Flame Graph feature, please enable the ’tracesEmbeddedFlameGraph’ feature toggle.

Note: in order to determine that there is a profile for a given span and render the ‘Profiles for this span’ button or the embedded flame graph in the span details, the ‘pyroscope.profile.id’ key-value pair must exist in your span tags.

Collect traces when you configure Kubernetes Monitoring, and then use Tempo to create search queries. Refer to Navigate to traces for more information. (Release 1.3.0)

You can use Grafana Cloud k6 cloud options to configure additional test parameters, such as how to distribute your test across load zones, or to use static IP addresses.

Before, you would use the options.ext.loadimpact object to pass any configuration parameters. Now, you can use the options.cloud object instead:

JavaScript Copy

export const options = {
  cloud: {
    name: 'Hello k6 cloud!',
    projectID: 123456,
    staticIPs: true,
  }
}

The options.ext.loadimpact object will be deprecated in the future. For more details about cloud options, refer to Cloud options.

CPU and memory usage graphs and details, along with associated costs, are available on these pages in Kubernetes Monitoring:

• Namespace details page
• Workload detail page

(Release 1.5.0)

You can now use generative AI to assist you in your Grafana dashboards. So far generative AI can help you with the following tasks:

• Generate panel and dashboard titles and descriptions - You can now generate a title and description for your panel or dashboard based on the data you’ve added to it. This is useful when you want to quickly visualize your data and don’t want to spend time coming up with a title or description.
• Generate dashboard save changes summary - You can now generate a summary of the changes you’ve made to a dashboard when you save it. This is great for effortlessly tracking the history of a dashboard.

To enable these features, you must first enable the dashgpt feature toggle. Then install and configure Grafana’s LLM app plugin. For more information, refer to the Grafana LLM app plugin documentation.

When enabled, look for the ✨ Auto generate option next to the Title and Description fields in your panels and dashboards, or when you press the Save button.

You can now use generative AI to assist you in your Grafana dashboards. So far generative AI can help you generate panel and dashboard titles and descriptions - You can now generate a title and description for your panel or dashboard based on the data you’ve added to it. This is useful when you want to quickly visualize your data and don’t want to spend time coming up with a title or description.

Make sure to enable and configure Grafana’s LLM app plugin. For more information, refer to the Grafana LLM app plugin documentation.

When enabled, look for the ✨ Auto generate option next to the Title and Description fields in your panels and dashboards, or when you press the Save button.

Save money and apply an extra layer of network security by using Azure Private Link to send metrics, logs, traces, and profiles to Grafana Cloud.

Normally when you send telemetry from Microsoft Azure to Grafana Cloud, you incur network egress fees above a certain traffic volume and your data, though encrypted, traverses the public internet. With Azure Private Link, traffic between your virtual network and Grafana Cloud travels the Microsoft backbone network, so exposing your service to the public internet is no longer necessary.

To set up Azure Private Link in Grafana Cloud, follow the instructions from our public documentation.

Control the automatic refresh interval of the GUI as well as disable the auto refresh until you are ready to do so manually. This is particularly useful for very large Kubernetes fleets that display a lot of data.

From the Clusters, Nodes, and Namespaces list pages, select multiple Clusters from the filters.

Grafana can now be used to manage both Prometheus and Mimir Alertmanagers with UTF-8 configurations. For more information, please see the release notes for Alertmanager 0.27.0.

You can now view percent change in stat visualizations. This makes it easier to understand your data by showing how metrics are changing over time.

You can now zoom in on the y-axis of your time series and candlestick visualizations. This is useful when you want to focus on a specific range of values. To zoom in on the y-axis on supported visualizations, hold the Shift key while clicking and dragging; double-click to reset the zoom.