Skip to content

v1.45.0

Compare
Choose a tag to compare
@netdatabot netdatabot released this 21 Mar 13:48
· 517 commits to master since this release

Table of Contents

Netdata Growth

  • 67.5k GitHub stars!
  • 626M Docker Hub pulls!

Thanks to your love ❤️, Netdata is leading the observability category in CNCF, having significantly more stars than Elasticsearch, Grafana, Prometheus and all other observability solutions listed in CNCF landscape.

We are committed to provide the most advanced and innovative observability solution, to help us minimize monitoring costs while providing AI-powered high-fidelity monitoring!

You like Netdata? Give Netdata a ⭐ too, on GitHub!

Release Summary

3 months have passed since the previous Netdata release. A lot has changed since then! Netdata now has a mobile app for alert notifications, new drag-and-drop custom dashboards, network connections monitoring, dynamic configuration for data collection jobs and alerts, and many more...

To see how Netdata stacks up against the most advanced commercial offerings available today, we did an analysis on how Dynatrace, Datadog, Instana, Grafana and Netdata commercial offerings compare.

It is nice to see that Netdata stands out for its:

  1. Excellent technology coverage

    Netdata's monitoring coverage is significantly higher compared to others, in all areas!

  2. High-fidelity, real-time insights

    Netdata is the only monitoring solution offering this kind of fidelity (per-second for all metrics), at this extend!

  3. Real AI and Machine-Learning.

    Netdata is the only monitoring system that offers real machine learning, running at the edge.

  4. Lightweight

    Netdata is among the lightest agents, despite the fact that it does a lot more than the others.

  5. Best cost efficiency

    Netdata's cost efficiency is unbeatable, making Netdata the most cost-efficient monitoring solution available today!

Read the full blog here.

Release Highlights

Netdata Mobile App

You can now receive Netdata alerts directly on your mobile phone!

Choose your space and see all the available notifications since you last signed in!

Check the full demo here.

The Mobile App is available for Homelab and Business plan users.

Custom Dashboards

You can now create advanced custom dashboards with Netdata!

  • Drag-and-Drop

    Easily move charts from Metrics or Single Node views straight to your dashboards. It's intuitive and fun.

  • New Chart Types

    Discover your data in new ways with Bar, Circle, Gauge, Pie, Value, and Group boxes.

  • Quick Dashboard Creation

    Hit the plus button, drag, and you've got a new dashboard. Simple as that.

  • Rename Charts

    Customize your dashboard by renaming charts to whatever makes sense to you.

  • Refreshed Text Cards

    We've upgraded text cards for better clarity and aesthetics.

On the Agent UI and the Community plan of Netdata Cloud 1 custom dashboard is allowed. The Homelab and Business plans of Netdata Cloud support an unlimited number of custom dashboards.

Network Viewer

Explore the network connections of your servers and processes!

302690196-5f71c102-9146-463e-acba-329094b136a5

Netdata got a network viewer (select network-connections from the Top tab inside the dashboard).

The tool reports all IPv4 and IPv6, TCP and UDP sockets a system and all its processes have. Also, it automatically and reliably classifies them as inbound, outbound, local (i.e. within the host itself), or listen (for daemons).

The visualization graph has 4 sides:

  • public (i.e. public IPs),
  • private (i.e. private and reserved IPs),
  • servers (i.e. listening and inbound sockets),
  • clients (i.e. sockets towards other servers).

The position of each application on the chart is determined by the classification of the sockets it has. To the top are clients, to the bottom are servers, to the right are internet facing applications, to the left is internal network applications.

The size of each application in the chart is determined by the number of sockets it has, and each application is a pie chart representing the percentage of each kind of sockets it has.

For servers with dozens of thousands of sockets, the tool provides an aggregated view, grouping similar sockets together and reporting the total. Users can switch to a detailed view from the UI.

User Settings

We've improved immensely the customization capabilities of Netdata with the introduction of User Settings.

Our first release on this front is focused on the customization of charts, either on the Metrics tab or the Single Node view tab. You can now create for any chart:

  • Personal views
  • Room specific views
  • Space dedicated views

With this, you can define what is best for your team to visualise a given chart but still allow each teammate to define their own. Users will be presented with the view they should see, based on setting hierarchy, but they
can interchangeably select which of the select views they want.

chrome_UPtwnDlNyd

More areas of customization will come soon, Filters saved views, Dashboards Table of Content (TOC) ordering, etc.

Dynamic Configuration (beta)

Netdata agents are now deployed with the ability to dynamically accept configuration from the UI, for data collection jobs and alerts. The feature is released in beta.

Alerts Configuration Manager

The Alerts Configuration Manager is transforming the way users configure and manage alerts in their Netdata environment. This powerful tool integrates directly into the Netdata Dashboard, offering a streamlined and intuitive interface for both novice and experienced users.

Check the full demo here

Alerts Silencing Rules

g6oHMtz

🔕Improvements done to make it easier to interact and see Alert Silencing Rules.
With this release, you will be able to:

  • See silencing rules status directly on entities like Alerts, Rooms, and Nodes
  • Immediately create a silencing rule for an Alert, a Room, or a Node

We hope this makes it easier for you to interact with the Alert Silencing Rule Manager.
Stay tuned for more improvements!

MacOS Processes Monitoring

Netdata's apps.plugin has been ported to macOS, allowing users to view processes information on Linux, FreeBSD and macOS!

Just install the latest Netdata on your macOS and enjoy full processes monitoring!

Homelab Plan

For non-professional use, get the whole and the latest of Netdata! For the cost of a beer per month, you can get access to all Business features of Netdata, for your home lab or personal project!

Our Homelab plan is available to technology enthusiasts and students, for non-professional user, offering the entire Netdata suite, for a small flat fee, under a fair usage policy.

  • Unlimited Access: Enjoy the freedom of unlimited usage, with no caps on nodes or custom dashboards.

  • Premium Features: Get your hands on business-level features, including enhanced alert integrations and access to our mobile app, all tailored for your personal projects.

  • Support Netdata: Support the open-source Netdata, to ensure it will be there for you, when you need it!

New Build Infrastructure

Starting with Netdata 1.45, we have completely removed our GNU Autotools based build system and replaced it with
CMake. The new CMake build system has a number of significant benefits for developers, package maintainers, and
those using local builds of Netdata.

See details
  • We now have proper support for out-of-tree builds, and this is now the preferred method for building Netdata.
  • Build configuration is now measurably faster than it was previously.
  • Netdata can now be built using Ninja instead of Make, further speeding up the build process. The installer and
    updater script will automatically use Ninja instead of Make when possible, and developers, package maintainers,
    and users who are building by hand are encouraged to explicitly use it themselves when building Netdata by
    specifying the -G Ninja option to CMake during the configuration process.
  • Overall maintenance of the build system will be significantly easier going forwards. This means we should be
    able to fix any issues involving it much more quickly, and contributions from external developers should be
    much easier.
  • A number of features we have wanted to add to our build infrastructure will be much easier to add now.

Most users should not be directly affected by this change other than benefiting from the faster build times,
only those who were building locally by hand (not using the netdata-installer.sh script or the kickstart script)
will need to change things.

Go Plugin Moved to Main Repository

Alongside the new CMake build system, we have also moved the go.d.plugin
code from the netdata/go.d.plugin repository to the main netdata/netdata
repository.

See details

We have made this change for three reasons:

  • It makes handling of bugs in the Go plugin much easier. Instead of possibly needing to track issues and PRs
    across two repositories, now everything should end up tracked coherently in one repository.
  • It lets us significantly simplify a number of parts of our CI and installation code, allowing for greater
    reliability and easier maintenance.
  • It provides a testbed for infrastructure for handling of Go in the main repo, which is significant as we have
    been internally looking at reimplementing some other components of the agent in Go.

Users of native packages and static builds should see no difference at all from this change.

Building the agent locally will now require a working Go toolchain supporting a particular minimum version of the
Go language (currently 1.21) if the Go plugin needs to be built. The plugin itself can still be disabled to avoid
this requirement, but this is not recommended.

The installer code will attempt to ensure that a sufficiently up-to-date Go toolchain is installed when installing
or updating the agent. If such a toolchain is not found, it will attempt to automatically install a copy of the
official toolchain from https://go.dev/dl/ in /usr/local/go. If that attempt fails, the Go plugin will be
DISABLED automatically at build time.

Acknowledgments

  • @candlerb for improving robustness of netdata-updater.sh.
  • @carrychair for removing unnecessary repetition of words in documentation.
  • @luisj1983 for adding "Backing up a Netdata Agent" documentation.
  • @moschlar for fixing --distro-override parameter name in kickstart documentation.
  • @pschaer for correcting instructions on creating a startup script in the "Install Netdata on Synology" guide.
  • @sepek for fixing description of "chart labels" in "Configure alerts".

Contributions

Collectors

Improvements
  • Add macOS support for collecting resource usage of processes (apps.plugin) (#17180, @ktsaou)
  • Improve identification of applications in docker service discovery (go.d.plugin) (#17174, @ilyam8)
  • Execute local-listeners periodically rather than just once at startup (go.d.plugin) (#17160, @ilyam8)
  • Add service discovery for applications running inside Docker containers (go.d.plugin) (#17152, @ilyam8)
  • Implement dynamic configuration for configuring data collection jobs (go.d.plugin) (#17064, @ilyam8)
  • Update message IDs for systemd and dbus (systemd-journal.plugin) (#16987, @ktsaou)
  • Report EDAC ECC errors s total counts since boot instead of rates (proc/sys_devices_system_edac_mc) (#16970, @ilyam8)
  • Add aggregated view (network-viewer.plugin) (#16940, @ktsaou)
  • Add filtering by username (network-viewer.plugin) (#16911, @ktsaou)
  • Add Network Viewer plugin (#16872, @ktsaou)
  • Add CPU throttling % column to the containers-vms function (cgroups.plugin) (#16800, @ilyam8)
  • Add the ndsudo binary, a helper tool for assisting in the execution of privileged commands (#16614, @ktsaou)
  • Disable CPU per core metrics by default (proc.plugin) (#16572, @ilyam8)
Bug fixes
  • Fix incorrect family value of the ZFS ZPool state chart (proc/proc_spl_kstat_zfs) (#17054, @ilyam8)
  • Fix race conditions (diskspace.plugin) (#16786, @ktsaou)
  • Fix allocated memory after it has been freed (diskspace.plugin) (#16784, @ktsaou)
  • Fix priority per-core CPU charts (proc/proc_stat) (#16749, @ilyam8)
  • Fix missing CPU frequency chart (proc/proc_stat) (#16732, @ilyam8)
  • Fix an issue where cgroup_check_for_new_every was incorrectly multiplied by update_every (cgroups.plugin) (#16719, @ilyam8)
Other
  • Add mongodb-community-server image to docker service discovery configuration (go.d.plugin) (#17173, @ilyam8)
  • Add an option to disable service discovery (go.d.plugin) (#17171, @ilyam8)
  • Allow array/object to be null json schemas (go.d.plugin) (#17166, @ilyam8)
  • Update file path pattern in jsonschema (go.d.plugin) (#17164, @ilyam8)
  • Add support for multi-config templates in the service discovery configuration (go.d.plugin) (#17157, @ilyam8)
  • Improve go.d.plugin dyncfg config schemas (#17124, @ilyam8)
  • Fix incorrect chart priority for discovered configs (go.d.plugin) (#17115, @ilyam8)
  • Add notice log level (go.d.plugin) (#17112, @ilyam8)
  • Fix pulsar tests (go.d/pulsar) (#17093, @ilyam8)
  • Set max chart id length to 1200 (go.d.plugin) (#17062, @ilyam8)
  • Improve aggregated view (network-viewer.plugin) (#16960, @ktsaou)
  • Show unknown container (network-viewer.plugin) (#16900, @ktsaou)
  • Reorganise code to prepare for functions (ebpf.plugin) (#16788, @thiagoftsm)
  • Fix missing aral_freez call (eBPF) (#16765, @thiagoftsm)
  • Cleanup network devices rename (proc/proc_net_dev) (#16745, @ktsaou)
  • Improve ebpf-socket function column names (ebpf.plugin) (#16727, @ilyam8)
  • Add double-linked network interfaces collection delay (#16701, @ilyam8)
  • Cleanup code and improve reliability (ebpf.plugin) (#16669, @thiagoftsm)
  • Update to create a separate chart for each systemd service rather than a chart dimension (ebpf.plugin) (#16630, @thiagoftsm)
  • Include 'lxcfs.service/.control' in the list of filtered cgroups (cgroups.plugin) (#16620, @ilyam8)
  • Exit if unable to locate journal data directories (systemd-journal.plugin) (#16592, @ilyam8)

Health

All changes
  • Remove deprecated alert fields from stock alarms (#17113, @ilyam8)
  • Fix filtering by severity for gotify notifications (#17069, @ilyam8)
  • Remove deprecated alert fields: "charts", "os", "host", "plugin" and "module" (#17048, @ktsaou)
  • Add a new alert to notify about systemd timer units that have failed (#16845, @tkatsoulas)
  • Implement dynamically configured alerts (#16779, @ktsaou)
  • Add a new alert to detect unexpected HTTP headers (#16736, @ilyam8)

Packaging/Installation

All changes

Documentation

All changes
  • Improve "Choose your Netdata Cloud theme" doc (#17172, @Ancairon)
  • Add instructions for monitoring NVIDIA GPUs to the Docker installation guide (#17167, @ilyam8)
  • Add documentation for the "Integration URL" field to PagerDuty Cloud integration doc (#17149, @juacker)
  • Bring back old docs that were containing missing information (#17146, @Ancairon)
  • Remove unnecessary repetition of words in docs (#17131, @carrychair)
  • Fix broken link in "Netdata Cloud On-Prem Installation" (#17118, @tkatsoulas)
  • Fix typos and improve wording in "Backing up a Netdata Agent" (#17117, @Ancairon)
  • Remove deprecated settings from "Configure alerts" (#17116, @ilyam8)
  • Fix broken links in go.d.plugin markdown files (#17108, @ilyam8)
  • Remove deprecated "foreach" from "Configure alerts" (#17106, @ilyam8)
  • Remove distributed-data-architecture.md (#17097, @Ancairon)
  • Fix broken links (#17095, @Ancairon)
  • Remove docs/netdata-security.md (#17094, @Ancairon)
  • Update "Plugin Functions Tables" docs (#17071, @car12o)
  • Update "Sizing Netdata Agents" doc (#17057, @ktsaou)
  • Fix links pointing to old go.d repo and update the integrations (#17040, @Ancairon)
  • Update links to Netdata Agent start-stop-restart docs (#17037, @Ancairon)
  • Include information on securing Netdata parent-child communication in "Configuring Metrics Centralization Points" (#17035, @Ancairon)
  • Restructure and update documentation (#17014, @Ancairon)
  • Add "Backing up a Netdata Agent" documentation (#17006, @luisj1983)
  • Correct instructions on creating a startup script in the "Install Netdata on Synology" guide (#16980, @pschaer)
  • Improve formatting in "How to optimize the Netdata Agent's performance" (#16925, @tkatsoulas)
  • Fix links to the energy efficiency screenshots to main readme file (#16904, @Aliki92)
  • Update "What's New and Coming?"based on Office Hours shared plans to main readme file (#16895, @hugovalente-pm)
  • Improve readability of Webhook Cloud notification documentation (#16882, @juacker)
  • Remove deprecated db mode "save" from "Database" (#16864, @Ancairon)
  • Fix CNCF link (#16851, @hugovalente-pm)
  • Add documentation on how to configure MS Teams Cloud notifications (#16834, @papazach)
  • Added instructions on calculating replication history to "Streaming and Replication Reference" (#16816, @thiagoftsm)
  • Update provisioning instructions in "Netdata Cloud On-Prem Light PoC" (#16811, @M4itee)
  • Add information about the new node permissions to "Role-Based Access model" (#16791, @vkuznecovas)
  • Add missing settings to "Streaming and replication reference" (#16778, @thiagoftsm)
  • Fix instructions for setting up Telegram notifications (#16777, @thiagoftsm)
  • Updated the kickstart URL to https://get.netdata.cloud/kickstart.sh (#16738, @ilyam8)
  • Fix --distro-override parameter name in "Install Netdata with kickstart.sh" (#16726, @moschlar)
  • Add the Mobile App notification Integration (#16715, @sashwathn)
  • Add "Require Cloud" column to the functions table in "Netdata Functions" (#16681, @ilyam8)
  • Fix typos and improve wording in "Creating Alerts with Netdata Alerts Configuration Manager" (#16679, @Ancairon)
  • Fix description of "chart labels" in "Configure alerts" (#16656, @sepek)
  • Fix formatting in "Creating Alerts with Netdata Alerts Configuration Manager" (#16651, @Ancairon)
  • Add practical examples showcasing how to utilize journalctl for querying Netdata logs to "Netdata Logging" (#16650, @ilyam8)
  • Add "Creating Alerts with Netdata Alerts Configuration Manager" (#16642, @sashwathn)
  • Add instructions for installing Netdata in a rootless Docker environment (#16632, @ilyam8)
  • Add energy efficiency image to main readme file (#16617, @Aliki92)
  • Remove deprecated memory mode "map" and "save" (#16604, @vkalintiris)
  • Update Splunk icon to a dark version for improved visibility (#16593, @juacker)
  • Add documentation on how to configure Splunk Cloud notifications (#16586, @juacker)
  • Add a new document explaining Gorilla compression and decompression techniques (#16553, @vkalintiris)
  • Add an initial version of the "Plugin Functions Tables" documentation (#16535, @ktsaou)

Other Notable Changes

Improvements
Bug fixes
  • Fix a crash occurring when failing to create the requested number of tiers (#16999, @stelfrag)
  • Fix an issue where Netdata plugins could inherit unintended sockets or file descriptors during the forking process (#16881, @ktsaou)
Other

Deprecation notice

Changed in this release

All depreciated items from the v1.44.0 notice have been addressed except for enabling gorilla compression by default.

Additionally, the following Alert options have been deprecated in this release. While Netdata will still understand these options when
reading existing alert configurations for now, we recommend updating your custom alert configurations to use the
replacements listed below. Compatibility with these deprecated options might be removed in a future release.

Option Use instead
foreach DIMENSIONS (lookup line) -
charts -
os host labels: _os=X
host host labels: _hostname=X
plugin chart labels: _collect_plugin=X
module chart labels: _collect_module=X

Where X is a simple pattern.

Netdata Release Meetup

Join the Netdata team on the 25th of March at 17:00 UTC for the Netdata Release Meetup.

Together we’ll cover:

  • Release Highlights.
  • Acknowledgments.
  • Q&A with the community.

RSVP now - we look forward to meeting you.

Support options

As we grow, we stay committed to providing the best support ever seen from an open-source solution. Should you encounter an issue with any of the changes made in this release or any feature in the Netdata Agent, feel free to contact us through one of the following channels:

  • Netdata Learn: Find documentation, guides, and reference material for monitoring and troubleshooting your systems with Netdata.
  • GitHub Issues: Make use of the Netdata repository to report bugs or open a new feature request.
  • GitHub Discussions: Join the conversation around the Netdata development process and be a part of it.
  • Community Forums: Visit the Community Forums and contribute to the collaborative knowledge base.
  • Discord Server: Jump into the Netdata Discord and hang out with like-minded sysadmins, DevOps, SREs, and other troubleshooters. More than 2000 engineers are already using it!