Netdata Versions Save

Netdata v1.45.2 is a patch release to address issues discovered since v1.45.1.

This patch release provides the following bug fixes and updates:

• Improved PostgreSQL/MySQL local listener discovery to automatically check for connections using both TCP and Unix sockets, enabling support for passwordless Unix socket connections (#17304 #17305, @ilyam8)
• Fixed an issue that prevented negative matching of host/chart labels in alert configurations (#17290 #17292, @ktsaou)
• Improved go.d.plugin stability by preventing Netdata from shutting down the entire plugin due to an issue with registering jobs for unregistered modules (#17289, @ilyam8)
• Improved go.d.plugin HTTP requests now include a UserAgent string, enhancing identification in server log (#17286, @ilyam8)
• Improved Nginx discovery in go.d.plugin by automatically trying multiple status endpoints when discovering Nginx containers (#17285, @ilyam8)
• Fixed a go.d.plugin panic that could occur when using the Unbound collector with TLS (#17283, @ilyam8)
• Fixed a libyaml linking issue (#17276, @Ferroin)
• Improved go.d.plugin configuration validation, preventing unexpected or invalid options through dynamic configurations (#17269, @ilyam8)

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!

Netdata v1.45.1 is a patch release to address issues discovered since v1.45.0.

This patch release provides the following bug fixes and updates:

• Ensured proper handling of default values for data collection jobs submitted via dynamic configuration. (#17255, @ilyam8)
• Optimized go.d.plugin service discovery by filtering out irrelevant docker-proxy listeners. (#17254, @ilyam8)
• Improved go.d.plugin's ability to find applications, including those using IPv6, and identify Apache processes more reliably. (#17252, @ilyam8)
• Improved OpenSSL discovery on macOS for Homebrew builds. (#17250, @Ferroin)
• Obsolete references to saving the internal database using the USR1 signal, reflecting the removal of save/map memory modes. (#17249, @ilyam8)
• Added ZSTD compression support for dbengine (disabled by default for now). This improves storage efficiency when available, automatically falling back to uncompressed pages for compatibility. (#17244, @ktsaou)
• Fixed a bug that caused metric reference count errors during release. (#17239, @ktsaou)
• Code cleanup. (#17237, @ktsaou)
• Enabled Gorilla compression by default for dbengine, reducing memory usage. (#17234, @ktsaou)
• Improved dbengine unit tests for better code coverage and maintainability. (#17232, @ktsaou)
• Fixed a database engine cache bug that could cause queries to stop prematurely under pressure. (#17231, @ktsaou)
• Implemented caching optimization to reduce the number of cache flushes following journal file v2 creation. (#17220, @stelfrag)
• Reduced clutter in MySQL/MariaDB query logs by disabling session query logging for the go.d/mysql collector. (#17219, @ilyam8)
• Improved go.d.plugin to correctly identify MariaDB databases. (#17218, @ilyam8)
• Enhanced macOS build stability by using native libraries and optimizing checks for dependencies. (#17216, @Ferroin)
• Suppressed unnecessary compiler warnings about redefined macros, improving build cleanliness and compatibility with stricter build flags. (#17209, @Ferroin)

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!

Netdata v1.44.3 is a patch release to address issues discovered since v1.44.2.

This patch release provides the following bug fixes and updates:

• Improved handling of slow queries and CPU usage of the ACLKSYNC thread. (#16838, @stelfrag)
• Improved error handling for listen bind failures. Instead of terminating fatally, Netdata now exits gracefully. (#16937, @stelfrag)
• Fixed invalid alert durations in health log entries. (#16931, @stelfrag)
• Fixed a race condition during analytics data setup, preventing potential Netdata crashes. (#16929, @stelfrag)
• The Netdata base image includes Debian backports for comprehensive security and stability. (netdata/helper-images#271, @tkatsoulas)

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 1900 engineers are already using it!

Netdata v1.44.2 is a patch release to address issues discovered since v1.44.1.

This patch release provides the following bug fixes and updates:

• Fixed an inconsistency where the NETDATA_LOG_LEVEL environment variable did not affect log level in Docker containers. (#16943, @ilyam8)
• Fixed inconsistent log severity across sources: log severity level setting now work for all Netdata log sources (daemon, collector, health, access, aclk). (#16922, @ilyam8)
• Fixed a bug in chartd.d.plugin that prevented loading of its modules configuration files. (#16939, @ilyam8)
• Fixed inaccurate server type identification in Netdata Cloud for FreeBSD jails. Jails are now recognized correctly. (#16858, @ilyam8)
• Fixed a bug that prevented the edit-config script from running correctly in Podman containers. The script now accurately identifies container environments. (#16825, @Ferroin)
• Fixed a bug that caused excessive logging of "Using host prefix directory" messages. (#16814, @ilyam8)
• Fixed incorrect label source for apps.plugin charts, ensuring they are now accessible when querying Prometheus metrics. (#16810, @boxjan)
• Fixed a bug in the cgroups.plugin that could lead to crashes. Additionally, addressed incorrect thread name during fatal Agent exits. (#16771, @ktsaou)
• Fixed a race condition related to pthread_detach() calls, preventing potential Netdata crashes during thread creation. (#16760, @ktsaou)
• Fixed a bug that caused "maximum number of cgroups reached" messages to spam logs. (#16730, @ilyam8)
• Fixed incorrect service file location during MacOS installation: now, launchctl commands can reliably start and stop Netdata. (#16693, @ilyam8)
• Fixed a bug that caused the Netdata claiming process to fail on macOS due to an inaccessible netdata-claim.sh script. (#16686, @ilyam8)
• Fixed missing host label streaming from child nodes: host labels are now transmitted reliably to parent nodes. (#16821, @stelfrag)
• Fixes a bug in clock resolution calculation that prevented some data collection plugins from working correctly. (#16720, @ktsaou)
• Fixed a bug that caused Netdata to crash when calculating database size due to missing or single datafiles. (#16699, @ktsaou)
• Fixed a bug that caused the cups.plugin to not terminate upon receiving a SIGPIPE (Broken Pipe) signal. (#16691, @ilyam8)
• Fixed a reference counting issue that could lead to Netdata crashes. (#16687, @ktsaou)
• Fixed charts context and family definitions of exporting engine. (#16683, @ilyam8)
• Fixed a bug that could cause crashes when processing web requests. (#16664, @ktsaou)
• Fixed improper handling of the dbengine event loop during shutdown. (#16658, @stelfrag)
• Fixed a potential memory corruption issue in database code. (#16654, @stelfrag)
• Fixed "response too big" error for Systemd-journal: addressed limitations by raising the maximum web response size. (#16649, @ktsaou)
• Fixed compilation issues with --disable-dbengine: addressed errors that prevented successful builds when this flag was used. (#16611, @stelfrag)
• Fixed labels corruption due to duplicate key/value pairs. Additionally, addressed logging errors that occurred during fatal Agent exits. (commit, @ktsaou)
• Update go.d.plugin to v0.58.0. (#16725, @ilyam8).

Acknowledgements

We would like to thank our dedicated, talented contributors who make up this amazing community. The time and expertise that you volunteer are essential to our success. We thank you and look forward to continuing to grow together to build a remarkable product.

• @boxjan for fixing incorrect label source for apps.plugin charts.

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 1900 engineers are already using it!

Netdata v1.44.1 is a patch release to address issues discovered since v1.44.0.

This patch release provides the following bug fixes and updates:

• Fixed an issue in the uninstall script that prevented log2journal and systemd-cat-native from being removed (#16585, @ilyam8).
• Fixed a bug that caused the debugfs.plugin to not terminate upon receiving a SIGPIPE (Broken Pipe) signal (#16569, @ilyam8).
• Fixed memory leak during host chart label cleanup (#16568, @stelfrag).
• Fixed incorrect cpu architecture/ram/disk values in build info (#16567, @ilyam8).
• Fixed a bug that prevented the parent from accepting streaming connections on systems with one CPU core (#16565, @stelfrag).
• Make the systemd-journal mandatory package on Centos 7 and Amazon linux 2 (#16562, @tkatsoulas).
• Fixed crash on reading memory clock speed of an AMD graphics card (#16561, @MrZammler).
• Fixed an unhandled error that occurred when setting file capabilities in the Debian postinst script of the perf.plugin (#16558, @tkatsoulas).
• Fixed an issue where the user's netdata home directory was set to an incorrect value (#16548, @ilyam8).
• Added the lightweight text editor to the Docker image (#254, @tkatsoulas).

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 1700 engineers are already using it!

Netdata v1.43.2 is a patch release to address issues discovered since v1.43.1.

This patch release provides the following bug fixes and updates:

• Fix rrdlabels type (1676de2, @stelfrag).
• Fix label copy to allow new keys with different values (6179213, @stelfrag).
• Fix internal label source propagation when streaming metrics (60cd86d, @ktsaou).
• Speed up queries when sending alerts to Cloud on parents with a large number of alerts per child (f80f0fc, @MrZammler).
• Fix filtering when selecting multiple fields in systemd-journal plugin (750ca8e, @stelfrag).
• Fix an issue where parents were missing chart labels of child instances (240f9e7, @ktsaou).
• Fix an issue where updated labels were not propagated to parents (644d432, @stelfrag).

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 1700 engineers are already using it!

Netdata v1.43.1 is a patch release to address issues discovered since v1.43.0.

This patch release provides the following bug fixes and updates:

• Prevent wrong optimization armv7l static build (#16274, @stelfrag).
• Fixed pattern matching in Functions Search (#16264, @ktsaou).
• Fixed an issue where the query planner was using the wrong dbengine tier that had no data for the selected time period (#16263, @ktsaou).
• Fixed invalid payload in Discord notifications (#16257, @luchaos).
• Fixed possible deadlock on discovery thread shutdown in cgroups plugin (#16246, @stelfrag).
• Fixed duplicate chart labels (#16249, @stelfrag).
• Fixed dimension HETEROGENEOUS check (#16234, @stelfrag).
• Updated go.d plugin version to v0.56.3 (#16228, @ilyam8).
• Fixed calculation of dbengine statistics on 32bit systems (#16222, @stelfrag).
• Improved handling of duplicate labels (#16172, @stelfrag).
• Improved cleanup on shutdown of collectors (#16023, @ktsaou)

Acknowledgements

We would like to thank our dedicated, talented contributors that make up this amazing community. The time and expertise that you volunteer are essential to our success. We thank you and look forward to continuing to grow together to build a remarkable product.

• @luchaos for fixing Discord notifications.

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 1700 engineers are already using it!

Groundbreaking: systemd-journal logs release!

Table of Contents

• Netdata Growth
• Release Summary
• Release Highlights
  • systemd journal improvements
  • Virtual Machine monitoring (VMWare vSphere)
• What is coming next
• Acknowledgments
• Contributions
  • Collectors
  • Packaging/Installation
  • Documentation
  • Health
  • Other Notable Changes
• Deprecation notice
• Netdata Release Meetup
• Support options

Steady to our schedule, this is another great Netdata release!

Netdata Growth

• 65.5 k GitHub Stars ⭐ Since October 2023, Netdata is leading the observability category in the CNCF landscape, surpassing Elasticsearch. Thank you for your love ❤️! Give Netdata a ⭐ too, on GitHub!

• 595 M docker hub pulls Netdata runs with about 200k docker hub downloads per day. Since June 2023 we are a Verified Publisher, so that Netdata pulls don't count against docker hub pull limits for our users, allowing all our users to integrate Netdata to their CI/CD toolchains.

Release Summary

This release is the most robust and reliable Netdata we have ever built.

These are the main areas Netdata has improved since the last release:

1. Logs Today we release an almost rewritten version of systemd-journal, to improve its performance and visualization capabilities. systemd-journal holds critical systems and security information and given the lack of systemd-journal visualization tools, we focused first on filling this gap. At the same time, we are standardizing the way logs should be as a part of Netdata, enabling us to support more log management engines, like Loki and Elasticsearch.

2. Instances Slice and Dice Given the capabilities of the new Netdata Agent UI (v2), we are changing the way some of our collectors collect and expose metrics, to allow easier slicing and dicing of the data and be more OpenTelemetry compatible in terms of specifications. So, in this release we changed the way apps.plugin exposes charts in the Applications section of the dashboard. Following the NIDL framework, each application group is now an instance, allowing better aggregation of processes utilization across nodes. Similarly, our systemd units charts have been updated to have an instance for each systemd unit. For the same reasons, disk charts now have additional labels (id, model and serial) to help us identify disks from the charts. Unfortunately, such changes tend to make the older dashboards (v1, v0) less usable, especially on servers with many hundreds of instances.

3. Stock Alerts A number of changes have been implemented to the Netdata Health engine, to allow better integration with the new dashboard. More changes in this area are about to come, as part of the next release: a) allow multi-node alerts on parents, b) allow evaluating and configuring alerts from the UI.

4. Alerts Accuracy Netdata has by default 3 tiers of metrics, each with a different resolution. The Netdata query planner is automatically picking the right tier to satisfy a query, based on the number of points requested in the response. For alerts there was a side effect. Since alerts request only 1 point of data in the response, the query planner was picking the "easier" tier to query, which is of course the one with the lower resolution. Now alerts are always run on tier 0, the higher resolution one.

5. Lower Resources Utilization Several changes have been implemented for Netdata to better take care of itself. That includes lower memory usage, lower disk footprint, self vacuuming of SQLite databases, and more. Probably the most notable change is that now Netdata needs only 1 pointer (8 bytes on 64 bit, 4 bytes on 32 bit) for each use of a label name-value combination. This improves drastically Netdata's memory requirements in setups like busy k8s clusters, that containers come and go all the time, increasing the labels cardinality significantly.

6. 32bit Netdata on 64bit IoT machines A common request when Netdata is installed on 64bit IoT devices, is to have a 32bit Netdata running there. Before this release, this was not possible. Now a 32bit Netdata will nicely run on a 64bit operating system.

7. Netdata Cloud on prem Netdata Cloud is now available to be installed on-prem! Several companies have already deployed it and are currently testing it. If you want to join them, submit this form.

Release Highlights

systemd-journal

systemd-journal was first included in Netdata v1.42.0. Immediately after release, we recognized the wider need for this feature, so we've rewritten the plugin almost entirely, to provide the best possible experience. This work is also fundamental for supporting more log monitoring integrations - stay tuned!

The major improvements done on systemd-journal logs function were:

• addition of the histogram for log entries over time, with a break down per field-value, for any field and any time-frame
• enable of the PLAY mode provides the same experience as journalctl -f, showing new logs entries immediately after they are received
• allow filtering on any journal field or field value, for any time-frame
• add support for coloring log entries, the same way journalctl does

If you want to take a look at a full presentation of the systemd-journal plugin, how it works, how you can take full advantage of this and even instructions on configuration of a logs centralization server, check the documentation for the plugin.

You can experience the power of systemd-journal logs function in one of our Netdata demo rooms here or check our latest YouTube video on it.

Want to know why you should untap the full potential of systemd-journal logs? Check out Netdata's founder, Costa Tsaousis @ktsaou, blogpost on it here.

Virtual Machine monitoring (VMWare vSphere)

With the increased feedback and requests on VMware vCenter Server collectors we have:

• Reviewed our out-of-the-box charts
• Added labels to the charts, e.g. host, datacenter, cluster, vm
• Reviewed the metadata on alerts
• Added summary charts section

It is with this feedback from the Community that we can keep working on improving Netdata to ensure it meets your needs!

What is coming next

We are currently working on the following areas, which we hope to release next month:

1. Logs Explorer for Loki and Elasticsearch Similar to systemd-journal, allow Netdata to explore, query and visualize logs from Loki and Elasticsearch.

2. Collectors Configuration from the UI In the last release we presented the Integrations Marketplace. Since then, we work to make all integrations configurable via the dashboard. This will allow all of us to configure our Netdata servers directly from the UI, without touching configuration files, improving significantly the usability and easiness of Netdata.

3. Alerts Configuration from the UI Similarly, we work to allow configuring alerts directly from the UI, without text file configurations, so the all of us can create powerful alerts on the spot.

4. Netdata Mobile App We are at the final stage of releasing our Netdata Mobile App (iOS and Android) for receiving mobile push notifications and exploring alerts statuses.

5. Scalability Given the wide adoption of Netdata, we are committed to make Netdata scale better in larger environments. Especially when it comes to Netdata parents, we aim to provide the best scalability possible. We are currently finalizing the necessary changes to allow Netdata achieve:

   • 1 CPU core per 1 million metrics/s for data collection
   • 1 CPU core per 1 million metrics/s for ML and health (alerts)
   • 1 CPU core per 1 million metrics/s for re-streaming (pushing metrics to another parent)

   Of course, the numbers depend on the CPU and its clock, but they shouldn't vary significantly on modern systems.

   At the same time, we work to integrate Gorilla compression to our database. This will provide a significantly better overall memory footprint for Netdata.

Acknowledgments

We would like to thank our dedicated, talented contributors that make up this amazing community. The time and expertise that you volunteer are essential to our success. We thank you and look forward to continuing to grow together to build a remarkable product.

• @MAH69IK for improving ntfy notification title.
• @chpfm for fixing slave/user metrics collection stopping when query times out in go.d/mysql.
• @k0ste for various installation improvements on CentOS-Stream.
• @kylemanna for fixing an issue where a properly functioning sensors was skipped due to limits in python.d/sensors.
• @miversen33 for adding access control configuration to ntfy notification method.
• @novotnyJiri for fixing the wrong path in ansible-playbook deployment guide.
• @theggs for adding installation description for Homebrew on Apple Silicon.
• @vpnable for fixing counting UNDEF as users in go.d/openvpn_status_log.
• @zhqu1148980644 for fixing docker-compose example.
• @luisj1983 for implementing molecule tests in the netdata/ansible playbook

Contributions

Collectors

All changes

Improvements

• Improve exposing metrics by creating a chart for each app group/user/user group (apps.plugin) (#16095, @thiagoftsm)
• Add env NETDATA_LOG_SEVERITY_LEVEL support to external collectors (#16089, @ilyam8)
• Add env NETDATA_LOG_SEVERITY_LEVEL support (charts.d.plugin) (#16085, @ilyam8)
• Add env NETDATA_LOG_SEVERITY_LEVEL support (python.d.plugin) (#16084, @ilyam8)
• Improve performance by reading files sequentially (systemd-journal.plugin) (#16038, @ktsaou)
• Add systemd-journal plugin to apps_groups.conf (apps.plugin) (#16024, @ilyam8)
• Improve exposing metrics by creating a chart for each systemd service (cgroups.plugin) (#15975, @thiagoftsm)
• Add disk labels (proc/diskstats) (#15949, @ktsaou)
• Add support for opening journal files when running inside a container (systemd-journal.plugin) (#15830, @ktsaou)
• Add env NETDATA_LOG_SEVERITY_LEVEL support (go.d.plugin) (#1351, @ilyam8)
• Add "network" config option that allows configuration of DNS resolution (go.d/ping) (#1348, @ilyam8)
• Add "custom_numeric_fields" config option (go.d/web_log) (#1343, @ilyam8)
• Add upsd (NUT) collector (go.d/upsd) (#1341, @ilyam8)
• Improve status chart by making it a dimension per status (go.d/vcsa) (#1332, @ilyam8)
• Add label to vm/host charts (go.d/vsphere) (#1331, @ilyam8)

Bug fixes

• Fix 1-second latency in play mode (systemd-journal.plugin) (#16123, @ktsaou)
• Fix an issue where ipv4 metrics were exposed as ip (proc/netstat) (#16122, @ilyam8)
• Fix an issue where OOMKill was created unconditionally (ebpf.plugin) (#16115, @thiagoftsm)
• Fix an issue where ebpf threads did not respect the enable/disable value in the configuration (ebpf.plugin) (#16083, @thiagoftsm)
• Fix using undefined var when loading job statuses (python.d.plugin) (#15965, @ilyam8)
• Fix an issue where a properly functioning sensor was skipped due to limits (python.d/sensors) (#15905, @kylemanna)
• Fix slave/user metrics collection stopping when query times out (go.d/mysql) (#1346, @chpfm)
• Fix counting UNDEF as users (go.d/openvpn_status_log) (#1334, @vpnable)
• Fix an issue where power metric were not collected due to renaming (go.d/nvidia_smi) (#1310, @ilyam8)

Other

• Remove mem_private chart on FreeBSD (apps.plugin) (#16166, @ilyam8)
• Improve eBPF exit (ebpf.plugin) (#16159, @thiagoftsm)
• Update eBPF to use event loop for functions (ebpf.plugin) (#16004, @thiagoftsm)
• Add socket function (ebpf.plugin) (#15850, @thiagoftsm)
• Update restart time to 1 day (nfacct.plugin) (#15801, @ilyam8)

Packaging / Installation

All changes

• Fix removing wrong directories when uninstalling on FreeBSD (#16167, @tkatsoulas)
• Fix repo path for openSUSE 15.5 packages (#16161, @tkatsoulas)
• Fix an issue running a Docker container when the default user was configured as a non-root user (#16156, @ilyam8)
• Fix an issue where the uninstaller script doesn't clean up properly (#16148, @ilyam8)
• Fix problem with the uninstaller script when executed as a regular user (#16146, @ilyam8)
• Skip trying to preserve file owners when bundling external code (#15966, @Ferroin)
• Cleanup Dockerfile (#15902, @Ferroin)
• Skip copying environment/install-type files when checking existing installations (#15876, @Ferroin)
• Add setuid fallback for perf and slabinfo plugins in the installer script (#15807, @ilyam8)
• Fix an issue where cleanup was not performed during the kickstart.sh dry run (#15775, @ilyam8)
• Add CentOS-Stream to distros (#15742, @k0ste)
• Fix build with --disable-https (#15395, @MrZammler)
• Enable building go.d plugin natively for CentOS-Stream (#14551, @k0ste)

Documentation

All changes

• Cleanup systemd-journal readme (#16096, @Ancairon)
• Change Twitter username to @netdatahq (#16082, @ralphm)
• Add "maintained by netdata/community" badge to collectors (#16073, @Ancairon)
• Improve tc plugin description (#16068, @thiagoftsm)
• Add doc about running a local dashboard through Cloudflare (#16043, @Ancairon)
• Merge docs about Netdata Charts (#16042, @Ancairon)
• Add installation description for Homebrew on Apple Silicon (#16027, @theggs)
• Add a script to create COLLECTORS.md from the integrations.js file and generate the document (#15995, @Ancairon)
• Clarify shipping repositories cases (#15960, @tkatsoulas)
• Clarify possible installation types (#15958, @tkatsoulas)
• Add specific info on how to access the dashboards (#15925, @hugovalente-pm)
• Cleanup "Change how long Netdata stores metrics" (#15896, @Ancairon)
• Update pfSense doc header (#15894, @Ancairon)
• Properly document issues with installing on IPv6-only hosts. (#15882, @Ferroin)
• Add new delete old models param to ML readme (#15873, @andrewm4894)
• Rename alarm to alert (#15812, @ilyam8)
• Fix typo in Readme (#15794, @shyamvalsan)
• Fix the wrong path in ansible-playbook deployment guide (#15786, @novotnyJiri)
• Fix docker-compose example (#15784, @zhqu1148980644)
• Mark integrations milestones as completed in README.md (#15783, @tkatsoulas)
• Add dev documentation for Dynamic Configuration (#15643, @underhood)

Health

All changes

• Add "summary" field to alert configurations (#16129, @MrZammler)
• Remove discontinued Hangouts and StackPulse notification methods (#16041, @Ancairon)
• Add go.d/upsd alerts (#16036, @ilyam8)
• Improve the accuracy of health queries (#16032, @MrZammler)
• Remove "family" field from alerts (#16025, @MrZammler)
• Add access control configuration to ntfy notification method (#15932, @miversen33)
• Improve ntfy notification title (#15909, @MAH69IK)
• Add "summary" field to alerts (#15886, @MrZammler)
• Enable "ml_1min_node_ar" alert by default (#14687, @andrewm4894)

Other Notable Changes

All changes

Improvements

• Improve ML database size management (#16046, @stelfrag)
• Add the ability for facets to generate histograms (#15846, @ktsaou)
• Add /api/v2/ilove.svg endpoint (#15815, @ktsaou)
• Reduce label memory usage (#15255, @stelfrag)
• Add severity level for logs (#14727, @thiagoftsm)
• Introduce molecule tests for the netdata/ansible (netdata/ansible#6, @luisj1983

Bug Fixes

• Fix MQTT crash when running 32-bit static build on 64-bit system on ARM (#16154, @underhood)
• Fix random crashes on pthread_detach() (#16137, @ktsaou)
• Fix crash on parsing clabel command with no source (#16114, @ilyam8)
• Fix crash on startup on busy parents (#16016, @ilyam8)
• Fix duplicate keys in labels (#16014, @stelfrag)
• Fix obsolete charts cleanup (#15892, @MrZammler)

Other

• Fix corrupting the index when doubling the hashtable in facets (#16171, @ktsaou)
• Fix compilation warnings (#16158, @stelfrag)
• Don't queue removed when there is a newer alert (#16157, @MrZammler)
• Batch ML model load commands (#16155, @stelfrag)
• Make io charts "write" negative in apps and cgroups (#16152, @ilyam8)
• Varius improvements in system-journal plugin facets (#16150, @ktsaou)
• Fix logging an unknown key error for "families" in health (#16145, @ilyam8)
• Fix journal help and mark debug keys in the output (#16133, @ktsaou)
• Fix an issue where newlines were removed when forwarding FUNCTION_PAYLOAD (#16120, @underhood)
• Fix an issue in systemd-journal plugin where anchor were not respected on non-data-only queries (#16109, @ktsaou)
• Improve systemd-journal plugin histogram and facets calculation (#16107, @ktsaou)
• Various code improvements (#16104, @stelfrag)
• Improve systemd-journal plugin logging (#16101, @ktsaou)
• Improve systemd-journal plugin performance (#16099, @ktsaou)
• Fix incremental queries in systemd-journal plugin (#16098, @ktsaou)
• Fix querying out of retention (#16094, @ktsaou)
• Fix an issue where buggy sd_journal_open_files_fd() were used on systems with old libsystemd (#16090, @ktsaou)
• Fix a busy wait loop in functions (#16086, @ktsaou)
• Skip database migration steps in new installation (#16071, @stelfrag)
• Fix coverity 402975 (#16058, @stelfrag)
• Send alerts summary field to cloud (#16056, @MrZammler)
• Fix summary field in table (#16050, @MrZammler)
• Fix overflow in storage engine stats on 32bit systems (#16048, @stelfrag)
• Fix wrong units in the anomaly_detection.detector_events chart (#16028, @andrewm4894)
• Fix crash in systemd-journal plugin when the uid or gid have no names (#16015, @ktsaou)
• Remove the line length limit from pluginsd (#16013, @ktsaou)
• Fix compilation warnings (#16006, @stelfrag)
• Add missing files to CMakeLists.txt (#16005, @stelfrag)
• Fix compilation warnings (#16001, @ktsaou)
• Add collectors restart support to functions (#15983, @ktsaou)
• Maintain node's last connected timestamp in the db (#15979, @stelfrag)
• Fix race in apps plugin Processes function (#15978, @ktsaou)
• Implement functions cancelling (#15977, @ktsaou)
• Various improvements to facets (#15976, @ktsaou)
• Remove sending db retention in facets (#15974, @ktsaou)
• Extend ml default training from ~24 to ~48 hours (#15971, @andrewm4894)
• Fix wrong facets when histogram is empty (#15970, @ktsaou)
• Fix shadowing local variable in facets (#15968, @ktsaou)
• Implement data-only queries in facets (#15961, @ktsaou)
• Fix direction parsing in systemd-journal plugin (#15957, @ktsaou)
• Various improvements in facets (#15956, @ktsaou)
• Fix CID 400366 (#15953, @stelfrag)
• Improve streaming logs (#15948, @ktsaou)
• Improve facets performance (#15940, @ktsaou)
• Improve facets info (#15936, @ktsaou)
• Add info and source facets to systemd-journal plugin (#15928, @ktsaou)
• Various improvements in systemd-journal and facets (#15926, @ktsaou)
• Reduce workload during metadata cleanup (#15919, @stelfrag)
• Improve shutdown of the metadata thread (#15901, @stelfrag)
• Make anomaly_detection.type_anomaly_rate stacked (#15895, @andrewm4894)
• Add better recovery for corrupted metadata (#15891, @stelfrag)
• Add index to ACLK table to improve update statements (#15890, @stelfrag)
• Limit atomic operations for statistics (#15887, @ktsaou)
• Allow any field to be a facet (#15880, @ktsaou)
• Improve hashing performance in facets by switching to the newer XXH3 128bits algorithm (#15878, @ktsaou)
• Update SQLITE version to 3.42.0 (#15870, @stelfrag)
• Add initialization fail reason to analytics (#15866, @stelfrag)
• Add a chart that groups anomaly rate by chart type (#15856, @vkalintiris)
• Fix "unrecognized options: --with-zlib" configure warning (#15840, @stelfrag)
• Fix compilation warning (#15839, @stelfrag)
• Fix warning when compiling with -flto (#15838, @stelfrag)
• Fix opening journal files on systems with old systemd (#15837, @ktsaou)
• Add ilove.html (#15836, @ktsaou)
• Fix CID 382964: Code maintainability issues (SIZEOF_MISMATCH) (#15833, @stelfrag)
• Fix coverity 393052: API usage errors (LOCK) (#15832, @stelfrag)
• Fix an issue where fd wasn't released if setsockopt or bind fails (#15826, @stelfrag)
• Fix memory leak when updating job status if job does not exist in Dyncfg. (#15822, @stelfrag)
• Improve ML initialization (#15819, @stelfrag)
• Update cmakelist (#15817, @stelfrag)
• Add streaming support to Dyncfg (#15791, @underhood)
• Various fixes in Dyncfg (#15785, @underhood)
• Improve ML models cleanup (#15720, @vkalintiris)
• Cleanup and refactoring of health and ACLK code (#15665, @stelfrag)
• Improve metadata cleanup (#15462, @stelfrag)

Deprecation notice

Changed in this release

In accordance with our previous deprecation notice, the following items in this release have been changed:

Component	Type	Change	Action
apps.plugin	collector	a dimension for each group/user/user group => a chart for each group/user/user group
cgroups.plugin	collector	a dimension for each systemd service => a chart for each systemd service
proc.plugin	collector	all "Networking Stack" metrics except "tcp" have been moved to "IPv4 Networking"
family attribute	alert configuration and Health API	deprecated	use chart labels

Will be changed in the next release

We plan to change in the next release (v1.44.0):

Component	Type	Change	Action
charts.d/nut	collector	deprecated	use go.d/upsd

Netdata Release Meetup

Join the Netdata team on the 18th of October at 16:30 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 1700 engineers are already using it!