Release v1.26.0

The v1.26.0 release of the Netdata Agent brings exciting new collectors written in Go, a new integration with the DevOps startup StackPulse, and massive improvements to the way users navigate Netdata's documentation. We've also added compatibility with an exciting new feature that's coming soon to Netdata Cloud—stay tuned!

This release contains 3 new collectors, 1 new notifcation method, 21 improvements, 13 documentation updates, and 12 bug fixes.

At a glance

The Netdata Agent can now collect metrics from files/directories, systemd units, and ISC DHCP servers. These new collectors are part of our larger effort to migrate all collectors to Go, which provides more extensibility compared to previous implementations. You can read about each of these new collectors in our docs: filecheck, systemd, isc_dhcpd.

We're excited to extend our health watchdog even further to integrate with StackPulse, which is designed to help SREs manage and respond to incidents with code and automation. You can read more about how to configure Netdata to send notifications to StackPulse in the docs.

We rearchitected our docs/education site, Netdata Learn, to focus on users actions rather than the Netdata Agent's hierarchy of code. The core docs now better guide users through the most important actions, such as configuring collectors and interacting with charts, independent of whether they use only the Netdata Agent, or the Agent in combination with Netdata Cloud. Of course, all of our reference documentation is still alive and kicking for those who want to dive into every configuration option or API query. We also revamped our guides page with better visuals, a search/filter, and more rational categories.

Acknowledgments

We're grateful to the Netdata community for their contributions for this release.

• @HolgerHees for fixing the comment syntax in Netdata's systemd file. (#10066,
• @Saruspete for fixing a file descriptor leak in the Infiniband colector (proc.plugin).
• @hamedbrd for adding a new Go-based systemd unit state collector and fixing gauges for the go.d.plugin/web_log collector.
• @chadknutson for adding chart for churn rates to python.d/rabbitmq.
• @hydrogen-mvm for adding a missing period in the Netdata dashboard.
• @roedie for adding a missing libelf-dev dependency.
• @Dim-P) and @disko for documentation improvements.

Improvements

• Add the ability to send Agent alarm notifications to StackPulse. (#9965, @thiagoftsm)
• Add a way to get build configuration info from the Agent. (#9913, @Ferroin)
• Add chart for churn rates to python.d/rabbitmq. (#10031, @chadknutson)
• Add failed dim to the connection_fails alarm in the Portcheck alarm. (#10048, @ilyam8)
• Improve the data query when using the context parameter (#9978, @stelfrag).
• Add a context parameter to the data endpoint. (#9931, @stelfrag)

Netdata Cloud

• Change default ACLK query thread count. (#10009, @underhood)
• Remove leading whitespace before JSON in ACLK. (#9998, @underhood)
• Allow using libwebsockets without SOCKS5. (#9973, @underhood)
• Add information about Cloud disabled status to -W buildinfo. (#9936, @underhood)

Collectors

• Update go.d.plugin version to v0.23.0. (#10046, @ilyam8)
  • Add new filecheck collector. (go.d.plugin/#445, @ilyam8)
  • Add new systemd unit state collector. (go.d.plugin/#439, @hamedbrd)
  • Add new ISC DHCP collector. (go.d.plugin/#451, @thiagoftsm)

Dashboard

• Add missing period in Netdata dashboard. (#9960, @hydrogen-mvm)
• Add missing tests to the web server. (#10008, @thiagoftsm)

Packaging/installation

• Rename NETDATA_PORT to NETDATA_LISTENER_PORT. (#10045, @knatsakis)
• Add a few changes that were missed by the systemd updater support. (#10007, @Ferroin)
• Switch to our installer's bundling code for libJudy in static installs. (#9988, @Ferroin)
• Add improved auto-update support. (#9966, @Ferroin)
• Add missing libelf-dev dependency. (#9974, @roedie)
• Update RPM spec file to use automatic dependency list generation. (#9937, @Ferroin)
• Add support for using /etc/cron.d for auto-updates. (#9598, @Ferroin)
• Add more stringent check for C99 support in configure script. (#9982, @Ferroin)

Documentation

• Add note about using nolock when debugging. (#10036, @andrewm4894)
• Update claiming document to instruct users to install uuidgen. (#9925, @OdysLam)
• Fix link in exporting document. (#10020, @Dim-P)
• Clean up and better cross-link new docsv2 documents. (#10015, @joelhans)
• Update FreeBSD documentation with updated packages. (#10005, @disko)
• Add documentation for claiming k8s parent pods and Prometheus service discovery. (#10001, @joelhans)
• Add docsv2 project to master branch. (#10000, @joelhans)
• Fix setting for disabling eBPF-apps.plugin integration. (#9967, @joelhans)
• Fix Stackpulse doc. (#9968, @thiagoftsm)
• Add persistent configuration details to Docker docs. (#9926, @joelhans)
• Add guide for monitoring Pi-hole and Raspberry Pi. (#9770, @joelhans)
• Add notice to Docker docs about systemd volumes. (#9927, @thiagoftsm)
• Add mirrored_hosts_status into Swagger docs. (#9867, @underhood)

Bug fixes

• Fix systemd comment syntax. (#10066, @HolgerHees)
• Fix file descriptor leak in Infiniband collector (proc.plugin). (#10013, @Saruspete)
• Fix the data endpoint to prioritize chart over context if both are present. (#10032, @stelfrag)
• Fix cleanup of obsolete charts. (#9985, @mfundul)
• Fix typos in installer functions. (#9992, @Ferroin)
• Fix typo inside netdata-installer.sh (#9962, @thiagoftsm)
• Fix build for the AWS Kinesis exporting connector. (#9823, @vlvkobal)
• Fix incorrect condition in updater type detection. (#10028, @Ferroin)
• Fix gauges for go.d.plugin/web_log collector. (#10029, @hamedbrd)
• Fix locking order to address CID_362348. (#9991, @stelfrag)
• Fix chart's last accessed time during context queries. (#9952, @stelfrag)
• Fix resource leak in case of malformed request to Netdata Cloud. (#9934, @underhood)