What netdata/netdata shipped
Written by FoxPlug from the repository's own releases; not affiliated with Netdata. Paste your own repository in the box below and FoxPlug writes the same weekly update for your product. This page is an automatic summary of the releases, pull requests and commits in github.com/netdata/netdata; Netdata did not write it and does not use or endorse FoxPlug. Every line links to the change it describes.
Get a weekly update like this for your product, free
Week of October 5, 2026
What shipped
- Netdata v2.12.0 was released, with the Windows Agent build removed from the release page because of startup crashes traced to an upstream bug in a dependency. Release
- Netdata v2.12.1 shipped as a patch release that fixes a Windows Agent startup crash and a timezone detection bug, stabilizes the eBPF collector across more kernels and libbpf versions, and reduces Docker collector load. Release
- The experimental Go IPMI plugin now supports all Go Linux architectures instead of only amd64 and arm64, and stays Linux-only and disabled by default. Pull request #24215
- The experimental IPMI plugin can now collect from a remote BMC over LAN and LAN+, so one job can monitor a local OpenIPMI device or a remote BMC. Pull request #24216
- The Jetson collector now collects current power per rail, adding power rail charts alongside the existing GPU and EMC charts. Pull request #24204
- SNMP traps from SNMPv3 users configured with authentication now require authentication, so a noAuthNoPriv message naming such a user is no longer accepted. Pull request #24188
- SNMP connection info no longer includes the v1/v2c community or SNMPv3 passphrases in clear text, as the string is built from security level and USM user instead. Pull request #24217
- Host labels are loaded from the config file the Agent actually loaded, including one passed with -c FILE, and labels deleted from that file are removed on reload-labels. Pull request #24117
- The DEM GeoIP sources now support Agent-aware cache and stock discovery, explicit overrides and automatic refresh, and expose source availability through rum-sites. Pull request #24192
- Ndsudo no longer waits forever on root commands that cannot be signaled, so a hung command with no output returns after its timeout instead of blocking the call. Pull request #24189
Why it matters
This week brought two Agent releases, including a patch that addresses the Windows startup crash and timezone bug from the previous build, plus wider eBPF stability and lower Docker collector load. Alongside that, the experimental Go IPMI plugin gained architecture support and remote BMC collection, and SNMP handling stopped leaking secrets into debug logs. A number of collectors and background paths were tuned, from Jetson power rails to proc file lookups, host label reloading and DEM GeoIP handling.
Changelog entry
- Published Netdata v2.12.0, removing the Windows Agent installer from the release page due to startup crashes from an upstream bug. Release
- Released Netdata v2.12.1 with fixes for a Windows Agent startup crash and timezone detection bug, wider eBPF kernel stability, and lower Docker collector load. Release
- Extended the experimental Go IPMI plugin to all Go Linux architectures by dropping the amd64/arm64 restriction. Pull request #24215
- Enabled LAN and LAN+ remote BMC collection in the experimental IPMI plugin. Pull request #24216
- Added per-rail power consumption charts to the Jetson collector. Pull request #24204
- Required authentication for SNMPv3 traps from users configured with an authentication protocol. Pull request #24188
- Removed SNMP community strings and SNMPv3 passphrases from the logged connection info string. Pull request #24217
- Reloaded host labels from the Agent's active config file, including one passed with -c FILE, and removed labels deleted from that file. Pull request #24117
- Reworked DEM GeoIP loading to use Agent-aware discovery, explicit overrides and automatic refresh, and exposed source availability via rum-sites. Pull request #24192
- Kept RunNDSudo's timeout for root commands that cannot be signaled. Pull request #24189
- Made RUM installation explicit and receiver-hosted, using the configured
public_urland listener address. Pull request #24187 - Avoided unnecessary POLLOUT polling for streaming senders and receiver. Pull request #24203
Two releases this week: v2.12.1 fixes the Windows Agent startup crash and timezone bug, stabilizes eBPF across more kernels, and cuts Docker collector load. Also: Go IPMI now covers all Linux architectures and remote LAN BMCs, and SNMP no longer logs secrets.
This week Netdata shipped v2.12.1, a patch release that fixes a Windows Agent startup crash and a timezone detection bug, stabilizes the eBPF collector across a wider range of kernels and libbpf versions, and reduces the load the Docker collector places on the system. v2.12.0 was also published, with its Windows installer removed from the release page because of startup crashes traced to an upstream bug. On the collectors side, the experimental Go IPMI plugin now supports all Go Linux architectures and can collect from remote BMCs over LAN and LAN+, the Jetson collector added per-rail power, and SNMP stops putting communities and v3 passphrases in debug logs. Host labels now reload from the config file the Agent loaded, and DEM gained Agent-aware GeoIP refresh and explicit RUM installation
Week of September 28, 2026
What shipped
- Released v2.12.0 with OpenTelemetry Traces in Early Access, Redfish out-of-band server hardware monitoring, SNMP Topology diagnostics, new S3 and SMBIOS collectors, and reworked Ceph monitoring. Release
- Dem.plugin now runs Playwright journeys and Lighthouse desktop audits natively alongside browser RUM pageviews, with per-attempt history and investigation features. Pull request #24137
- Introduced dem.plugin using the Go framework with managed receiver and site jobs, RUM charts, Functions, SQLite history, OTLP export, and alerts. Pull request #24129
- Traces queries now read evicted files back from remote storage, and span searches report span count per service. Pull request #24088
- The redis collector now supports Redis-compatible servers including Valkey, Dragonfly, KeyDB, Kvrocks, and Garnet. Pull request #24130
- Prometheus collector adds
counter_raw_chartsoption to display a counter's raw value alongside its per-second rate. Pull request #24103 - Nd-run now offers opt-in Linux process-tree supervision with Go support for cancellation and verified cleanup of descendants. Pull request #24135
- Windows.plugin gains missing.NET CLR metrics and switches CLR collection from object-level totals to per-instance metrics. Pull request #22219
- Native scripts now declare metric and check metadata inside each collection snapshot, eliminating the need for separate metrics and checks sections in manifests. Pull request #24096
- Windows monitoring now includes support for Cluster Shared Volumes (CSV) in failover clustering environments. Pull request #23720
Why it matters
Version 2.12.0 brings significant monitoring expansions: native browser journey and audit collection through DEM, OpenTelemetry traces in early access, and out-of-band hardware monitoring with Redfish. Support for Redis-compatible servers and Windows clustering workloads extends platform coverage. The refactored native scripting system simplifies integration authoring by eliminating separate metadata files.
Changelog entry
- Released v2.12.0 with OpenTelemetry Traces Early Access, Redfish, SNMP Topology, S3, SMBIOS, Ceph rework, Windows Hyper-V/RDP/Cluster Shared Volumes Release
- Dem.plugin: add native Playwright journeys and Lighthouse desktop audits with per-attempt history and investigation features Pull request #24137
- Dem.plugin: introduce Go framework with managed jobs, RUM charts, Functions, SQLite history, OTLP export, alerts (experimental, disabled by default) Pull request #24129
- Go.d/redis: support Redis-compatible servers (Valkey, Dragonfly, KeyDB, Kvrocks, Garnet) Pull request #24130
- Go.d/prometheus: add
counter_raw_chartsto display counter raw values alongside per-second rates Pull request #24103 - Otel-traces: read evicted files from remote storage; search rows report span count per service Pull request #24088
- Nd-run: add opt-in Linux process-tree supervision with Go cancellation support Pull request #24135
- Windows.plugin: add missing.NET CLR metrics; switch from object-level totals to per-instance collection Pull request #22219
- Scripts.d: native scripts declare metadata inside collection snapshots, eliminating separate manifest sections Pull request #24096
- Windows.plugin: add Cluster Shared Volumes (CSV) monitoring Pull request #23720
- Plugins.d: restart external plugins after protocol errors instead of disabling permanently Pull request #24051
- Fix global SQLite stalls by replacing global lock with lifetime leases per connection Pull request #24099
Netdata v2.12.0: OpenTelemetry Traces, Playwright journeys and Lighthouse audits in dem.plugin, Redfish hardware monitoring, Redis-compatible servers, and Windows Cluster Shared Volumes support.
Netdata v2.12.0 ships with OpenTelemetry Traces in Early Access, enabling observability across tracing, metrics, and logs. dem.plugin now runs Playwright journeys and Lighthouse audits natively. Redfish brings out-of-band hardware monitoring. New support for Redis-compatible servers (Valkey, Dragonfly, KeyDB, Kvrocks, Garnet). Windows gains Hyper-V, RDP, and Cluster Shared Volumes monitoring. The native scripting system now eliminates separate metadata files.
Week of September 21, 2026
What shipped
- The Go StatsD collector now runs as a standalone statsd.d.plugin with its own entry point and configuration instead of as part of go.d. Pull request #24016
- Source archives are now available compressed with zstd, producing roughly 30% smaller files than gzip while decompressing faster. Pull request #22440
- StatsD listener configuration now supports a 'both' protocol option to bind UDP and TCP on the same address, and adds duration idle timeout configuration. Pull request #24028
- Database engine flush deadlock in the libuv worker pool is fixed by bounding flush workers and deferring excess work. Pull request #24027
- Host label changes now forward through streaming parents without reconnecting while keeping Cloud node info and alert matching current. Pull request #24015
- Idle NetIPC CPU use on 32-bit time64 Linux is fixed by updating the vendored NetIPC library to handle the futex syscall correctly. Pull request #24049
- Function responses from freeipmi.plugin are now serialized with chart output to prevent chart lines from being read as part of the response. Pull request #24052
- Function responses from apps.plugin are now serialized with chart output using proper mutex protection. Pull request #24050
- SMBIOS memory collector now identifies slots by both bank and device locator to handle boards that reuse device locators across channels. Pull request #24043
- Go.d collectors can now classify Init/Check failures as permanent or temporary errors with shared lifecycle error classification. Pull request #24004
Why it matters
This week focused on fixing synchronization issues between chart output and function responses that could corrupt protocol streams, improving StatsD as a standalone plugin, and addressing database and SNMP reliability across streaming and time handling. Several fixes target flaky tests and race conditions in the configuration and secret management systems.
Changelog entry
- StatsD collector now runs as standalone statsd.d.plugin Pull request #24016
- Zstd-compressed source archives available alongside gzip Pull request #22440
- StatsD: 'both' protocol option for UDP+TCP and duration idle timeout Pull request #24028
- Fix database engine flush deadlock in libuv worker pool Pull request #24027
- Host label forwarding through streaming parents without reconnect Pull request #24015
- Fix idle NetIPC CPU use on 32-bit time64 Linux Pull request #24049
- Serialize freeipmi.plugin function responses with chart output Pull request #24052
- Serialize apps.plugin function responses with chart output Pull request #24050
- SMBIOS memory: identify slots by bank and device locator Pull request #24043
- Go.d collectors: PermanentError and TemporaryError lifecycle classification Pull request #24004
- SNMP: handle sysObjectID as OctetString instead of ObjectIdentifier Pull request #24031
- NetFlow plugin: publish nothing until UDP listeners are bound Pull request #24029
- Fix SQLite status checks and ACLK protobuf serialization failures Pull request #24012
- Fix 64-bit timestamp formatting on 32-bit targets Pull request #23991
- Correct const qualifiers and pointer casts in C code Pull request #23998
New: StatsD runs as statsd.d.plugin. Fixes: function response serialization, database flush deadlock, idle CPU on 32-bit time64, SMBIOS slot identification, host label streaming. zstd archives now available.
This week's updates strengthen Netdata's reliability across multiple fronts. StatsD is now a standalone plugin with better configuration options. We fixed critical synchronization bugs where function responses could be corrupted by chart output, database flush deadlocks in the libuv pool, and idle CPU consumption on 32-bit time64 systems. SNMP and SMBIOS collectors are more robust, host labels stream properly through parents, and zstd-compressed archives reduce download sizes by 30%. Over 40 pull requests landed, mostly addressing race conditions, test flakiness, and protocol correctness.
Week of September 14, 2026
What shipped
- New
smbios_memorycollector for Linux reads SMBIOS firmware tables and reports physical memory inventory with persistent loss detection. Pull request #23965 - V2 collectors can now add, remove, and replace named native chart templates at runtime while preserving unchanged entries' lifecycle state. Pull request #23971
- Redfish collector chart template rebuilt from 10,223 lines of generated defaults to 4,000 lines of curated configuration. Pull request #23957
- Legacy shell-format alarm notification configuration now maps to 23 Go providers with ordered overlays and recipient policies. Pull request #23930
- Redfish collector enriched with optional hardware columns for processor counts, memory, storage details, firmware dates, and assembly revisions. Pull request #23950
- Redfish collector adds on-demand BMC logs viewer with event filtering by time, severity, and text. Pull request #23943
- V2 collector per-job collection status and duration charts restored, shared with V1 rendering. Pull request #23946
- Journal access diagnostics expanded to capture all three permission-granting mechanisms with optional plugin debug mode. Pull request #23947
- Discovered targets can now reference secrets when explicitly trusted by their collection pipeline. Pull request #23918
- Documentation added for secret-resolution execution privileges and SNMP scan credential exposure on Linux and Unix systems. Pull request #23923
Why it matters
This week adds hardware inventory collection via SMBIOS, significantly improves Redfish functionality with logs viewing and enriched hardware details, and maps legacy alarm notification systems to modern providers. Support and debugging capabilities also improve for journal access and secret handling.
Changelog entry
- Add
smbios_memorycollector for Linux memory inventory and persistent loss detection Pull request #23965 - Support named active chart template sets in V2 collectors with runtime add/remove/replace Pull request #23971
- Refactor Redfish chart template from 10,223 to 4,000 lines, removing generated defaults Pull request #23957
- Add optional Redfish hardware columns for processor, memory, storage, and firmware details Pull request #23950
- Add Redfish on-demand BMC logs viewer with time, severity, and text filtering Pull request #23943
- Map legacy alarm notification settings to 23 Go providers with config overlays Pull request #23930
- Restore V2 collector per-job status and duration charts Pull request #23946
- Expand journal access diagnostics to capture all three permission mechanisms Pull request #23947
- Allow per-pipeline trust of discovered targets for secret references Pull request #23918
- Document secret-resolution execution privileges and SNMP scan exposure Pull request #23923
New smbios_memory collector reports Linux physical memory inventory. Redfish gains BMC logs viewer and hardware enrichment. Legacy alarm notifications now map to Go providers. V2 charts restored.
This week's updates strengthen hardware monitoring and legacy system integration. A new smbios_memory collector provides Linux memory inventory tracking. The Redfish collector gains on-demand BMC logs viewing and expanded hardware details. Most significantly, legacy shell-format alarm notification configurations now map to 23 modern Go providers, enabling smoother migration paths for existing deployments.