1 netdata (1.6.0) - 2017-03-20
3 * birthday release: 1 year netdata
5 netdata was first published on March 30th, 2016.
6 It has been a crazy year since then:
8 225.000 unique netdata users
9 currently, at 1.000 new unique users per day
11 80.000 unique netdata installations
12 currently, at 500 new installation per day
14 610.000 docker pulls on docker hub
16 4.000.000 netdata sessions served
17 currently, at 15.000 sessions served per day
24 * central netdata is here
26 This is the first release that supports real-time streaming of
27 metrics between netdata servers.
31 - autonomous host monitoring
32 (like it always has been)
34 - headless data collector
35 (collect and stream metrics in real-time to another netdata)
38 (collect metrics from multiple netdata and stream them to another netdata)
40 - store and forward proxy
41 (like headless proxy, but with a local database)
44 (metrics from multiple hosts are aggregated)
46 metrics databases can be configured on all nodes and each node maintaining
47 a database may have a different retention policy and possibly run
48 (even different) alarms on them.
50 * monitoring ephemeral nodes
52 netdata now supports monitoring autoscaled ephemeral nodes,
53 that are started and stopped on demand (their IP is not known).
55 When the ephemeral nodes start streaming metrics to the central
56 netdata, the central netdata will show register them at "my-netdata"
57 menu on the dashboard.
59 For more information check:
60 https://github.com/firehol/netdata/wiki/monitoring-ephemeral-nodes
62 * monitoring ephemeral containers and VM guests
64 netdata now cleans up container, guest VM, network interfaces and mounted
65 disk metrics, disabling automatically their alarms too.
67 For more information check:
68 https://github.com/firehol/netdata/wiki/monitoring-ephemeral-containers
70 * apps.plugin ported for FreeBSD
72 @vlvkobal has ported "apps.plugin" to FreeBSD. netdata can now provide
73 "Applications", "Users" and "User Groups" on FreeBSD.
77 @l2isbad has done a wonderful job creating a unified web log parsing plugin
78 for all kinds of web server logs. With it, netdata provides real-time
79 performance information and health monitoring alarms for web applications
82 For more information check:
83 https://github.com/firehol/netdata/wiki/The-spectacles-of-a-web-server-log-file
87 netdata can now archive metrics to `JSON` backends
88 (both push, by @lfdominguez, and pull modes).
92 netdata now has an IPMI plugin (based on freeipmi)
93 for monitoring server hardware.
95 The plugin creates (up to) 8 charts:
97 1. number of sensors by state
98 2. number of events in SEL
99 3. Temperatures CELCIUS
100 4. Temperatures FAHRENHEIT
106 It also supports alarms (including the number of sensors in critical state).
108 For more information, check:
109 https://github.com/firehol/netdata/wiki/monitoring-IPMI
113 @l2isbad builds python data collection plugins for netdata at an wonderfull
116 - **web_log** for monitoring in real-time all kinds of web server log files @l2isbad
117 - **freeipmi** for monitoring IPMI (server hardware)
118 - **nsd** (the [name server daemon](https://www.nlnetlabs.nl/projects/nsd/)) @383c57
119 - **mongodb** @l2isbad
120 - **smartd_log** (monitoring disk S.M.A.R.T. values) @l2isbad
124 - **nfacct** reworked and now collects connection tracker information using netlink.
125 - **ElasticSearch** re-worked @l2isbad
126 - **mysql** re-worked to allow faster development of custom mysql based plugins (MySQLService) @l2isbad
128 - **tomcat** @NMcCloud
129 - **ap** (monitoring hostapd access points)
130 - **php_fpm** @l2isbad
131 - **postgres** @l2isbad
132 - **isc_dhcpd** @l2isbad
133 - **bind_rndc** @l2isbad
135 - **apps.plugin** improvements and freebsd support @vlvkobal
136 - **fail2ban** @l2isbad
137 - **freeradius** @l2isbad
138 - **nut** (monitoring UPSes)
139 - **tc** (Linux QoS) now works on qdiscs instead of classes for the same result (a lot faster) @t-h-e
140 - **varnish** @l2isbad
142 * new and improved alarms
143 - **web_log**, many alarms to detect common web site/API issues
144 - **fping**, alarms to detect packet loss, disconnects and unusually high latency
145 - **cpu**, cpu utilization alarm now ignores `nice`
147 * new and improved alarm notification methods
148 - **HipChat** to allow hosted HipChat @frei-style
149 - **discordapp** @lowfive
151 * dashboard improvements
152 - dashboard now works on HiDPi screens
153 - dashboard now shows version of netdata
154 - dashboard now resets charts properly
155 - dashboard updated to use latest gauge.js release
158 - thanks to @rlefevre netdata now uses a lot of different high resolution system clocks.
160 netdata has received a lot more improvements from many more contributors!
165 netdata (1.5.0) - 2017-01-22
167 * yet another release that makes netdata the fastest
170 * netdata runs on FreeBSD, FreeNAS and MacOS !
172 Vladimir Kobal (@vlvkobal) has done a magnificent work
173 porting netdata to FreeBSD and MacOS.
175 Everyhing works: cpu, memory, disks performance, disks space,
176 network interfaces, interrupts, IPv4 metrics, IPv6 metrics
177 processes, context switches, softnet, IPC queues,
178 IPC semaphores, IPC shared memory, uptime, etc. Wow!
180 * netdata supports data archiving to backend databases:
186 and of course all the compatible ones
187 (KairosDB, InfluxDB, Blueflood, etc)
191 Ilya Mashchenko (@l2isbad) has created most of the python
192 data collection plugins in this release !
194 - systemd Services (using cgroups!)
195 - FPing (yes, network latency in netdata!)
196 - postgres databases @facetoe, @moumoul
197 - Vanish disk cache (v3 and v4) @l2isbad
198 - ElasticSearch @l2isbad
200 - FreeRadius @l2isbad, @lgz
201 - mdstat (RAID) @l2isbad
202 - ISC bind (via rndc) @l2isbad
203 - ISC dhcpd @l2isbad, @lgz
205 - OpenVPN status log @l2isbad, @lgz
208 - gunicorn log @deltaskelta
209 - ECC memory hardware errors
211 - uptime plugin (with a nice badge too)
215 - netfilter conntrack
216 - mysql (replication) @l2isbad
222 - nginx_log @paulfantom
229 - /proc/interrupts @rlefevre
230 - /proc/softirqs @rlefevre
231 - /proc/vmstat (system memory charts)
232 - /proc/net/snmp6 (IPv6 charts)
233 - /proc/self/meminfo (system memory charts)
234 - /proc/net/dev (network interfaces)
237 * new/improved alarms:
239 - MySQL / MariaDB alarms (incl. replication)
245 - netfilter connections alarms
248 * new alarm notifications:
250 - messagebird.com @tech-no-logical
251 - pagerduty.com @jimcooley
252 - pushbullet.com @tperalta82
253 - twilio.com @shadycuz
259 - shell scripts can now query netdata easily!
261 * dashboard improvements:
262 - dashboard is now faster on firefox, safari, opera, edge
263 (edge is still the slowest)
264 - dashboard now has a little bigger fonts
265 - SHIFT + mouse wheel to zoom charts, works on all browsers
266 - perfect-scrollbar on the dashboard
267 - dashboard 4K resolution fixes
268 - dashboard compatibility fixes for embedding charts in
269 third party web sites
270 - charts on custom dashboards can have common min/max
271 even if they come from different netdata servers
272 - alarm log is now saved and loaded back so that
273 the alarm history is available at the dashboard
275 * other improvements:
276 - python.d.plugin has received way to many improvements
277 from many contributors!
278 - charts.d.plugin can now be forked to support
279 multiple independent instances
280 - registry has been re-factored to lower its memory
281 requirements (required for the public registry)
282 - simple patterns in cgroups, disks and alarms
283 - netdata-installer.sh can now correctly install
284 netdata in containers
285 - supplied logrotate script compatibility fixes
286 - spec cleanup @breed808
287 - clocks and timers reworked @rlefevre
289 netdata has received a lot more improvements from many more
295 netdata (1.4.0) - 2016-10-04
299 - the fastest netdata ever (with a better look too)!
300 - improved IoT and containers support!
301 - alarms improved in almost every way!
305 extended TCP metrics,
307 NFS v2, v3 client (server was there already),
308 NFS v4 server & client,
324 Many new alarms have been added to detect common kernel
325 configuration errors and old alarms have been re-worked
326 to avoid notification floods.
328 Alarms now support notification hysteresis (both static
329 and dynamic), notification self-cancellation, dynamic
330 thresholds based on current alarm status
332 * improved alarm notifications
334 netdata now supports:
336 - email notifications
337 - slack.com notifications on slack channels
338 - pushover.net notifications (mobile push notifications)
339 - telegram.org notifications
341 For all the above methods, netdata supports role-based
342 notifications, with multiple recipients for each role
343 and severity filtering per recipient!
345 Also, netdata support HTML5 notifications, while the
346 dashboard is open in a browser window (no need to be
349 All notifications are now clickable to get to the chart
350 that raised the alarm.
352 * improved IoT support!
354 netdata builds and runs with musl libc and runs on systems
357 * improved containers support!
359 netdata runs on alpine linux (a low profile linux distribution
362 * Dozens of other improvements and bugfixes
365 netdata (1.3.0) - 2016-08-28
369 - netdata has health monitoring / alarms!
370 - netdata has badges that can be embeded anywhere!
371 - netdata plugins are now written in Python!
372 - new plugins: redis, memcached, nginx_log, ipfs, apache_cache
375 Since netdata now uses Python plugins, new packages are
376 required to be installed on a system to allow it work.
377 For more information, please check the installation page:
379 https://github.com/firehol/netdata/wiki/Installation
383 * netdata has alarms!
385 Based on the POLL we made on github
386 (https://github.com/firehol/netdata/issues/436),
387 health monitoring was the winner. So here it is!
389 netdata now has a poweful health monitoring system embedded.
390 Please check the wiki page:
392 https://github.com/firehol/netdata/wiki/health-monitoring
394 * netdata has badges!
396 netdata can generate badges with live information from the
398 Please check the wiki page:
400 https://github.com/firehol/netdata/wiki/Generating-Badges
402 * netdata plugins are now written in Python!
404 Thanks to the great work of Paweł Krupa (@paulfantom), most BASH
405 plugins have been ported to Python.
407 The new python.d.plugin supports both python2 and python3 and
408 data collection from multiple sources for all modules.
410 The following pre-existing modules have been ported to Python:
425 The following new modules have been added:
434 * other data collectors:
436 - Thanks to @simonnagl netdata now reports disk space usage.
438 * dashboards now transfer a certain settings from server to server
439 when changing servers via the my-netdata menu.
441 The settings transferred are the dashboard theme, the online
442 help status and current pan and zoom timeframe of the dashboard.
446 - reduction functions now support 'min', 'sum' and 'incremental-sum'.
448 - netdata now offers a multi-threaded and a single threaded
449 web server (single threaded is better for IoT).
451 * apps.plugin improvements:
453 - can now run with command line argument 'without-files'
454 to prevent it from enumating all the open files/sockets/pipes
455 of all running processes.
457 - apps.plugin now scales the collected values to match the
458 the total system usage.
460 - apps.plugin can now report guest CPU usage per process.
462 - repeating errors are now logged once per process.
464 * netdata now runs with IDLE process priority (lower than nice 19)
466 * netdata now instructs the kernel to kill it first when it starves
469 * netdata listens for signals:
471 - SIGHUP to netdata instructs it to re-open its log files
472 (new logrotate files added too).
474 - SIGUSR1 to netdata saves the database
476 - SIGUSR2 to netdata reloads health / alarms configuration
478 * netdata can now bind to multiple IPs and ports.
480 * netdata now has new systemd service file (it starts as user
481 netdata and does not fork).
483 * Dozens of other improvements and bugfixes
486 netdata (1.2.0) - 2016-05-16
490 - netdata is now 30% faster
491 - netdata now has a registry (my-netdata dashboard menu)
492 - netdata now monitors Linux Containers (docker, lxc, etc)
495 This version requires libuuid. The package you need is:
497 - uuid-dev (debian/ubuntu), or
498 - libuuid-devel (centos/fedora/redhat)
502 * netdata is now 30% faster !
504 - Patches submitted by @fredericopissarra improved overall
505 netdata performance by 10%.
507 - A new improved search function in the internal indexes
508 made all searches faster by 50%, resulting in about
509 20% better performance for the core of netdata.
511 - More efficient threads locking in key components
512 contributed to the overal efficiency.
514 * netdata now has a CENTRAL REGISTRY !
516 The central registry tracks all your netdata servers
517 and bookmarks them for you at the 'my-netdata' menu
520 Every netdata can act as a registry, but there is also
521 a global registry provided for free for all netdata users!
523 * netdata now monitors CONTAINERS !
525 docker, lxc, or anything else. For each container it monitors
526 CPU, RAM, DISK I/O (network interfaces were already monitored)
528 * apps.plugin: now uses linux capabilities by default
529 without setuid to root
531 * netdata has now an improved signal handler
534 * API: new improved CORS support
536 * SNMP: counter64 support fixed
538 * MYSQL: more charts, about QCache, MyISAM key cache,
539 InnoDB buffer pools, open files
541 * DISK charts now show mount point when available
543 * Dashboard: improved support for older web browsers
544 and mobile web browsers (thanks to @simonnagl)
546 * Multi-server dashboards now allow de-coupled refreshes for
547 each chart, so that if one netdata has a network latency
548 the other charts are not affected
550 * Several other minor improvements and bugfixes
553 netdata (1.1.0) - 2016-04-20
555 Dozens of commits that improve netdata in several ways:
557 - Data collection: added IPv6 monitoring
558 - Data collection: added SYNPROXY DDoS protection monitoring
559 - Data collection: apps.plugin: added charts for users and user groups
560 - Data collection: apps.plugin: grouping of processes now support patterns
561 - Data collection: apps.plugin: now it is faster, after the new features added
562 - Data collection: better auto-detection of partitions for disk monitoring
563 - Data collection: better fireqos intergation for QoS monitoring
564 - Data collection: squid monitoring now uses squidclient
565 - Data collection: SNMP monitoring now supports 64bit counters
566 - API: fixed issues in CSV output generation
567 - API: netdata can now be restricted to listen on a specific IP
568 - Core and apps.plugin: error log flood protection
569 - Dashboard: better error handling when the netdata server is unreachable
570 - Dashboard: each chart now has a toolbox
571 - Dashboard: on-line help support
572 - Dashboard: check for netdata updates button
573 - Dashboard: added example /tv.html dashboard
574 - Packaging: now compiles with musl libc (alpine linux)
575 - Packaging: added debian packaging
576 - Packaging: support non-root installations
577 - Packaging: the installer generates uninstall script
579 netdata (1.0.0) - 2016-03-22
581 - first public release
583 netdata (1.0.0-rc.1) - 2015-11-28