new
Costa Tsaousis (ktsaou) committed
Apr 27, 2014 at 16:40 UTC
bd247168b5bcde5234079cd04c80e4d1bf24ebac
1 file changed
+49
-40
README.md
+49
-40
@@ -3,13 +3,44 @@ netdata
3
4
linux real time system monitoring
5
6
-This program is a daemon that collects system information from /proc and other sources.
7
-It is heavily optimized and lightweight.
6
+This program is a daemon that collects system information from a linux system and presents a web site to view the data.
7
+This presentation is full of charts that precisely render all system values, in realtime, for a short time (1 hour by default).
8
9
-It updates everything every second!
10
-But tt only needs a few microseconds (just a fraction of a millisecond) of one of your cores and a few megabytes of memory.
9
+# Features
10
+
11
+- highly optimized C code
12
+ it only needs a few milliseconds per second to collect all the data.
13
+
14
+- extremely lightweight
15
+ it only needs a few megabytes of memory to store all its round robin database.
16
+
17
+- per second data collection
18
+ every chart, every value, is updated every second.
19
+
20
+- visualizes QoS classes automatically
21
+ if you also use fireqos for QoS, it even collects class names automatically.
22
+
23
+- the generated web site uses bootstrap and google charts for a very appealing result
24
+ it works even on mobile devices, adapts to screen size changes and rotation.
25
+
26
+- web charts do respect your browser resources
27
+ the charts adapt to show only as many points are required to have a clear view.
28
+
29
+- highly configurable
30
+ all charts and all features can be enabled or disabled.
31
+
32
+- it reads and renders charts for all these:
33
+ - /proc/net/dev (all netwrok interfaces for all their values)
34
+ - /proc/diskstats (all disks for all their values)
35
+ - /proc/net/snmp (total IPv4, TCP and UDP usage)
36
+ - /proc/net/netstat (more IPv4 usage)
37
+ - /proc/net/stat/nf_conntrack (connection tracking performance)
38
+ - /proc/net/ip_vs/stats (IPVS connection statistics)
39
+ - /proc/stat (CPU utilization)
40
+ - /proc/meminfo (memory information)
41
+ - /proc/vmstat (system performance)
42
+ - tc classes (QoS classes)
43
12
-If listens on port 19999 and it will give you a full featured web interface using google charts.
44
45
Check it live at:
46
@@ -19,6 +50,7 @@ Here is a screenshot:
50
51
52
53
+
54
# How it works
55
56
1. You run a daemon on your linux: netdata.
@@ -39,7 +71,7 @@ Here is a screenshot:
71
The above is equivalent to:
72
73
```
42
- http://127.0.0.1:19999/data/net.eth0/3600/1/average
74
+ http://127.0.0.1:19999/data/net.eth0/3600/1/average/0/0
75
```
76
77
where:
@@ -47,6 +79,7 @@ Here is a screenshot:
79
- 3600 is the number of entries to generate (3600 is a default which can be overwritten by -l).
80
- 1 is grouping count, 1 = every single entry, 2 = half the entries, 3 = one every 3 entries, etc
81
- `average` is the grouping method. It can also be `max`.
82
+ - 0/0 they are `before` and `after` timestamps, allowing panning on the data
83
84
85
2. On your web page, you add a few javascript lines and a DIV for every graph you need.
@@ -60,49 +93,25 @@ Here is a screenshot:
93
94
## Automatic installation
95
63
-1. Download the git.
64
-2. cd to that directory
65
-3. run:
96
+Before you start, make sure you have `zlib` development files installed.
97
+To install it in Ubuntu, you need to run:
98
99
```sh
68
-./netdata.start
100
+apt-get install zlib1g-dev
101
```
102
71
-Once you run it, the file netdata.conf will be created. You can edit this file to set options for each graph.
72
-To apply the changes you made, you have to run netdata.start again.
73
-
74
-To access the web site for all graphs, go to:
75
-
76
- ```
77
- http://127.0.0.1:19999/
78
- ```
79
-
80
-
81
-
82
----
83
-
84
-## Installation by hand
85
-
86
-### Compile netdata
87
-step into the directory you downloaded everything and compile netdata
103
+Then do this to install and run netdata:
104
105
```sh
90
-gcc -Wall -O3 -o netdata netdata.c -lpthread -lz
106
+git clone https://github.com/ktsaou/netdata.git netdata
107
+cd netdata
108
+./netdata.start
109
```
110
93
-### run netdata
94
-Run netdata:
95
-
96
-```sh
97
-./netdata -u nobody -d -l 1200 -u 1 -p 19999
98
-```
99
- - -d says to go daemon mode
100
- - -l 1200 says to keep 1200 history entries (20 minutes)
101
- - -t 1 says to add a new history entry every second
102
- - -p 19999 is the port to listen for web clients
103
- - -u nobody to run as nobody
111
+Once you run it, the file netdata.conf will be created. You can edit this file to set options for each graph.
112
+To apply the changes you made, you have to run netdata.start again.
113
105
-### open a web browser
114
+To access the web site for all graphs, go to:
115
116
```
117
http://127.0.0.1:19999/