-
Notifications
You must be signed in to change notification settings - Fork 4
/
README.rtst
82 lines (63 loc) · 2.96 KB
/
README.rtst
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
#
# LPCPU (Linux Performance Customer Profiler Utility): ./README.rtst
#
# (C) Copyright IBM Corp. 2018
#
# This file is subject to the terms and conditions of the Eclipse
# Public License. See the file LICENSE.TXT in the main directory of the
# distribution for more details.
#
RTST - Real Time Server Telemetry
1. Overview
2. Dependencies
3. Running RTST
4. Viewing the data
1. Overview
===========
RTST is a framework for viewing server telemetry in real time via a
remote web browser. By using standard Internet technologies (HTTP,
HTML, Javascript, etc.) RTST aims to be easily portable across
environments.
RTST currently supports 4 charts: system wide CPU, system wide disk
IO, selectable network interface monitoring, and system wide memory
usage.
2. Dependencies
===============
RTST requires that Python and sar (sysstat package) be installed.
3. Running RTST
===============
RTST aims to be flexibly configured via the command line. All
configuration options are setup as required command line options.
Here is the help output for RTST:
./rtst.py --help
Required options for ./rtst.py are:
--server-interval=<int> How often should the server collect samples
--client-interval=<int> How often should the client request new data from the server
--server-history=<int> How many samples should the server buffer for sample history
--client-history=<int> How many samples should the client buffer for sample history by default (client can dynamically change this value)
--server-name=<string> What is the accessible hostname the client should use to access the server
--server-port=<int> What TCP port should the server listen on (must be open through any/all firewalls)
--selected-interfaces=<string> Comma separated list of interfaces to monitor
Available interfaces for monitoring are: lo eth4 eth0 eth1 eth2 eth3 usb0
As discussed in the LPCPU README file, the charting technology used by
LPCPU/RTST must be cognizant of some browser security issues in order
to function properly. The --server-name=<string> parameter is what
allows this to be handled. The DNS hostname that the client will use
to connect to the server must be supplied as the value to this
argument otherwise the browser may refuse to load the chart data.
Here is an example invocation:
./rtst.py --server-interval=1 --client-interval=1 --server-history=120 --client-history=900 --server-name=my.server.hostname.com --server-port=1338 --selected-interfaces=eth0,eth1,eth4
Main Thread: Started...
Monitoring eth0,eth1,eth4
Monitoring Interval: 1
Client Update Interval: 1
History Length: 120
Client History Length: 900
Server Name: my.server.hostname.com
Server Port: 1338
To view the statistics, load http://my.server.hostname.com:1338 in your web browser
To exit the server, use CTRL-C
4. Viewing the data
===================
To view the data, simply open the URL that the script provides in your
web browser.