{"id":11699,"date":"2019-03-15T22:38:54","date_gmt":"2019-03-15T22:38:54","guid":{"rendered":"http:\/\/www.appservgrid.com\/paw92\/?p=11699"},"modified":"2019-03-15T22:38:54","modified_gmt":"2019-03-15T22:38:54","slug":"how-to-install-and-configure-collectd-and-collectd-web-to-monitor-server-resources-in-linux","status":"publish","type":"post","link":"https:\/\/www.appservgrid.com\/paw92\/index.php\/2019\/03\/15\/how-to-install-and-configure-collectd-and-collectd-web-to-monitor-server-resources-in-linux\/","title":{"rendered":"How to Install and Configure \u2018Collectd\u2019 and \u2018Collectd-Web\u2019 to Monitor Server Resources in Linux"},"content":{"rendered":"

Collectd-web<\/b>\u00a0is a web front-end monitoring tool based on RRDtool (R<\/b>ound-R<\/b>obin\u00a0D<\/b>atabase\u00a0Tool)<\/b>, which interprets and graphical outputs the data collected by the\u00a0Collectd<\/strong>\u00a0service on Linux systems.<\/p>\n

Collectd<\/strong>\u00a0service comes by default with a huge collection of available plug-ins into its default configuration file, some of them being, by default, already activated once you have installed the software package.<\/p>\n

Collectd-web CGI<\/strong>\u00a0scripts which interprets and generates the graphical html page statistics can be simply executed by the\u00a0Apache CGI<\/strong>\u00a0gateway with minimal of configurations required on Apache web server side.<\/p>\n

However, the graphical web interface with the generated statistics can, also, be executed by the standalone web server offered by\u00a0Python CGIHTTPServer<\/strong>\u00a0script that comes pre-installed with the main\u00a0Git<\/strong>\u00a0repository.<\/p>\n

This tutorial will cover the installation process of\u00a0Collectd<\/strong>\u00a0service and\u00a0Collectd-web<\/strong>\u00a0interface on\u00a0RHEL\/CentOS\/Fedora<\/strong>\u00a0and\u00a0Ubuntu\/Debian<\/strong>\u00a0based systems with the minimal configurations needed to be done in order to run the services and to enable a\u00a0Collectd<\/strong>\u00a0service plug-in.<\/p>\n

Please go through the following articles of\u00a0collectd<\/strong>\u00a0series.<\/p>\n

Part 1<\/b>:\u00a0Install and Configure \u2018Collectd\u2019 and \u2018Collectd-Web\u2019 to Monitor Linux Resources<\/b><\/div>\n
Part 2<\/b>:\u00a0Monitor Linux Resources with Collectd-web and Apache CGI<\/a><\/div>\n
Part 3<\/b>:\u00a0Configure Collectd as a Central Monitoring Server for Clients<\/a><\/div>\n

Step 1: \u2013 Install Collectd Service<\/h3>\n

1.<\/strong>\u00a0Basically, the\u00a0Collectd<\/strong>\u00a0daemon task is to gather and store data statistics on the system that it runs on. The\u00a0Collectd<\/strong>\u00a0package can be downloaded and installed from the default Debian based distribution repositories by issuing the following command:<\/p>\n

On Ubuntu\/Debian<\/h5>\n
# apt-get install collectd\t\t\t[On Debian<\/strong> based Systems]\r\n<\/pre>\n
\n
\"Install<\/a><\/p>\n
Install Collectd on Debian\/Ubuntu<\/p>\n<\/div>\n
On RHEL\/CentOS 6.x\/5.x<\/h5>\n
On older\u00a0RedHat<\/strong>\u00a0based systems like\u00a0CentOS\/Fedora<\/strong>, you first need to\u00a0enable epel repository<\/a>\u00a0under your system, then you can able to install\u00a0collectd<\/strong>\u00a0package from the epel repository.<\/p>\n
# yum install collectd\r\n<\/pre>\n
On RHEL\/CentOS 7.x<\/h5>\n
On latest version of RHEL\/CentOS 7.x, you can install and enable epel repository from default yum repos as shown below.<\/p>\n
# yum install epel-release\r\n# yum install collectd\r\n<\/pre>\n
\n
\"Install<\/a><\/p>\n
Install Collectd on CentOS\/RHEL\/Fedora<\/p>\n<\/div>\n
Note:<\/strong>\u00a0For Fedora users, no need to enable any third party repositories, simple yum to get the collectd package from default yum repositories.<\/p>\n
2.<\/strong>\u00a0Once the package is installed on your system, run the below command in order to start the service.<\/p>\n
# service collectd start\t\t\t[On Debian<\/strong> based Systems]\r\n# service collectd start                        [On RHEL\/CentOS 6.x\/5.x<\/strong> Systems]\r\n# systemctl start collectd.service              [On RHEL\/CentOS 7.x<\/strong> Systems]\r\n<\/pre>\n
Step 2: Install Collectd-Web and Dependencies<\/h3>\n
3.<\/strong>\u00a0Before starting to import the\u00a0Collectd-web<\/strong>\u00a0Git repository, first you need to assure that\u00a0Git<\/strong>\u00a0software package and the following required dependencies are installed on your machine:<\/p>\n
----------------- On Debian \/ Ubuntu<\/strong> systems -----------------\r\n# apt-get install git\r\n# apt-get install librrds-perl libjson-perl libhtml-parser-perl\r\n<\/pre>\n
\n
\"Install<\/a><\/p>\n
Install Git on Debian\/Ubuntu<\/p>\n<\/div>\n
----------------- On RedHat\/CentOS\/Fedora<\/strong> based systems -----------------\r\n# yum install git\r\n# yum install rrdtool rrdtool-devel rrdtool-perl perl-HTML-Parser perl-JSON\r\n<\/pre>\n
\n
\"Install<\/a><\/p>\n
Install Git and Dependencies<\/p>\n<\/div>\n
Step 3: Import Collectd-Web Git Repository and Modify Standalone Python Server<\/h3>\n
4.<\/strong>\u00a0On the next step choose and change the directory to a system path from the Linux tree hierarchy where you want to import the Git project (you can use\u00a0\/usr\/local\/<\/code>\u00a0path), then run the following command to clone\u00a0Collectd-web<\/strong>\u00a0git repository:<\/p>\n
# cd \/usr\/local\/\r\n# git clone https:\/\/github.com\/httpdss\/collectd-web.git\r\n<\/pre>\n
\n
\"Git<\/a><\/p>\n
Git Clone Collectd-Web<\/p>\n<\/div>\n
5.<\/strong>\u00a0Once the Git repository is imported into your system, go ahead and enter the\u00a0collectd-web<\/strong>\u00a0directory and list its contents in order to identify the Python server script (runserver.py<\/code>), which will be modified on the next step. Also, add execution permissions to the following CGI script:\u00a0graphdefs.cgi<\/code>.<\/p>\n
# cd collectd-web\/\r\n# ls\r\n# chmod +x cgi-bin\/graphdefs.cgi\r\n<\/pre>\n
\n
\"Set<\/a><\/p>\n
Set Execute Permission<\/p>\n<\/div>\n
6.<\/strong>\u00a0Collectd-web<\/strong>\u00a0standalone Python server script is configured by default to run and bind only on\u00a0loopback address (127.0.0.1)<\/strong>.<\/p>\n
In order to access\u00a0Collectd-web<\/strong>\u00a0interface from a remote browser, you need to edit the\u00a0runserver.py<\/code>\u00a0script and change the\u00a0127.0.1.1 IP<\/strong>\u00a0Address to\u00a00.0.0.0<\/strong>, in order to bind on all network interfaces IP Addresses.<\/p>\n
If you want to bind only on a specific interface, then use that interface IP Address (not advised to use this option in case your network interface Address is dynamically allocated by a DHCP server). Use the below screenshot as an excerpt on how the final\u00a0runserver.py<\/code>\u00a0script should look like:<\/p>\n
# nano runserver.py\r\n<\/pre>\n
\n
\"Configure<\/a><\/p>\n
Configure Collect-web<\/p>\n<\/div>\n
If you want to use another network port than\u00a08888<\/strong>, modify the PORT variable value.<\/p>\n
Step 4: Run Python CGI Standalone Server and Browse Collectd-web Interface<\/h3>\n
7.<\/strong>\u00a0After you have modified the standalone Python server script IP Address binding, go ahead and start the server in background by issuing the following command:<\/p>\n
# .\/runserver.py &\r\n<\/pre>\n
Optional, as an alternate method you can call the Python interpreter to start the server:<\/p>\n
# python runserver.py &\r\n<\/pre>\n
\n
\"Start<\/a><\/p>\n
Start Collect-Web Server<\/p>\n
8.<\/strong>\u00a0To visit\u00a0Collectd-web<\/strong>\u00a0interface and display statistics about your host, open a browser and point the URL at your server IP Address and port\u00a08888<\/strong>\u00a0using HTTP protocol.<\/p>\n
By default you will see a number of graphics about CPU, disk usage, network traffic, RAM, processes and other system resources by clicking on the hostname displayed on Hosts form.<\/p>\n
http:\/\/192.168.1.211:8888\r\n<\/pre>\n
\n
\"Access<\/a><\/p>\n
Access Collect-Web Panel<\/p>\n<\/div>\n
\n
\"Linux<\/a><\/p>\n
Linux Disk Monitoring<\/p>\n<\/div>\n
9.<\/strong>\u00a0To stop the standalone Python server issue the below command or you may cancel or stop the script by hitting\u00a0Ctrl+c<\/strong>\u00a0key:<\/p>\n
# killall python\r\n<\/pre>\n
Step 5: Create a Custom Bash Script to Manage the Standalone Python Server<\/h3>\n
10.<\/strong>\u00a0To manage the standalone\u00a0PyhtonCGIServer<\/strong>\u00a0script more easily (start<\/strong>,\u00a0stop<\/strong>\u00a0and view\u00a0status<\/strong>), create the following\u00a0collectd-server<\/code>\u00a0Bash script at a system executable path with the following configurations:<\/p>\n
# nano \/usr\/local\/bin\/collectd-server\r\n<\/pre>\n