Set up Splunk on linux

0. Download splunk install package 

http://www.splunk.com/download/splunk-6.1.2-213098-linux-2.6-x86_64.rpm

1. Install enterprise :

    npm -i splunk-6.1.2-213098-linux-2.6-x86_64.rpm

2. start splunk:

    cd /opt/splunk/bin

    ./splunk start --accept-license

   ./splunk enable boot-start -user root

3. Install universal forwarder:

    npm -i splunkforwarder-6.1.2-213098-linux-2.6-x86_64.rpm

4. start splunk forwarder:

      cd /opt/splunkforwarder/bin

    ./splunk start --accept-license

   ./splunk enable boot-start -user root

5. Change forwarder admin password:

     cd /opt/splunkforwarder/bin

    ./splunk edit user admin -password <new password> -role admin admin: changeme <changed to forwardme>

6. configure universal forwarder act as a deplyment client:

    ./splunk set deploy-poll 127.0.0.1:8089

7. configure universal forwarder to forward a specific receiving indexer:

  ./splunk add forward-server 127.0.0.1:9997 admin:forwardme

8. configure forwarder inputs.conf:

   cd /opt/splunkforwarder/etc/system/local

   gedit inputs.conf

[monitor:<the directory you would like to monitor> ] //my sample: /home/aimqa/Desktop/SG_JobsResults

disabled=false

sourcetype=<your sourcetype name that you need to set up on server> //my sample: sg_production

9. Additional setting:

     if you want to clone your data to the end sever, you may clone data to another server by:

    cd /opt/splunkforwarder/etc/system/local

    gedit outputs.conf

before edit, you should get the outputs.conf like this:

   [tcpout]

  defaultGroup=<target_group>

 [tcpout:<target_group>]

server=<receiving_server1>:<port><attribute1> = <val1><attribute2> = <val2>

to set up date clone, modify the outputs.conf to: 

 [tcpout]
  defaultGroup=<target_group1>,<target_group2>
 [tcpout:<target_group1>]
   server=<receiving_server1>:<port>
 
 [tcpout:<target_group2>]
   server=<receiving_server2>:<port>
10. add more monitor directory to added to different sourcetype:
    
   cd /opt/splunkforwarder/etc/system/local
   gedit inputs.conf
   add another line:
   [monitor:<the directory you would like to monitor> ] //my sample: /home/aimqa/Desktop/SG_JobsResults
11. Now you could set up a data monitor to grasp data you want to monitor and added it to the monitor directory so that data is forwarded to server for deeper search.

    I used crontab on linux to repeat query data and populate to monitor directory: 

    /sbin/service crond stop

    crontab -e

    0 * * * * /bin/sh <your bash file .sh>

    //before I start the timely update log, I queried all history data and forwarded it to server, then I start crond to query data once an hour at sharp time 

    /sbin/service crond start