Logs from linux machine is unreadable

I have logserver running on ubuntu 14.04. ELK versions are 2 or above(kibana 4.4.x).
I am also using nagios to monitor systems and applications. Now I need to sent error messages from nagios to logserver.
Nagios is installed on centos . Nagios error messages are writing into /usr/local/nagios/var/archives/ . I am planning to sent this error logs to logserver. (Using a shell script I keep only one log file in that folder . Errors will write into that log file)

For that, I installed filebeat on centos.

Logstash configuration :

input{
     tcp{
         port => 5550
         type => "nagios"
         codec => json {
          charset => "UTF-8"
            }
      }
}

output {
 if [type] == "nagios"  {
    elasticsearch {
        hosts => ["ipaddress:9200"]
        }
     }
 }

Now my problem is there is no readable information in logs getting in kibana .
Below is an example. I am getting only this type of logs.

How to get the actual error information in kibana. ??

So it's Filebeat that sends to port 5550? Use a beats input, not a tcp input.

No logs are getting with beats input plugin

But that's not what the configuration snippet you posted shows.