Alert Central (Availability)

Version 4

    This template collects Alert Central general statistics such as processes and TCP ports.


    Prerequisites:

    • This template uses one command (snmptable) from net-snmp-utils. These utilities should be installed on the target server.
    • SNMP installed on the Alert Central server. Additional configurations should be added into SNMPD configuration (snmpd.conf) to allow monitoring of the Alert Central processes:
      rocommunity public
      proc postmaster 30 1
      proc java 2 1
      proc vami-lighttpd 1 1
      proc ntpd 1 1
      load 4 4 4

    Following is a test showing that process monitoring works on Alert Central:

    $ snmpget -Os -c public -v2c 10.199.18.97 .1.3.6.1.2.1.1.4.0

    1. sysContact.0 = STRING: Alert Central support@solarwinds.com

    $ snmptable -Os -c public -v 1 10.199.18.97 .1.3.6.1.4.1.2021.2

    SNMP table: prTable

    prIndex prNames prMin prMax prCount prErrorFlag prErrMessage prErrFix prErrFixCmd

    1    postmaster     1 30      55       error Too many postmaster running (# = 55)  noError

    2 java     1 2       1 noError noError

    3 vami-lighttpd     1 1       1 noError noError

    4 ntpd     1 1       1 noError noError


    Credentials: Use the default Alert Central Admin account


    Monitored Components

    Process: postmaster

    This monitor returns the CPU and memory usage of the postmaster process. The database service spawns separate postmaster processed for each connection.

     

    Process: java

    This monitor returns the CPU and memory usage of the java process. This process runs a web application for the Alert Central user interface and business logic running on port 80.

     

    Process: vami-lighttpd

    This monitor returns the CPU and memory usage of vami-lighttpd process. This process runs a web application for the virtual appliance management interface running on port 5480.

     

    Process: ntpd

    This monitor returns the CPU and memory usage of ntpd process. This process keeps the computer clock in sync with internet time servers.

     

    TCP Port: HTTP Web Console

    This component monitor tests the ability of the Alert Central Web Server to accept incoming sessions. By default, it monitors port 80.

     

    TCP Port: HTTPS Web Console

    This component monitor tests the ability of the Alert Central Web Server to accept incoming sessions. By default, it monitors port 443.

     

    TCP Port: SSH

    This component monitor tests the ability of the SSH service to accept incoming sessions. By default, it monitors port 22.

     

    TCP Port: HTTPS VAMI Web Console

    This component monitor tests the ability of the VAMI Web Console to accept incoming sessions. This is a web application for the virtual appliance management interface. By default, it monitors port 5480.


    TCP Port: JMX

    This component monitor tests the ability of JMX to accept incoming sessions. These are java management extensions for monitoring statistics for the Alert Central service. By default, it monitors port 9004.

     

    Alert Central Processes

    This monitor returns the number of specific Alert Central processes:

      - postmaster  - A database service that spawns separate processes for each connection. The number of processes should be less than 50.
      - java – A web application for the Alert Central user interface and business logic. The number of processes should be 1 or 2.
      - vami-lighttpd – A web application for the virtual appliance management interface. The number of processes should be 1.
      - ntpd – This keeps the computer clock in sync with internet time servers. The number of processes should be 1. This monitor requires the following argument:
       SNMP_community_string – The community string which allows monitoring of these processes. By default the argument set to public.

    Below is an example using the Scripts Arguments field:
       perl ${SCRIPT} public

     

    Alert Central System Load Average

    This monitor returns the System Load Average for 1, 5, and 15 minutes. The returned values should be less than 4.

    This monitor requires the following argument:
       SNMP_community_string – The community string which allows monitoring of these processes. By default the argument set to public.

    Below is an example using the Scripts Arguments field:
       perl ${SCRIPT} public