Troubleshooting

If all the SNMP-based tests pertaining to the Cache database server are in an UNKNOWN state, then do the following:

  • Check whether the eG agent is up and running
  • Ensure that there are no issues in the agent-manager communication
  • Check whether the Windows SNMP service has been installed and started on the Cache host
  • Verify whether the Cache Monitoring Service has been enabled (refer to Pre-requisites for monitoring an Intersystems Cache Database Server for details)
  • Verify whether the Cache SNMP Agent has been set to start automatically at Cache startup (refer to Pre-requisites for monitoring an Intersystems Cache Database Server for more details)
  • Check whether the critical SNMP base classes such as %Monitor.System.Freespace, %Monitor.System.Database, %Monitor.System.Processes, and %Monitor.System.SystemMetrics were activated from the Cache Terminal. 

  • If one/more SNMP-based tests pertaining to the Cache database server are not reporting measures, then go to the command prompt of the eG agent host, and execute the following command from the <eg_agent_install_dir>\bin directory, to see if the desired statistics are being retrieved from the SNMP MIB of the Cache database server:

    snmpwalk nfq -O 161 <SNMP_community_string_configured_for_the_tests> <IP/hostname_of_the_CacheDB> <OID_of_the_measure>

    For instance, to view the value of a particular measure reported by the Cache Buffer test for a Cache database server with IP 192.16.10.51 (community string: public), your command would be:

    snmpwalk nfq -O 161 public 192.168.10.51 .1.3.6.1.4.1.16563.1.1.7.1.2

  • Given below are the details of OIDs that are associated with the tests mapped to the Cache database server model. You will find that most tests are associated with a wide range of OIDs, one for every measure reported by the test. To check whether a test is reporting measures or not, it would suffice to run the above-mentioned command using any one of the OIDs in the given range, and not each one of them.

    • For the Cache Buffer test, use the OIDs in the range, .1.3.6.1.4.1.16563.1.1.7.1.2 to .1.3.6.1.4.1.16563.1.1.7.1.17

    • For the Cache Database test, use any of the following OIDs:

      .1.3.6.1.4.1.16563.1.1.3.1.5

      .1.3.6.1.4.1.16563.1.1.3.1.6

      .1.3.6.1.4.1.16563.1.1.3.1.7

    • For the Cache Performance test, use the OIDs in the range .1.3.6.1.4.1.16563.1.1.2.1.2 to .1.3.6.1.4.1.16563.1.1.2.1.18
    • For the Cache Resource test, use the OID, .1.3.6.1.4.1.16563.1.1.6.1.3
    • For the Cache Systems test, use the OIDs in the range .1.3.6.1.4.1.16563.1.1.1.1.6 to .1.3.6.1.4.1.16563.1.1.1.1.11
    • For the Ecp Application Server test, use the OIDs in the range, .1.3.6.1.4.1.16563.1.1.4.1.1 to .1.3.6.1.4.1.16563.1.1.4.1.3
    • For the Ecp Data Server test, use the OIDs in the range, .1.3.6.1.4.1.16563.1.1.5.1.1 to .1.3.6.1.4.1.16563.1.1.5.1.5

  • Similarly, if one/more tests using the cstat utility are not reporting measures, then issue the cstat command from the command prompt to see if it returns a valid output. For that, follow the steps given below:

    • Go to the command prompt on the Cache host.
    • Switch to the <CACHE_INSTALL_DIR>.
    • Issue the command: cstat –s <CACHE_INSTALL_DIR>\mgr -e2 –m-1 –n3 –j5 –g1 –m3 –L1 –u-1 –v1 –p-1 –c-1 –q1 –w2 –S –1

    For example, if the Cache instance being monitor is installed in c:\CacheSys, then the command would be:

    cstat –s c:\CacheSys\mgr -e2 –m-1 –n3 –j5 –g1 –m3 –L1 –u-1 –v1 –p-1 –c-1 –q1 –w2 –S –1