Removing supergroups

Remove a supergroup for multiple groups that have a common event.

1. To access the master.cea_properties file, use the rsh command to go to the ncoprimary pod as shown in the following example:

   oc rsh $(oc get pods|grep ncoprimary|head -1|awk '{print $1}')

2. From within the pod, run the nco_sql command:

   /opt/IBM/tivoli/netcool/omnibus/bin/nco_sql -user root -password $OMNIBUS_ROOT_PWD -server AGG_P

3. Check for the existence of the CEASupergroupEnabled property:

   select IntValue from  master.cea_properties  where Name ='CEASupergroupEnabled';

   If 0 rows are returned, the CEASupergroupEnabled property doesn't exist, and you can insert it with the following command:

   insert into master.cea_properties (Name, IntValue) values('CEASupergroupEnabled', 0);

4. In the master.cea_properties file, set the value of the CEASupergroupEnabled property to 0 with the following command:

   UPDATE master.cea_properties set IntValue=0 where Name ='CEASupergroupEnabled';

5. To notify the collator that supergrouping is removed, set the COLLATER_SUPER_GROUPING_ENABLED environment variable to false by using one of the following methods.
   • Find the analytics-dev-collater-aggregationservice collator deployment. Add the COLLATER_SUPER_GROUPING_ENABLED variable and set it to false, or set the existing COLLATER_SUPER_GROUPING_ENABLED variable to false. Then, restart the collator.
   • Alternatively, enable the normalizer swagger. For more information about enabling the normalizer swagger, see Configure aggregation settings. Then, add the property by using the post method. Format the JSON property first and add "superGroupingEnabled": false as shown in the following example:

     {
     "groupAggregationConfiguration": {
     "windowType": "rolling",
     "durationSeconds": 1200
     },
     "supergroupAggregationConfiguration": {
     "windowType": "rolling",
     "durationSeconds": 1200,
     "superGroupingEnabled": false
     }
     }

     With this method, you don't need to restart the deployment for the change to take effect.

6. Additionally, you can set group ranking so that the common event is moved to the highest-ranked group that uses the deployment or normalizer connection. Update the group ranking only when necessary. Without an update, a group ranking uses the default ranking settings.
   • For a deployment, add the following environment variable:

     COLLATER_SUPER_GROUPING_TYPE_RANK_LIST=analytics.temporal-patterns,related-events,scope,topological-correlation

     After you set the environment variable, restart the collator.

   • For a normalizer connection, add the following property to the JSON string as shown in the following example:

     {
         "groupAggregationConfiguration": {
             "windowType": "rolling",
             "durationSeconds": 1200
         },
         "supergroupAggregationConfiguration": {
             "windowType": "rolling",
             "durationSeconds": 1200,
             "superGroupingEnabled": false,
             "superGroupingTypeRankList": "analytics.temporal-patterns,related-events,scope,topological-correlation"
         }
     }