Steps to remove unneeded Metric Instances from MOI 2.1 MPMUI Differential Analysis display

Products

• Mainframe Operational Intelligence

Issue/Introduction

NOTE:  This is an internal process and should ONLY BE DONE with MOI Support/Engineering assistance!

 A MOI Customer attempts to update the metric filters using the MOI Appliance Controller Configuration >  Metrics option.   The update attempt times out with page not responding error or metric instances are displayed in the MPMUI that no longer have metric data associated with them and are causing clutter for the Customer.    This Knowledge Document details the steps to resolve the page not responding error as follows:

• Export the current metric filters to save filters for import later in the steps.
• Stop the maas microservice to stop metric data flow to MOI and then shut down MOI.
• Perform a backup of the RDF database just to be safe.
• Remove the RDF database to clear out voluminous metric instances.
• Restart MOI
• Stop the maas microservice to prevent more metric instances from being created.
• Import the save metric filters.
• Update the metric filters as required
• Start maas microservice.

Environment

Release:  2.1.00

Component:  MF OPERATIONAL INTELLIGENCE

Resolution

NOTE:  This process should ONLY BE DONE with MOI Support/Engineering assistance!   Please contact Support before proceeding with the following steps.

 

1. Download the moi-filters-util_1651237168310.zip file attached to this Knowledge Document to the MOI appliance Linux system in any directory.  Recommendation is that you can place the zip file in /tmp.

 

2. Logon to the MOI appliance Linux system as root and cd to the directory that you downloaded the zip file to.  Extract the zip file contents by running the following command:

            >tar -xvf moi-filters-util.zip

3. Run the following command to export the filters to a file

            >sh export-filters.sh

4. On successful completion of the script, you will see the following message

            “Saved filters to file: /var/opt/CA/itoa/profiler/rdf/backup/metric-filters.json”

5. Once the filters are exported, first bring down the maas microservice using the following commands. This is to stop all the incoming data.
                                

                         > docker stop oi-engine_profiler-maas_1    
   
   Once the  maas micro service is down we need to make sure that all the data that was inflight is processed. To check this, run the kafka-lag.sh script using the below command and check the lag column for OI-TIMESERIES-RAW-PAYLOAD-RECORDS, OI-SMF-RECORDS, OI-DATASTORE.0 topics. If the lag is not zero then give it a few minutes and check the lag again until the lag is cleared.  

                        

   > ./kafka-lag.sh 

    

   Once the lag is cleared, proceed on to the next step to shutdown MOI.

 
                        > cd /opt/moi/
                        > bin/plex shutdown --bg=wait

                      

6. Run the following command to verify that the MOI microservices are down:

            >docker ps –a

7. Once the services are down, first take a backup of RDF using the following commands:
   
   >cd /tmp
               >tar -czvf rdf-backup.zip /var/opt/moi/data2/profiler/var/opt/CA/itoa/profiler/rdf/tdb
   
   
   
8. Next remove the RDF database using the command:
   
   > rm -rf /var/opt/moi/data2/profiler/var/opt/CA/itoa/profiler/rdf/tdb
9. Start up MOI using the following commands:

            >cd /opt/moi

            >bin/plex startup --bg=wait

10. Using the following command, check to see if the maas container is created:

            > docker ps -a | grep maas

11. Once the maas microservice is created, stop the maas microservice

            >docker stop oi-engine_profiler-maas_1

12. Check if the MTC AC (appliance controller) is healthy using the following command:

            >docker ps -a | grep mtcac     

13. Once the appliance controller is healthy, cd to the directory where the moi-filters-util.zip file was extracted.
14. Now import the filters using the following command:  (Please note that this process has taken over 1 hour on resource constrained Linux machines)  

           >sh import-filters.sh

15. Once the filters are imported, you will see the following message:

            “Filters imported successfully!!”

16. Now login to the appliance controller and update the SMF 101 filters and any others that are needed using the Configuration> Metrics option on the MOI Appliance Controller menu.

 

17. Once the filters are set, start maas using the following command:

            >docker start oi-engine_profiler-maas_1

 

Scripts/Directories

The attached moi-filters-util_1651237168310.zip file contains the following components:

 

kafka-lag.sh

export-filters.sh

import-filters.sh

backup-graphdb/

backup-graphdb/oi-engine-jena-backup-1.0.01-SNAPSHOT.jar

backup-graphdb/install.sh

backup-graphdb/backup-graphdb.sh