AIOps

This article gives insights about Topology Match Tokens in Watson AIOps.

The detailed documentation about defining merge rules is available in IBM documentation https://www.ibm.com/docs/en/cloud-paks/cloud-pak-watson-aiops/3.4.2?topic=elements-configuring-rules

1. Create Match Token

1. Create match token with the below condition.

• Tokens : name
• Resource Types : deployment

2. Merge Token Created 

2. View Resource Details

The merge token is created. Let us review the Resource details with the merge token.

1. Right click on the ilender-user deployment resource and click on the Resource details menu. 

2. Here is the resource details of the ilender-user deployment.

You can see the properties like _id, resourcetype and etc.

Click on the Show Json option button. 

3. The json view of the ilender-user deployment resource properties is displayed.

The json file is available here.

You can see the properties like _id, resourcetype and etc.

You can also note that the matchTokens property contains the values of

• unique id : f35675ff-be21-4851-a892-3cd1a5f7721b
• deployment resource name : ilender-user

    "matchTokens": [
        "f35675ff-be21-4851-a892-3cd1a5f7721b",
        "ilender-user"
    ],

3. Pushing Events with Match Token

Lets push an event to kafka topic using match token and see in Alert Viewer.

1. Create Kafka topic like this 

2. Create Event file like the below. Sample file is available here

The ilender-user value from this "application":"ilender-user" property would be taken and matched with fields _id and matchTokens. 

3. Push the events to kafka.

Refer this article https://github.com/GandhiCloudLab/ibm-watson-aiops/tree/main/09-push-events-to-kafka-topic.

4. The pushed event is super embossed on the topology as an alert like the below. 

5. There is an alert created in the Alert Viewer. 

4. Pushing Events with _id

By default, _id field is used to map the event to topology resource. When the resources are recreated, then this _id would change, so event also to be modified. To overcome this probelm only the match token concept is available.

Lets see how _id field is used here.

1. Create Event file like the below. Sample file is available here

The f35675ff-be21-4851-a892-3cd1a5f7721b value from this "application":"f35675ff-be21-4851-a892-3cd1a5f7721b" property would be taken and matched with field _id in the topology resources.

2. Push the events to kafka.

3. There is an alert created in the Alert Viewer.