Cloud Services


ChatOps configuration
Published On May 16, 2024 - 1:59 PM

ChatOps configuration

Learn about the IT modernization approach that connects people, processes and automation to help ensure availability of critical IT systems and timely resolution of issues.
ChatOps helps you track your trouble tickets without the need to spend a lot of time doing manual discovery and information collection to address a client situation or issue, which requires hours of data analysis, team collaboration, and communication to resolve the issues.
ChatOps collects the contextually relevant information automatically and presents it in a Slack channel or a conversation within a Microsoft Teams channel, allowing teams to collaborate easily. This helps you reduce the time required to address client situations from hours to minutes by engaging teams and providing them with the appropriate information and insights.

ChatOps available services

ChatOps consists of the following services:
  • AIOps Data Lake:
    Real time data for incident tickets, events, changes, configurations, metrics, and so on flows into the AIOps Data Lake. From there, the data is processed, reconciled, and analyzed in real time, extracting entities and creating entity relationships across data sets. These relationships can be first class or probabilistic inferential linkages. The Data Lake stores all of the raw data, processed data including derived/inferences, computations, and recommendations.
  • AIOps Chatbot:
    The AIOps ChatOps bot retrieves relevant Kyndryl Modern Operations Applications AIOps console tickets, and creates, manages, and archives Slack channels from those tickets, including adding any other users as desired. It manages the full lifecycle of the incident ticket. For more information, see Chatbot incident tracker.
  • ChatOps Framework:
    An API microservices framework that provides atomic services to create, update, and archive channels.
  • Collaboration tool:
    The following collaboration tools are currently supported:
    • Slack
    • Microsoft Teams
  • Sakhi:
    A question and answer widget in the
    Security and Compliance
    dashboard that can answer common questions in a conversational format from users. This service uses the Azure Cognitive Questioning and Answering Service and Azure Cognitive Search Service. Sakhi draws from documents that Kyndryl has uploaded into its knowledge base and is continuously improved by user and administrator input. For more information, see Sakhi widget.
In Microsoft Teams, the number of channels is limited to 200. Therefore, there will be a channel created indicating the severity, and a new conversation in that channel will be generated for each incident ticket with that severity.

Languages supported in ChatOps

ChatOps can function in the following languages:
  • Arabic
  • Chinese
  • Dutch
  • English, United States and United Kingdom
  • French, France and Canada
  • German
  • Gujarati*
  • Hindi*
  • Italian
  • Japanese
  • Korean
  • Marathi*
  • Portuguese, Brazil
  • Spanish, Mexico and Spain
  • Tamil*
  • Telugu*
  • Turkish

Managing the ChatOps knowledge base

In order to train the ChatBot, you need to add files and web pages to the knowledge base. The system will use the supplied information to create question and answer pairs that the ChatBot can use to provide possible solutions to your users. The following tasks are all done from the
ChatBot Knowledge Base Management
page. To reach it, click the
Open menu
icon in the upper left and select
ChatBot Knowledge Base Management

Adding information to the ChatBot knowledge base

To add information to the knowledge base, complete these steps from the
ChatBot Knowledge Base Management
  1. Click
    Add knowledge base documents
  2. On the
    Add knowledge base documents
    page, add any files that you want by either dragging and dropping them into the field or clicking the field and navigate to them. You can add a maximum of ten files at a time. The following formats are supported:
    Maximum file size in MB
  3. Add any web pages that you want by entering the URL in the
    Attach from URL
    field and then clicking
  4. (optional) For each document that you enter, click the
    field and select one or more domains to classify the document under.
  5. Click
  6. The documents are uploaded in the Draft status. To deploy them, click the
    Retrain knowledge base
    icon next to the
    Add knowledge base documents
    button. The documents are now part of the knowledge base that the ChatBot draws on. All documents must be activated at the same time.

Managing knowledge base documents

ChatBot Knowledge Base Management
page also allows you to manage your uploaded documents. All documents are listed. You can find specific documents by using the following filters, all of which are drop-down menus. You can make multiple selections in each category.
  • Select document(s) type:
    Select the type of document that you want to search for.
  • Select domain(s):
    Select one or more domains, which are categories that documents fall into such as
  • Select uploaded by:
    Filter by the person who uploaded the document.
To delete a document so that it is no longer used by the ChatBot, complete these steps:
  1. Click the
    icon next to the document.
  2. Click the
    Retrain knowledge base
    icon next to the
    Add knowledge base documents
    Until you retrain the knowledge base, the deleted document will still be listed and its information used by ChatBot.
Do you have two minutes for a quick survey?
Take Survey