Deep discovery for Service Graph Connector for GCP

  • Release version: Washingtondc
  • Updated July 24, 2025
  • 2 minutes to read
    • Summarize
    Summarized using AI
    This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.

    Summary of Deep Discovery for Service Graph Connector for GCP

    Deep discovery for the Service Graph Connector for Google Cloud Platform (GCP) allows customers to fetch and populate server data into the Configuration Management Database (CMDB). This process involves executing patch jobs at the GCP project level to retrieve critical information about virtual machines (VMs), including host names, serial numbers, CPU data, TCP connections, and running processes.

    Show full answer Show less

    Key Features

    • Patch Job Execution: The connector utilizes OS Config Patch jobs to run deep discovery commands independently on both Linux and Windows VMs.
    • Scheduled Imports: Two scheduled imports facilitate the process:
      • SG-GCP Execute Patch Job: Executes the patch job commands based on the specified shell and PowerShell scripts stored in Cloud Storage.
      • SG-GCP Get Patch Job: Reads the output of the commands executed on the VMs, processes the data, and deletes the output files from Cloud Storage.
    • Connection Properties: Essential connection properties are defined, including bucket names for scripts and outputs, paths for storage, and versioning numbers for scripts.

    Key Outcomes

    By implementing deep discovery, customers can expect:

    • Detailed and accurate VM data populated in the CMDB, enhancing configuration visibility.
    • Improved Application Dependency Mapping (ADM) capabilities, allowing for the classification of various applications such as Apache, MySQL, and PostgreSQL instances based on the imported middleware information.
    • Streamlined management of server configurations and dependencies, leading to better resource allocation and operational efficiency.

    The SG-GCP Execute Patch Job and SG-GCP Get Patch Job scheduled import jobs enable deep discovery by running patch jobs at the GCP project level.

    Deep discovery involves fetching server data. The deep discovery data populates host name, serial number, CPU data, TCP connections, and running processes data into the CMDB.

    The Service Graph Connector for GCP uses OS Config Patch jobs on GCP projects to execute the deep discovery commands on the VMs. These patch jobs run independently on Linux and Windows VMs.

    The following scheduled imports execute the OS Config Patch jobs, and then read and delete the output:

    Deep discovery connection properties

    Table 1. Deep discovery connection properties
    Connection property name Connection property value
    Scripts Bucket Name Name of the Cloud Storage Bucket where the deep discovery scripts are saved.
    Outputs Bucket Name Name of the output Cloud Storage Bucket.
    Outputs Storage Path Path of the output Cloud Storage Bucket.
    Shell Script Object Path Path of the deep discovery Shell script file.
    Shell Script Generation Number Number that is used to identify the versioning of the cloud storage object.
    PowerShell Script Object Path Path of the deep discovery PowerShell script file.
    PowerShell Script Generation Number Number that is used to identify the versioning of the cloud storage object.

    SG-GCP Execute Patch Job

    The SG-GCP Execute Patch Job scheduled import reads the path of the Shell and PowerShell scripts uploaded to the Cloud Storage Bucket and the generation numbers from the deep discovery connection properties. The scheduled import then runs the commands on Linux and Windows VMs separately.

    In the Import Set table for the SG-GCP Execute Patch Job import schedule, each project will have two rows, one for Linux VMs and the other for Windows VMs. The Import Set rows remain in Pending state because no transformations are required for these rows. The Import Set rows include the patch job ID and the list of instances on which the deep discovery script is run. The output is stored in the Cloud Storage Bucket with the Object ID of the VMs on which the deep discovery commands are run.

    SG-GCP Get Patch Job

    The SG-GCP Get Patch Job scheduled import reads the output of the commands that are run on the Linux and Windows VMs. The scheduled import reads the output from the output bucket and folder path that are specified in the deep discovery connection properties. The SG-GCP Get Patch Job scheduled import then deletes the output from the Cloud Storage Bucket. The Import Set rows are processed using Application Dependency Mapping (ADM) to populate the Server records with the data of the VMs on which deep discovery was run. The Import Set rows include VM data such as running processes, TCP connections, and CPU information.

    Application Dependency Mapping

    Using the deep discovery information, Application Dependency Mapping (ADM) creates and classifies the following applications from the imported middleware information:
    • Apache Web Server
    • MongoDB Instance
    • MSFT SQL Instance
    • MySQL Instance
    • Nginx Web Server
    • Oracle Database Listener
    • Oracle Instance
    • PostgreSQL Instance
    • Tomcat

    Additional resource

    Service Graph Connector for GCP - Deep Discovery Setup Instructions [KB2213095] article in the Now Support Knowledge Base