Scan and Introspect vCenter Server

You must scan a vCenter server to discover all available VMs or VMs within a inventory like data centers, folders, clusters, resource pools. You can run the introspection process on the discovered VMs to identify running components.

During vCenter scan, Application Transformer for VMware Tanzu collects the following information using the vim API:

• Datacenter name and moid
• Cluster name and moid
• Folder name, moid and tags
• Resource pool name and moid
• VM list and respective parent moids
• Name, power state, OS type, networks, disks, VMware VMtools version, VMware VMtools installation status, VMware VMtools running status, tags, RAM usage, CPU usage for all VMs.

During introspection of:

• Linux VMs: Application Transformer for VMware Tanzu scans the /proc directory to collect the status of running process and socket information to establish network topology.
  Note: Application Transformer for VMware Tanzu assumes Python installation on all Linux ditsros and uses a Python script to collect these information.
• Windows VMs: Application Transformer for VMware Tanzu runs Get-Process powershell command to get running process information and the netstat command to get socket information. Application Transformer for VMware Tanzu also runs additional commands to collect metadata for components, and make use of respective command line utilities. For example, to get the JAVA version details, it uses java -version.

Prerequisites

• Register vCenter.

• Ensure that you have the required permissions. For more information, see Overview Application Transformer for VMware Tanzu Roles.

• For introspection, you must have the valid credentials associated with the VMs. The VM user account should have the required permissions.

Procedure

1. Click Discovery from the main menu.
2. On the left navigation page, go to Inventory > vCenters.
3. Select a vCenter server from the table.
4. Click Scan and on the Scanning Options page.
   1. Select Apply Credential Policy to apply credential policy on the discovered VMs.
   2. Based on the your requirement, select one of the following:
   • Click No, Scan Only for discovering the VMs
   • Click Yes, Scan and Introspect for discovering the VMs and then discovering components running on those VMs.
     Note: If you scan more than 2000 VMs together, you might encounter the java.lang.OutOfMemoryError. Verify if the atvt_discovery_outOfMemory.hprof file is present at /log/iris/ and share with the VMware Support team.

     To resolve this error, run the following commands and increase the java heap memory space.

     1. vim /home/admin/.config/systemd/user/iris-discovery.service
     2. Change values for parameters -Xms512M -Xmx4096M to Xms1024M -Xmx8192M
     3. systemctl --user daemon-reload
     4. systemctl restart --user iris-discovery
     5. systemctl status --user iris-discovery

What to do next

• View VM details.
• Assign Tags.
• Introspect the VMs.
• Discover Application Topology.