Inventory management is a core feature of kubectl-mtv that allows you to discover, explore, and query resources from various virtualization providers. This chapter covers comprehensive inventory operations for migration planning and resource discovery.

Overview of Resources Available for Querying

kubectl-mtv provides access to a comprehensive inventory of resources across different provider types:

General Resources (All Providers)

Resource Aliases Description
vm vms Virtual machines and instances
host hosts Physical hosts and hypervisors
network networks Network configurations and segments
storage storages Storage systems and configurations
datacenter datacenters Data center structures
cluster clusters Compute clusters
disk disks Individual disk resources
namespace namespaces Logical resource groupings

Profile Resources (oVirt/RHV)

Resource Aliases Description
diskprofile diskprofiles, disk-profiles Disk performance profiles
nicprofile nicprofiles, nic-profiles Network interface profiles

OpenStack-Specific Resources

Resource Aliases Description
instance instances OpenStack compute instances
image images VM images and templates
flavor flavors Instance sizing configurations
project projects OpenStack projects/tenants
volume volumes Block storage volumes
volumetype volumetypes, volume-types Volume type definitions
snapshot snapshots Volume and instance snapshots
subnet subnets Network subnets

vSphere-Specific Resources

Resource Aliases Description
datastore datastores vSphere storage datastores
resourcepool resourcepools, resource-pools vSphere resource pools
folder folders vSphere organizational folders

Kubernetes-Specific Resources

Resource Aliases Description
pvc pvcs, persistentvolumeclaims Persistent Volume Claims
datavolume datavolumes, data-volumes KubeVirt DataVolumes

EC2-Specific Resources

Resource Aliases Description
ec2-instance ec2-instances EC2 compute instances
ec2-volume ec2-volumes EBS volumes
ec2-network ec2-networks VPCs and subnets
ec2-volume-type ec2-volume-types EBS volume types

Note: Generic resources (vms, networks, storage) also work with EC2 providers and display EC2-specific information.

Provider Resources

Resource Aliases Description
provider providers Provider inventory and status

General Syntax

The general syntax for inventory commands follows this pattern:

1

kubectl mtv get inventory <resource> --provider <provider> [flags]

Basic Command Structure

  • Resource: The type of resource to query (vm, network, storage, etc.)
  • Provider: The name of the provider to query
  • Flags: Optional parameters for filtering, output format, etc.

Common Flags

  • -o, --output: Output format (table, json, yaml, planvms for VMs)
  • -q, --query: Query filter using Tree Search Language (TSL)
  • -w, --watch: Watch for real-time changes
  • --extended: Show extended information (where supported)
  • --inventory-url: Custom inventory service URL

Common Inventory Examples

Virtual Machines

VMs are the most commonly queried inventory resource for migration planning:

1
2
3
4
5
6
7
8
9
10
11
12
13

# List all VMs from a vSphere provider
kubectl mtv get inventory vms --provider vsphere-prod

# List VMs from different provider types
kubectl mtv get inventory vms --provider ovirt-prod
kubectl mtv get inventory instances --provider openstack-prod
kubectl mtv get inventory vms --provider openshift-source

# List VMs with extended information
kubectl mtv get inventory vms --provider vsphere-prod --extended

# Watch for VM changes in real-time
kubectl mtv get inventory vms --provider vsphere-prod --watch

Networks

Network inventory helps plan network mappings:

1
2
3
4
5
6
7
8
9
10
11

# List all networks from vSphere
kubectl mtv get inventory networks --provider vsphere-prod

# List networks from oVirt
kubectl mtv get inventory networks --provider ovirt-prod

# List subnets from OpenStack
kubectl mtv get inventory subnets --provider openstack-prod

# View network details in YAML format
kubectl mtv get inventory networks --provider vsphere-prod --output yaml

Storage

Storage inventory assists with storage mapping configuration:

1
2
3
4
5
6
7
8
9
10
11

# List storage from vSphere (datastores)
kubectl mtv get inventory datastores --provider vsphere-prod

# List storage from oVirt
kubectl mtv get inventory storages --provider ovirt-prod

# List volume types from OpenStack
kubectl mtv get inventory volumetypes --provider openstack-prod

# View storage details in JSON format
kubectl mtv get inventory datastores --provider vsphere-prod --output json

Hosts and Infrastructure

Discover infrastructure layout for planning:

1
2
3
4
5
6
7
8
9
10
11
12
13
14

# List ESXi hosts in vSphere
kubectl mtv get inventory hosts --provider vsphere-prod

# List oVirt hosts
kubectl mtv get inventory hosts --provider ovirt-prod

# List datacenters
kubectl mtv get inventory datacenters --provider vsphere-prod

# List resource pools (vSphere)
kubectl mtv get inventory resource-pools --provider vsphere-prod

# List clusters
kubectl mtv get inventory clusters --provider vsphere-prod

Provider Status

Check provider health and connectivity:

1
2
3
4
5
6
7
8

# List inventory from all providers
kubectl mtv get inventory providers

# Get detailed inventory from specific provider
kubectl mtv get inventory providers --name vsphere-prod

# Monitor provider status
kubectl mtv get inventory providers --name vsphere-prod --watch

Output Formats

Table Format (Default)

The default table format provides a concise overview:

1
2
3
4
5
6
7
8

# Default table output
kubectl mtv get inventory vms --provider vsphere-prod

# Example output:
# NAME          POWER STATE    MEMORY(MB)    CPU    DISKS
# web-server-01 poweredOn      4096          2      2
# db-server-01  poweredOn      8192          4      3
# test-vm-01    poweredOff     2048          1      1

JSON Format

JSON output provides complete data for automation:

1
2
3
4
5
6
7
8
9

# JSON output for scripting
kubectl mtv get inventory vms --provider vsphere-prod --output json

# Extract specific fields using jq
kubectl mtv get inventory vms --provider vsphere-prod --output json | jq '.items[].name'

# Complex data extraction
kubectl mtv get inventory vms --provider vsphere-prod --output json | \
  jq '.items[] | select(.powerState == "poweredOn") | .name'

YAML Format

YAML format for human-readable structured data:

1
2
3
4
5

# YAML output
kubectl mtv get inventory vms --provider vsphere-prod --output yaml

# Save to file for analysis
kubectl mtv get inventory vms --provider vsphere-prod --output yaml > vms-inventory.yaml

Extended Output

Extended output provides additional details where supported:

1
2
3
4
5
6
7
8

# Extended information for VMs
kubectl mtv get inventory vms --provider vsphere-prod --extended

# Extended output shows additional fields like:
# - Guest OS information
# - IP addresses
# - Detailed disk information
# - Network adapter details

How-To: Exporting VMs for Migration Planning

The special planvms output format is designed specifically for migration planning:

PlanVMs Output Format 

1
2
3
4
5

# Export VMs in planvms format
kubectl mtv get inventory vms --provider vsphere-prod --output planvms

# Save to file for migration plan creation
kubectl mtv get inventory vms --provider vsphere-prod --output planvms > migration-vms.yaml

Using PlanVMs with Migration Plans 

1
2
3
4
5
6
7
8
9
10
11
12
13

# Create migration plan using exported VMs
kubectl mtv create plan --name production-migration \
  --source vsphere-prod \
  --vms @migration-vms.yaml

# Or use the planvms format directly in plan creation
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where name ~= 'prod-.*' and powerState = 'poweredOn'" \
  --output planvms > prod-vms.yaml

kubectl mtv create plan --name prod-migration \
  --source vsphere-prod \
  --vms @prod-vms.yaml

PlanVMs Format Structure

The planvms format provides a structured list suitable for plan creation:

1
2
3
4
5
6
7
8
9
10
11

# Example planvms output
apiVersion: v1
kind: List
items:
- name: web-server-01
  targetName: ""  # Can be customized
  rootDisk: ""    # Can be specified
- name: db-server-01
  targetName: ""
  rootDisk: ""
# Additional VMs...

Advanced Inventory Querying

Query-Based VM Discovery

Use the Tree Search Language (TSL) for sophisticated filtering:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

# Find powered-on VMs with high memory
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where powerState = 'poweredOn' and memoryMB > 8192"

# Find VMs matching name patterns
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where name ~= 'web-.*' or name ~= 'app-.*'"

# Find VMs with multiple disks
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where len(disks) > 1"

# Complex queries with multiple conditions
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where powerState = 'poweredOn' and memoryMB >= 4096 and name ~= 'prod-.*'"

Sorting and Limiting Results

Use order by and limit to control how many results are returned and in what order:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

# Top 10 largest VMs by memory (sorted descending, limited to 10)
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where powerState = 'poweredOn' order by memoryMB desc limit 10"

# First 50 powered-on VMs alphabetically
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where powerState = 'poweredOn' order by name limit 50"

# Smallest VMs first (candidates for quick migration)
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where powerState = 'poweredOn' order by memoryMB asc limit 20"

# Datastores with >100 GB free (use free in bytes with SI units)
kubectl mtv get inventory datastores --provider vsphere-prod \
  --query "where free > 100Gi order by free asc limit 5"

Selecting Specific Fields

Use the select clause to return only the fields you need, reducing output size:

1
2
3
4
5
6
7
8
9
10
11

# Return only name, memory, and CPU (compact output)
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "select name, memoryMB, cpuCount where powerState = 'poweredOn' limit 10"

# Select with aliases and combined with order/limit
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "select name, memoryMB as mem, cpuCount where memoryMB > 4096 order by memoryMB desc limit 5"

# Sum of disk capacity with select
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "select name, sum(disks[*].capacityGB) as totalDisk where powerState = 'poweredOn' order by totalDisk desc limit 10"

Provider-Specific Resource Queries

vSphere Resource Discovery 

1
2
3
4
5
6
7
8
9
10

# Find datastores with available space
kubectl mtv get inventory datastores --provider vsphere-prod \
  --query "where freeSpaceGB > 100"

# List resource pools by availability
kubectl mtv get inventory resource-pools --provider vsphere-prod

# Find hosts in specific clusters
kubectl mtv get inventory hosts --provider vsphere-prod \
  --query "where cluster.name = 'Production-Cluster'"

oVirt Resource Discovery 

1
2
3
4
5
6
7
8
9
10

# Find VMs with specific OS types
kubectl mtv get inventory vms --provider ovirt-prod \
  --query "where guestOS ~= 'rhel.*'"

# List disk profiles
kubectl mtv get inventory disk-profiles --provider ovirt-prod

# Find storage domains
kubectl mtv get inventory storages --provider ovirt-prod \
  --query "where type = 'data'"

OpenStack Resource Discovery 

1
2
3
4
5
6
7
8
9
10
11

# Find instances by flavor
kubectl mtv get inventory instances --provider openstack-prod \
  --query "where flavor.name = 'm1.large'"

# List available images
kubectl mtv get inventory images --provider openstack-prod \
  --query "where status = 'active'"

# Find volumes by size
kubectl mtv get inventory volumes --provider openstack-prod \
  --query "where size >= 100"

Inventory Automation and Scripting

Automated Resource Discovery 

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

#!/bin/bash
# Script to discover migration candidates

PROVIDER="vsphere-prod"
OUTPUT_DIR="inventory-$(date +%Y%m%d)"

mkdir -p "$OUTPUT_DIR"

# Export all VMs
kubectl mtv get inventory vms --provider "$PROVIDER" --output json > "$OUTPUT_DIR/all-vms.json"

# Export powered-on VMs only
kubectl mtv get inventory vms --provider "$PROVIDER" \
  --query "where powerState = 'poweredOn'" \
  --output planvms > "$OUTPUT_DIR/active-vms.yaml"

# Export large VMs (>50GB total disk)
kubectl mtv get inventory vms --provider "$PROVIDER" \
  --query "where sum(disks[*].capacityGB) > 50" \
  --output json > "$OUTPUT_DIR/large-vms.json"

# Export network information
kubectl mtv get inventory networks --provider "$PROVIDER" --output yaml > "$OUTPUT_DIR/networks.yaml"

# Export storage information
kubectl mtv get inventory datastores --provider "$PROVIDER" --output yaml > "$OUTPUT_DIR/datastores.yaml"

echo "Inventory exported to $OUTPUT_DIR/"

Migration Planning Automation 

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

#!/bin/bash
# Automated migration plan generation

PROVIDER="vsphere-prod"
PLAN_NAME="auto-migration-$(date +%Y%m%d)"

# Discover migration candidates
kubectl mtv get inventory vms --provider "$PROVIDER" \
  --query "where powerState = 'poweredOn' and memoryMB <= 8192 and len(disks) <= 2" \
  --output planvms > "small-vms.yaml"

# Create migration plan
kubectl mtv create plan --name "$PLAN_NAME" \
  --source "$PROVIDER" \
  --vms @small-vms.yaml \
  --migration-type cold

echo "Created migration plan: $PLAN_NAME"

Provider-Specific Inventory Features

vSphere Provider Inventory

vSphere provides the richest inventory with hierarchical structure:

1
2
3
4
5
6
7
8
9

# Complete vSphere infrastructure discovery
kubectl mtv get inventory datacenters --provider vsphere-prod
kubectl mtv get inventory clusters --provider vsphere-prod
kubectl mtv get inventory hosts --provider vsphere-prod
kubectl mtv get inventory resource-pools --provider vsphere-prod
kubectl mtv get inventory folders --provider vsphere-prod
kubectl mtv get inventory datastores --provider vsphere-prod
kubectl mtv get inventory networks --provider vsphere-prod
kubectl mtv get inventory vms --provider vsphere-prod

oVirt Provider Inventory

oVirt provides enterprise virtualization resources:

1
2
3
4
5
6
7

# oVirt-specific resources
kubectl mtv get inventory vms --provider ovirt-prod
kubectl mtv get inventory hosts --provider ovirt-prod
kubectl mtv get inventory storages --provider ovirt-prod
kubectl mtv get inventory networks --provider ovirt-prod
kubectl mtv get inventory disk-profiles --provider ovirt-prod
kubectl mtv get inventory nic-profiles --provider ovirt-prod

OpenStack Provider Inventory

OpenStack provides cloud-native resource discovery:

1
2
3
4
5
6
7
8
9

# OpenStack-specific resources
kubectl mtv get inventory instances --provider openstack-prod
kubectl mtv get inventory images --provider openstack-prod
kubectl mtv get inventory flavors --provider openstack-prod
kubectl mtv get inventory projects --provider openstack-prod
kubectl mtv get inventory volumes --provider openstack-prod
kubectl mtv get inventory volumetypes --provider openstack-prod
kubectl mtv get inventory snapshots --provider openstack-prod
kubectl mtv get inventory subnets --provider openstack-prod

OpenShift/KubeVirt Provider Inventory

For KubeVirt-to-KubeVirt migrations:

1
2
3
4
5

# KubeVirt-specific resources
kubectl mtv get inventory vms --provider openshift-source
kubectl mtv get inventory pvcs --provider openshift-source
kubectl mtv get inventory data-volumes --provider openshift-source
kubectl mtv get inventory namespaces --provider openshift-source

Real-Time Monitoring and Watching

Watch Mode

Monitor inventory changes in real-time:

1
2
3
4
5
6
7
8

# Watch VM state changes
kubectl mtv get inventory vms --provider vsphere-prod --watch

# Watch provider status
kubectl mtv get inventory providers --name vsphere-prod --watch

# Monitor network changes
kubectl mtv get inventory networks --provider vsphere-prod --watch

Automated Monitoring 

1
2
3
4
5
6
7
8
9
10
11
12
13

#!/bin/bash
# Monitor for migration readiness

PROVIDER="vsphere-prod"

echo "Monitoring VMs for migration readiness..."

kubectl mtv get inventory vms --provider "$PROVIDER" \
  --query "where powerState = 'poweredOn'" \
  --watch | while read -r line; do
    echo "$(date): $line"
    # Add notification logic here
done

Inventory Performance and Optimization

Large Environment Optimization

For large virtualization environments:

1
2
3
4
5
6
7
8
9
10
11
12
13

# Use specific queries to reduce data transfer
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where datacenter.name = 'DC-East'" \
  --output json

# Focus on specific clusters
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where cluster.name = 'Production-Cluster'"

# Limit results with targeted queries (order by + limit)
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where memoryMB >= 4096 and powerState = 'poweredOn' order by memoryMB desc limit 25" \
  --extended

Inventory Caching 

1
2
3
4
5
6
7

# Cache frequently accessed inventory
kubectl mtv get inventory vms --provider vsphere-prod --output json > vms-cache.json
kubectl mtv get inventory networks --provider vsphere-prod --output yaml > networks-cache.yaml
kubectl mtv get inventory datastores --provider vsphere-prod --output yaml > storage-cache.yaml

# Use cached data for planning
jq '.items[] | select(.powerState == "poweredOn")' vms-cache.json

Troubleshooting Inventory Issues

Common Inventory Problems

Provider Connectivity Issues 

1
2
3
4
5
6
7
8

# Check provider status
kubectl mtv describe provider --name vsphere-prod

# Test inventory service connectivity
kubectl mtv get inventory providers --name vsphere-prod

# Check inventory service URL
echo $MTV_INVENTORY_URL

Query Syntax Errors 

1
2
3
4
5
6
7
8

# Test queries with simple examples first
kubectl mtv get inventory vms --provider vsphere-prod --query "where name = 'test-vm'"

# Use JSON output to understand available fields
kubectl mtv get inventory vms --provider vsphere-prod --output json | jq '.items[0]' | head -20

# Check query syntax documentation
kubectl mtv get inventory vms --provider vsphere-prod --help

Performance Issues 

1
2
3
4
5
6
7
8
9

# Use specific queries instead of listing all resources
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where cluster.name = 'specific-cluster'"

# Enable debug logging
kubectl mtv get inventory vms --provider vsphere-prod -v=2

# Check inventory service performance
kubectl get pods -n konveyor-forklift -l app=forklift-inventory

Debug and Logging 

1
2
3
4
5
6
7
8

# Enable verbose logging
kubectl mtv get inventory vms --provider vsphere-prod -v=2

# Check inventory service logs
kubectl logs -n konveyor-forklift deployment/forklift-inventory -f

# Monitor network traffic
kubectl exec -it forklift-inventory-pod -- netstat -an

Integration with Migration Workflows

Inventory-Driven Migration Planning 

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

# 1. Discover migration candidates
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where powerState = 'poweredOn' and memoryMB <= 16384" \
  --output planvms > candidates.yaml

# 2. Analyze resource requirements
kubectl mtv get inventory vms --provider vsphere-prod \
  --query "where name in ['vm1', 'vm2', 'vm3']" \
  --output json | jq '.items[] | {name, memory: .memoryMB, disks: [.disks[].capacityGB]}'

# 3. Check network and storage availability
kubectl mtv get inventory networks --provider vsphere-prod --output yaml
kubectl mtv get inventory datastores --provider vsphere-prod \
  --query "where freeSpaceGB > 100" --output yaml

# 4. Create migration plan
kubectl mtv create plan --name inventory-driven-migration \
  --source vsphere-prod \
  --vms @candidates.yaml

Validation and Verification 

1
2
3
4
5
6
7
8
9
10

# Verify VMs exist before migration
VM_NAMES=$(kubectl get plan migration-plan -o json | jq -r '.spec.vms[].name')
for vm in $VM_NAMES; do
  kubectl mtv get inventory vms --provider vsphere-prod --query "where name = '$vm'" --output json | \
    jq -r '.items[0].name // "NOT FOUND: '$vm'"'
done

# Check resource availability
kubectl mtv get inventory datastores --provider vsphere-prod \
  --query "where freeSpaceGB > 500" --output table

Next Steps

After mastering inventory management:

  1. Learn Query Language: Master advanced filtering in Chapter 10: Query Language Reference and Advanced Filtering
  2. Create Mappings: Define resource mappings in Chapter 11: Mapping Management
  3. Plan Migrations: Use inventory data for planning in Chapter 13: Migration Plan Creation
  4. Optimize Performance: Apply inventory insights in Chapter 16: Migration Process Optimization

Previous: Chapter 8: VDDK Image Creation and Configuration
Next: Chapter 10: Query Language Reference and Advanced Filtering