This ticket is for documenting the new counter metric, mtv_migration_plan_vms_total, which will be used to gather and forward telemetry data about Virtual Machine (VM) migrations facilitated by the Migration Toolkit for Virtualization (MTV).

The goal of this metric is to provide a concise way of counting the total number of VM migrations on the cluster based on their key attributes, which currently relies on unreliable customer input.

Metric to Document

 

Label Details and Possible Values

The documentation should clearly define each label and list the possible values it can take.

• vm_status

• Description: The final state of the VM after the migration attempt.

• Possible Values:

• • Succeeded

• • Failed

• provider

• • Description: The source virtualization platform the VM was migrated from.

• • Possible Values (Current):

• • • vsphere

• • • ova

• • • Other currently supported providers (e.g., OpenStack, oVirt, OpenShift) should be included.

• Future Values (Note for Inclusion):

• • awsec2 (Not yet implemented, but planned)

• mode

• • Description: The type of migration performed.

• • Possible Values (Current):

• • • Cold: The source VM is shut down while data is copied.

• • • Warm: Most data is copied while the VM is running (pre-copy stage), and the VM is shut down only for the cutover stage.

• • Future Values (Note for Inclusion):

• • • RCM (Not yet implemented, but planned)

• target

• • Description: The destination cluster for the migration.

• • Possible Values:

• • • local

• • • remote

Example Use Cases

Please include clear examples to illustrate how this counter aggregates data.

• Example 1: Successful Warm Migrations

• • If the metric reports: succeeded, vsphere, warm, local = 2

• • This means: 2 VMs were successfully migrated using warm migration from a vSphere source provider to a local cluster.

• Example 2: Failed Cold Migrations

• • If the metric reports: failed, vsphere, cold, local = 3

• • This means: 3 VMs failed cold migration from a vSphere source provider to a local cluster.

Content Journey for MTV Telemetry Metrics Documentation

 

Jobs to be Done" (JTBD) statement:

 

"When I am running VM migrations with MTV, I want to automatically and reliably gather comprehensive data on the status, type, and outcomes of the migrations and individual VMs, so that I can report on usage, identify trends, and address failures without relying on manual customer input."

Breakdown

• Job (What the user wants to accomplish): Gather comprehensive data on migration status, type, and outcomes.
   
• Circumstance (The situation or context): When running VM migrations with MTV.
   
• Motivation/Need (Why they want it): To report on usage, identify trends, and address failures.
   
• Desired Outcome (What success looks like): Accessing this data automatically and reliably, without relying on manual customer input.

Personas:

1. Product Manager / Business Analyst

• As a Product Manager, I want to perform trend analysis on migration success rates, provider usage, and popular OSes , to achieve the goal of understanding MTV adoption, identifying pain points, and making data-driven decisions for feature prioritization.

2. Reliability Engineer (SRE) / Operations Specialist

• As a Reliability Engineer, I want to perform real-time monitoring of migration status, duration, and failures by plan and VM, to achieve the goal of proactively detecting and alerting on widespread issues, ensuring migration stability, and measuring the Mean Time To Recovery (MTTR).

3. Support Engineer / QE Specialist

• As a Support Engineer, I want to perform detailed lookups of metrics (like data transferred and duration) associated with a specific failed migration plan ID , to achieve the goal of quickly diagnosing specific customer issues and verifying the root cause of migration failures or cancellations.