Microsoft AZ-900: Support, Service Lifecycle & Marketplace — Study Guide

By Laurynas Brazys · 22 July 2026 · #microsoft #az-900 #certification #exam-prep #study-guide

Part of the Microsoft Azure AZ-900 — Study Guide. Practice with verified answers in the Microsoft exam hub, or take timed practice tests on ExamRoll.io.

Reliable operations on Azure depend on choosing the right support model, understanding service health signals, designing to published service-level agreements (SLAs), and sourcing third‑party solutions responsibly through the commercial marketplace. This domain focuses on how Microsoft supports cloud workloads throughout their lifecycle and how customers operationalize that support with case management, quota increases, health alerts, and incident handling. Equally important is governance around continuous change. Azure services follow a modern lifecycle with frequent, safe updates and advance notice for breaking changes. The Azure Marketplace extends the platform with validated partner solutions, enabling rapid adoption of security, networking, data, and application components under consolidated Azure billing and policy control.

Azure support plans and case management

Microsoft offers several Azure support plans aligned to workload criticality and the level of advisory guidance required. Basic support is included with every subscription and covers billing and subscription management only. Developer support targets non‑production environments and provides technical guidance during business hours for lower‑severity issues. Standard support is the entry point for 24x7 technical support and is appropriate for most production workloads. Professional Direct (ProDirect) adds faster response targets and access to proactive advisory guidance, architecture reviews, and web seminars to accelerate deployments. Premier/Unified is an enterprise umbrella across Microsoft cloud and on‑premises products, pairing your organization with a Customer Success Account Manager and proactive services such as service reviews, workshops, readiness planning, and incident management coordination. Support requests are created in the Azure portal under Help + support. Case types include technical, billing, subscription management, and service and subscription limits (quotas). Technical cases require a paid plan (Developer, Standard, ProDirect, or Unified). The built‑in Support Request Contributor role lets users open and manage cases without granting broader resource permissions, while Owners and Contributors can also file cases. Cases can be integrated with ITSM tooling via Azure Monitor action groups and the ITSM connector to maintain a single incident queue.

• Basic
  • Technical support availability: No (billing/subscription only)
  • Initial response targets (Sev A/B/C): N/A
  • Advisory guidance: Documentation, community only
  • Proactive services: None
  • Typical use: Trials, learning, non‑critical accounts
• Developer
  • Technical support availability: Business hours (email/web)
  • Initial response targets (Sev A/B/C): N/A / N/A / < 8 hours
  • Advisory guidance: General how‑to guidance
  • Proactive services: None
  • Typical use: Dev/test and pilots
• Standard
  • Technical support availability: 24x7 for Sev A; business hours for lower sev
  • Initial response targets (Sev A/B/C): < 1 hr / < 4 hrs / < 8 hrs
  • Advisory guidance: Best‑effort guidance with cases
  • Proactive services: None
  • Typical use: Production workloads with standard response needs
• Professional Direct
  • Technical support availability: 24x7 for Sev A
  • Initial response targets (Sev A/B/C): < 1 hr / < 2 hrs / < 4 hrs
  • Advisory guidance: Proactive advisory, architecture reviews, onboarding
  • Proactive services: Webinars, service reviews, escalation management
  • Typical use: Business‑critical production needing faster response and guidance
• Premier/Unified
  • Technical support availability: 24x7 (broad Microsoft scope)
  • Initial response targets (Sev A/B/C): < 1 hr / < 2 hrs / < 4 hrs
  • Advisory guidance: Designated CSAM, solution‑level guidance
  • Proactive services: Workshops, assessments, service reviews, incident coordination
  • Typical use: Enterprise portfolios spanning multiple products

Creating support requests and raising quotas and limits

Support requests originate in the Azure portal under Help + support > Create a support request. The workflow prompts for problem type, subscription, service, resource, region, and severity. For technical cases you select an impact level (severity) that aligns with business impact and your plan’s response targets. Include diagnostics (timestamps, correlation IDs, metrics, and change history) to accelerate triage. For billing and subscription management, Basic support is sufficient; for technical issues, a paid plan is required. Quota and limit increases are filed as Service and subscription limits (quotas). Common requests include per‑region compute vCPU quotas (by VM family), public IP addresses, cores for specialized GPU or HPC families, and Azure Machine Learning or Cognitive Services quotas. Compute quotas are enforced per subscription, per region, and per VM‑series family; requesting a higher quota enables capacity requests but does not guarantee capacity availability. For predictable capacity, combine quota increases with on‑demand capacity reservations or plan zonal deployments where appropriate. Access to file requests is controlled through Azure RBAC. The Support Request Contributor role can create and manage tickets without granting resource modification rights. Larger organizations often route quota workflows through internal approval and track them in ITSM; the Azure Quota API can automate monitoring and raising requests at predefined thresholds. After submission, cases are triaged to engineering when needed, and communications occur through the portal, email, and optionally phone for higher severities.

Service Health, planned maintenance, and SLAs in incident management

Azure provides multiple health lenses. Azure Status is the public, global view of platform‑wide issues. Azure Service Health is personalized for your subscriptions, showing service issues, planned maintenance, and health advisories affecting your regions and resources. Resource Health drills into the state of individual resources, providing recent availability events and root causes such as platform events, user‑initiated actions, or underlying hardware faults. Service Health alerts integrate with action groups for email, SMS, voice, webhooks, ITSM connectors, and Azure Functions. Planned maintenance is communicated via Service Health with expected windows and impact details. For virtual machines, Azure Scheduled Events expose in‑guest notifications (via the instance metadata service) of upcoming actions like reboots, allowing applications to drain, checkpoint, or fail over. Maintenance Control is available for Azure Dedicated Hosts and certain isolated VM sizes, enabling administrators to delay platform maintenance within a defined deferral window to align with change control processes. SLAs define the monthly uptime or transaction success targets for each service. When architecting multi‑service solutions, combined availability is multiplicative across dependencies. Higher availability tiers typically require zone‑redundant or multi‑instance deployments. For example, running virtual machines across two or more availability zones in a region achieves a higher SLA than a single VM or an availability set. If an SLA is not met, service credits are available by claim; submit through the portal with evidence (timestamps, affected resources) within the required window, and track the incident and post‑incident reviews in Service Health.

• Virtual Machines (single instance)
  • Key requirement for SLA: Premium SSD or Ultra Disk for OS and data
  • Published SLA (typical): 99.9%
  • Notes: Applies to eligible sizes and storage; no availability set/zone
• Virtual Machines in Availability Set
  • Key requirement for SLA: Two or more instances across fault/update domains
  • Published SLA (typical): 99.95%
  • Notes: Reduces single‑rack and update domain risk
• Virtual Machines across Availability Zones
  • Key requirement for SLA: Two or more instances across zones in one region
  • Published SLA (typical): 99.99%
  • Notes: Best regional resiliency without cross‑region failover
• Azure SQL Database (single database)
  • Key requirement for SLA: Standard deployment
  • Published SLA (typical): 99.99%
  • Notes: Zone redundancy can improve resiliency to zonal faults
• App Service (multi‑instance)
  • Key requirement for SLA: Two or more instances
  • Published SLA (typical): 99.95%
  • Notes: Requires Basic or higher tiers with multiple instances

Modern Lifecycle Policy and the Azure Marketplace

Azure services follow the Modern Lifecycle Policy, which emphasizes continuous service updates and customer responsibility to stay current. Breaking changes that require customer action are communicated with advance notice—typically at least 12 months for online services—through formal deprecation announcements, Service Health advisories, and documentation updates. Retirement timelines, API versioning, and migration paths are provided so workloads can be remediated within change windows that align to enterprise governance. This continuous delivery model leverages safe deployment practices with staged rollouts and automatic mitigation. Operational readiness means tracking service advisories, validating new runtime or SDK versions in pre‑production, and using feature flags or blue‑green patterns for application updates. Security and compliance updates arrive without separate installation cycles, reducing exposure windows while placing emphasis on observability and release discipline. The Azure Marketplace extends the platform with third‑party solutions: virtual machine images, managed applications, Kubernetes apps, and SaaS offerings. Offers are billed through your Azure subscription with options such as pay‑as‑you‑go metering, annual/term plans, and bring‑your‑own‑license (BYOL). Enterprises can curate a private marketplace to allow only approved publishers or offers, apply Azure Policy to constrain deployment footprint (regions, SKUs, networking), and negotiate custom pricing through private offers. Network virtual appliances (e.g., firewalls, load balancers), data platforms, security analytics, and observability stacks are common patterns; publisher support is the primary channel for these solutions, with billing consolidated in Azure.

• Virtual machine image (NVA or server)
  • Deployment model: ARM template/VM in your VNet
  • Billing model: PAYG or BYOL
  • Example use cases: Firewalls, WAFs, IDS/IPS, packaged apps
  • Operational notes: You manage VM lifecycle, scaling, and patching
• Managed Application
  • Deployment model: Deployed to your subscription, managed by publisher
  • Billing model: PAYG plus managed fee
  • Example use cases: Turnkey solutions with managed operations
  • Operational notes: Publisher updates core components; you manage data/config
• Kubernetes app (AKS add‑on)
  • Deployment model: Helm/ARM into your AKS cluster
  • Billing model: PAYG or BYOL
  • Example use cases: Ingress controllers, service meshes, operators
  • Operational notes: You manage AKS; publisher maintains app container images
• SaaS
  • Deployment model: Runs in publisher’s tenant; integrates with your tenant
  • Billing model: Subscription/consumption
  • Example use cases: APIs, analytics platforms, security SaaS
  • Operational notes: Fastest time‑to‑value; minimal infra to manage

Practical Problem: Coastal Outfitters: Operationalizing support, health, and marketplace solutions for a seasonal e‑commerce peak

Scenario: Coastal Outfitters runs an Azure‑hosted e‑commerce site with steady traffic most of the month and a four‑day surge at month‑end. The architecture uses Azure Application Gateway, two VM‑based web tiers, and a managed database. Security requires a next‑generation firewall, and operations must guarantee 99.99% availability for the web tier during surge periods while receiving proactive guidance during rollout.

Challenge: Choose the right support plan for proactive architecture guidance, deploy a marketplace firewall under governance, raise regional compute quotas ahead of the surge, and implement health alerts and incident workflows that align to 99.99% availability targets.

Recommended Approach:

1. Upgrade the subscription to the Professional Direct support plan to secure 24x7 critical response and proactive advisory guidance for the production rollout.
2. Assign the built‑in Support Request Contributor role to the NOC group so they can open and manage tickets without broad resource permissions.
3. Deploy a Palo Alto Networks or Fortinet next‑generation firewall from Azure Marketplace into a dedicated hub VNet; leverage a private marketplace so only approved, negotiated private offers are visible to subscription owners.
4. Place two or more web VMs in separate availability zones behind Application Gateway to meet the 99.99% VM SLA; store OS/data on Premium SSDs.
5. File a Service and subscription limits (quotas) request two weeks before the surge to increase per‑region vCPU quotas for the required VM families; add capacity reservations for the web tier to guarantee burst capacity.
6. Configure Azure Service Health alerts for service issues, planned maintenance, and health advisories in the site’s regions; route alerts to an action group that notifies on‑call staff via email/SMS and opens an incident in ServiceNow via the ITSM connector.
7. Enable Azure Scheduled Events handling on the web VMs to gracefully drain connections during host maintenance; validate failover behavior in staging using blue‑green swaps.
8. Document the SLA credit claim workflow in the runbook, including evidence capture (timestamps, resource IDs, metrics), and ensure claims are filed through the portal within the allowed window if availability targets are not met.

Azure Rationale: Professional Direct combines fast response with proactive advisory benefits without the overhead of an enterprise‑wide Unified agreement, aligning to a single critical production workload. Zonal VMs raise the availability tier to 99.99% with minimal instance count, while Premium SSDs meet single‑instance storage requirements. A curated private marketplace and negotiated private offers keep security appliances compliant and cost‑controlled, and publisher support remains the first line for NVA issues. Early quota increases and capacity reservations eliminate deployment friction during the surge. Service Health alerts, scheduled event handling, and an SLA claim runbook complete the incident management loop with clear responsibilities and evidence paths.

← Monitoring · All domains

Practice these questions → · Timed practice on ExamRoll.io →

Browse Microsoft →