{"id":59602,"date":"2026-11-06T13:01:58","date_gmt":"2026-11-06T03:01:58","guid":{"rendered":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/11\/06\/how-azure-site-recovery-for-vmss-flex-changes-your-dr-runbook\/"},"modified":"2026-11-06T13:03:42","modified_gmt":"2026-11-06T03:03:42","slug":"how-azure-site-recovery-for-vmss-flex-changes-your-dr-runbook","status":"publish","type":"post","link":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/11\/06\/how-azure-site-recovery-for-vmss-flex-changes-your-dr-runbook\/","title":{"rendered":"How Azure Site Recovery for VMSS Flex Changes Your DR Runbook"},"content":{"rendered":"\n

In this blog post How Azure Site Recovery for VMSS Flex Changes Your DR Runbook<\/strong> we will explain why an older disaster recovery plan may no longer match the way your Azure environment operates. If your runbook simply says \u201cfail over the scale set\u201d, it may hide gaps that only become visible during an outage.<\/p>\n\n\n\n\n\n\n\n

At a high level, Azure Site Recovery copies protected virtual machines to a secondary Azure location and coordinates their recovery when the primary location is unavailable. Recent support for Virtual Machine Scale Sets with Flexible orchestration, commonly called VMSS Flex, improves this process\u2014but each virtual machine still needs to be treated as an individual recovery item.<\/p>\n\n\n\n

What VMSS Flex and Azure Site Recovery actually do<\/h2>\n\n\n\n

VMSS Flex is a way to manage a group of Azure virtual machines that can increase or decrease in number as demand changes. Unlike older scale-set models, Flexible orchestration gives your team more control over each machine, including its disks, network settings and lifecycle.<\/p>\n\n\n\n

Azure Site Recovery is Microsoft\u2019s disaster recovery service. It continuously sends disk changes from a protected machine through a temporary cache and into replicated storage at the recovery location. It also creates recovery points, allowing the machine to be restarted from a recent copy during an outage.<\/p>\n\n\n\n

For VMSS Flex, Site Recovery protects each virtual machine separately. Before enabling replication, you create a target Flexible scale set in the recovery region or availability zone. During failover, Site Recovery creates the recovered machine and places it into that target scale set.<\/p>\n\n\n\n

Microsoft\u2019s October 2026 update made the end-to-end workflow generally available, including improved target configuration, failover, reprotection and failback. That removes several earlier limitations, but it does not make disaster recovery automatic at the scale-set level.<\/p>\n\n\n\n

Why your existing runbook probably needs rebuilding<\/h2>\n\n\n\n

1. Scale-set membership does not mean every machine is protected<\/h3>\n\n\n\n

Site Recovery cannot protect an entire Flexible scale set with one selection. Your team must enable replication for the individual machines that need recovery.<\/p>\n\n\n\n

This matters when autoscaling creates new machines. A runbook written around the original five instances may be incomplete when the production environment has grown to nine. Your operating process must check that every machine requiring protection is actually registered and healthy in Site Recovery.<\/p>\n\n\n\n

The business outcome is simple: fewer unpleasant surprises when an incident occurs. It also prevents you from paying to replicate temporary machines that could be rebuilt quickly from a standard image.<\/p>\n\n\n\n

2. The recovery scale set must exist before the emergency<\/h3>\n\n\n\n

The target Flexible scale set needs to be prepared in the recovery location before replication is enabled. Its zones, fault-domain design and capacity settings should reflect how the application must run after failover.<\/p>\n\n\n\n

Do not stop at creating an empty scale set. Confirm that the recovery subscription has enough processor quota, suitable machine sizes and access to the required storage services. Capacity shortages during a regional event can turn a planned 60-minute recovery into an open-ended outage.<\/p>\n\n\n\n

For region-to-region recovery, the secondary region protects against a large Azure regional disruption. Zone-to-zone recovery can protect against an availability-zone failure within one region, but it does not cover the loss of that entire region. Not every source and target layout supports both options, so the design needs to match the business risk.<\/p>\n\n\n\n

3. Starting the machines is only part of recovery<\/h3>\n\n\n\n

A successful Azure status screen does not mean employees or customers can use the application. The recovered machines still need working routing, load balancing, name resolution, security rules, certificates, identity access and connections to databases or third-party services.<\/p>\n\n\n\n

Document these dependencies in the order they must be restored. For example, database access may need to be validated before application nodes start, followed by the load balancer, DNS change and user testing.<\/p>\n\n\n\n

Our guide to designing Azure network resilience for Site Recovery failover covers this issue in more detail. A replicated server has little business value if traffic cannot reach it.<\/p>\n\n\n\n

4. Failback needs its own procedure<\/h3>\n\n\n\n

Many runbooks describe how to leave the primary region but say almost nothing about returning. With VMSS Flex, your team must reprotect recovered machines before failing them back to the original location.<\/p>\n\n\n\n

The runbook should define who decides when the original environment is stable, how data changes are synchronised and when traffic can safely return. It should also include a rollback decision if the primary location fails again during the process.<\/p>\n\n\n\n

Agree on the recovery time objective\u2014the maximum acceptable downtime\u2014and the recovery point objective\u2014the maximum amount of recent data the business can afford to lose. These targets should drive the design rather than being added after the technical configuration is complete.<\/p>\n\n\n\n

A practical VMSS Flex recovery runbook<\/h2>\n\n\n\n
    \n
  1. Classify every machine.<\/strong> Record whether it is permanent, temporary, stateful or easy to rebuild. Protect only the machines the business genuinely needs to recover.<\/li>\n
  2. Prepare the target environment.<\/strong> Create the target Flexible scale set, networks, subnets, security controls, load balancers and required access.<\/li>\n
  3. Enable individual replication.<\/strong> Confirm each selected machine appears as a healthy replicated item and that recovery points are being created.<\/li>\n
  4. Create a recovery plan.<\/strong> Group related machines and define their startup order, manual approvals and application validation steps.<\/li>\n
  5. Include autoscaling checks.<\/strong> Add a recurring control to identify newly created production machines that require protection.<\/li>\n
  6. Run an isolated test failover.<\/strong> Validate sign-in, application transactions, monitoring, security controls and external integrations without affecting production.<\/li>\n
  7. Test reprotection and failback.<\/strong> A drill is incomplete until your team has proved it can return safely.<\/li>\n
  8. Record evidence.<\/strong> Keep timings, screenshots, failed checks and improvement actions for management, auditors and future exercises.<\/li>\n<\/ol>\n\n\n\n

    Do not confuse replication with backup<\/h2>\n\n\n\n

    Site Recovery is designed to restore service quickly. It is not a replacement for backup because damaged, encrypted or deleted data can also be replicated to the recovery location.<\/p>\n\n\n\n

    Your plan still needs separate backups with suitable retention and access controls. For a serious tenant compromise, consider the clean recovery approach described in our cross-tenant Azure Backup cyber resilience playbook.<\/p>\n\n\n\n

    This separation also supports stronger alignment with the Essential Eight, the Australian Government\u2019s cybersecurity framework that many organisations use to reduce common attack risks. Disaster recovery, protected backups and tested restoration should work together rather than being treated as one control.<\/p>\n\n\n\n

    A realistic cost and recovery scenario<\/h2>\n\n\n\n

    Consider a 200-person business running twelve VMSS Flex machines during busy periods. Eight are temporary front-end machines that can be recreated from a standard image, while four perform specialised integration and processing work.<\/p>\n\n\n\n

    Replicating all twelve would add unnecessary storage and Site Recovery costs. Protecting the four harder-to-rebuild machines, while using automated deployment for the temporary front end, creates a faster and more economical recovery plan.<\/p>\n\n\n\n

    The runbook then tests the complete service, not just the four replicated machines. This gives leadership a clearer recovery time, avoids paying for protection that adds little value and reduces pressure on the IT team during an incident.<\/p>\n\n\n\n

    Build the runbook around the business service<\/h2>\n\n\n\n

    The biggest change with Azure Site Recovery for VMSS Flex is not a new button in the Azure portal. It is the need to manage protection at the individual machine level while recovering the application as one connected business service.<\/p>\n\n\n\n

    CloudProInc is a Melbourne-based Microsoft Partner with more than 20 years of enterprise IT experience. We help Australian organisations design and test practical Azure recovery plans without the overhead of a large, faceless managed service provider.<\/p>\n\n\n\n

    If you are not sure whether your VMSS Flex environment would recover within the timeframe your business expects, we are happy to review the current design and runbook\u2014no strings attached.<\/p>\n\n\n","protected":false},"excerpt":{"rendered":"

    VMSS Flex recovery now supports stronger end-to-end workflows. Learn how to update your disaster recovery runbook, control costs and avoid gaps during failover.<\/p>\n","protected":false},"author":1,"featured_media":59604,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":true,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_hide_featured_image":false,"_jetpack_featured_image_duplicate_dismissed":0,"_jetpack_feature_clip_id":0,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_post_was_ever_published":false},"categories":[16,131,13],"tags":[],"class_list":["post-59602","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-microsoft-azure","category-backup-disaster-recovery","category-blog"],"jetpack-related-posts":[{"id":59474,"url":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/10\/30\/cross-tenant-restore-with-azure-backup-cyber-resilience-playbook\/","url_meta":{"origin":59602,"position":0},"title":"Cross Tenant Restore with Azure Backup Cyber Resilience Playbook","author":"CPI Staff","date":"October 30, 2026","format":false,"excerpt":"Cross-tenant restore gives Azure customers a clean recovery path when their primary tenant is compromised. This playbook explains how to prepare, test and govern it without creating unnecessary cost.","rel":"","context":"In "Azure"","block_context":{"text":"Azure","link":"https:\/\/www.cloudproinc.com.au\/index.php\/category\/microsoft-azure\/"},"img":{"alt_text":"","src":"\/wp-content\/uploads\/2026\/10\/cross-tenant-restore-with-azure-backup-cyber-resilience-playbook.png","width":350,"height":200,"srcset":"\/wp-content\/uploads\/2026\/10\/cross-tenant-restore-with-azure-backup-cyber-resilience-playbook.png 1x, \/wp-content\/uploads\/2026\/10\/cross-tenant-restore-with-azure-backup-cyber-resilience-playbook.png 1.5x, \/wp-content\/uploads\/2026\/10\/cross-tenant-restore-with-azure-backup-cyber-resilience-playbook.png 2x, \/wp-content\/uploads\/2026\/10\/cross-tenant-restore-with-azure-backup-cyber-resilience-playbook.png 3x, \/wp-content\/uploads\/2026\/10\/cross-tenant-restore-with-azure-backup-cyber-resilience-playbook.png 4x"},"classes":[]},{"id":59538,"url":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/11\/04\/azure-site-recovery-for-azure-local-is-ga-what-it-teams-should-do-next\/","url_meta":{"origin":59602,"position":1},"title":"Azure Site Recovery for Azure Local Is GA What IT Teams Should Do Next","author":"CPI Staff","date":"November 4, 2026","format":false,"excerpt":"Azure Site Recovery for Azure Local is ready for production. Here is how Australian IT leaders can turn simpler deployment into a tested, secure recovery capability.","rel":"","context":"In "Azure"","block_context":{"text":"Azure","link":"https:\/\/www.cloudproinc.com.au\/index.php\/category\/microsoft-azure\/"},"img":{"alt_text":"","src":"\/wp-content\/uploads\/2026\/11\/azure-site-recovery-for-azure-local-is-ga-what-it-teams-should-do-next.png","width":350,"height":200,"srcset":"\/wp-content\/uploads\/2026\/11\/azure-site-recovery-for-azure-local-is-ga-what-it-teams-should-do-next.png 1x, \/wp-content\/uploads\/2026\/11\/azure-site-recovery-for-azure-local-is-ga-what-it-teams-should-do-next.png 1.5x, \/wp-content\/uploads\/2026\/11\/azure-site-recovery-for-azure-local-is-ga-what-it-teams-should-do-next.png 2x, \/wp-content\/uploads\/2026\/11\/azure-site-recovery-for-azure-local-is-ga-what-it-teams-should-do-next.png 3x, \/wp-content\/uploads\/2026\/11\/azure-site-recovery-for-azure-local-is-ga-what-it-teams-should-do-next.png 4x"},"classes":[]},{"id":59161,"url":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/10\/17\/azure-arc-and-site-recovery-without-re-platforming-your-servers\/","url_meta":{"origin":59602,"position":2},"title":"Azure Arc and Site Recovery Without Re-platforming Your Servers","author":"CPI Staff","date":"October 17, 2026","format":false,"excerpt":"Protect hybrid servers with Azure Arc and Azure Site Recovery, reducing downtime without rebuilding applications or committing to a full cloud migration.","rel":"","context":"In "Azure"","block_context":{"text":"Azure","link":"https:\/\/www.cloudproinc.com.au\/index.php\/category\/microsoft-azure\/"},"img":{"alt_text":"","src":"\/wp-content\/uploads\/2026\/10\/azure-arc-and-site-recovery-without-re-platforming-your-servers.png","width":350,"height":200,"srcset":"\/wp-content\/uploads\/2026\/10\/azure-arc-and-site-recovery-without-re-platforming-your-servers.png 1x, \/wp-content\/uploads\/2026\/10\/azure-arc-and-site-recovery-without-re-platforming-your-servers.png 1.5x, \/wp-content\/uploads\/2026\/10\/azure-arc-and-site-recovery-without-re-platforming-your-servers.png 2x, \/wp-content\/uploads\/2026\/10\/azure-arc-and-site-recovery-without-re-platforming-your-servers.png 3x, \/wp-content\/uploads\/2026\/10\/azure-arc-and-site-recovery-without-re-platforming-your-servers.png 4x"},"classes":[]},{"id":59167,"url":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/10\/17\/azure-arc-makes-hybrid-disaster-recovery-simpler-in-australia\/","url_meta":{"origin":59602,"position":3},"title":"Azure Arc Makes Hybrid Disaster Recovery Simpler in Australia","author":"CPI Staff","date":"October 17, 2026","format":false,"excerpt":"Microsoft has simplified disaster recovery setup for Arc-enabled Windows servers. Here is what Australian mid-market organisations should know before using it across a hybrid environment.","rel":"","context":"In "Azure"","block_context":{"text":"Azure","link":"https:\/\/www.cloudproinc.com.au\/index.php\/category\/microsoft-azure\/"},"img":{"alt_text":"","src":"\/wp-content\/uploads\/2026\/10\/azure-arc-makes-hybrid-disaster-recovery-simpler-in-australia.png","width":350,"height":200,"srcset":"\/wp-content\/uploads\/2026\/10\/azure-arc-makes-hybrid-disaster-recovery-simpler-in-australia.png 1x, \/wp-content\/uploads\/2026\/10\/azure-arc-makes-hybrid-disaster-recovery-simpler-in-australia.png 1.5x, \/wp-content\/uploads\/2026\/10\/azure-arc-makes-hybrid-disaster-recovery-simpler-in-australia.png 2x, \/wp-content\/uploads\/2026\/10\/azure-arc-makes-hybrid-disaster-recovery-simpler-in-australia.png 3x, \/wp-content\/uploads\/2026\/10\/azure-arc-makes-hybrid-disaster-recovery-simpler-in-australia.png 4x"},"classes":[]},{"id":59562,"url":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/11\/05\/azure-resilience-checklist-for-australian-mid-market-cios-in-2026\/","url_meta":{"origin":59602,"position":4},"title":"Azure Resilience Checklist for Australian Mid-Market CIOs in 2026","author":"CPI Staff","date":"November 5, 2026","format":false,"excerpt":"Understand where Site Recovery, restore points and immutable backups fit, then use this practical checklist to reduce downtime, ransomware risk and costly recovery gaps.","rel":"","context":"In "Azure"","block_context":{"text":"Azure","link":"https:\/\/www.cloudproinc.com.au\/index.php\/category\/microsoft-azure\/"},"img":{"alt_text":"","src":"\/wp-content\/uploads\/2026\/11\/azure-resilience-checklist-for-australian-mid-market-cios-in-2026.png","width":350,"height":200,"srcset":"\/wp-content\/uploads\/2026\/11\/azure-resilience-checklist-for-australian-mid-market-cios-in-2026.png 1x, \/wp-content\/uploads\/2026\/11\/azure-resilience-checklist-for-australian-mid-market-cios-in-2026.png 1.5x, \/wp-content\/uploads\/2026\/11\/azure-resilience-checklist-for-australian-mid-market-cios-in-2026.png 2x, \/wp-content\/uploads\/2026\/11\/azure-resilience-checklist-for-australian-mid-market-cios-in-2026.png 3x, \/wp-content\/uploads\/2026\/11\/azure-resilience-checklist-for-australian-mid-market-cios-in-2026.png 4x"},"classes":[]},{"id":59417,"url":"https:\/\/www.cloudproinc.com.au\/index.php\/2026\/10\/26\/azure-backup-recovery-checklist-for-premium-ssd-v2-and-ultra-disks\/","url_meta":{"origin":59602,"position":5},"title":"Azure Backup Recovery Checklist for Premium SSD v2 and Ultra Disks","author":"CPI Staff","date":"October 26, 2026","format":false,"excerpt":"Confirm whether Azure Backup can recover individual files from Premium SSD v2 and Ultra Disk workloads before an outage turns into an expensive full-system restore.","rel":"","context":"In "Azure"","block_context":{"text":"Azure","link":"https:\/\/www.cloudproinc.com.au\/index.php\/category\/microsoft-azure\/"},"img":{"alt_text":"","src":"\/wp-content\/uploads\/2026\/10\/azure-backup-recovery-checklist-for-premium-ssd-v2-and-ultra-disks.png","width":350,"height":200,"srcset":"\/wp-content\/uploads\/2026\/10\/azure-backup-recovery-checklist-for-premium-ssd-v2-and-ultra-disks.png 1x, \/wp-content\/uploads\/2026\/10\/azure-backup-recovery-checklist-for-premium-ssd-v2-and-ultra-disks.png 1.5x, \/wp-content\/uploads\/2026\/10\/azure-backup-recovery-checklist-for-premium-ssd-v2-and-ultra-disks.png 2x, \/wp-content\/uploads\/2026\/10\/azure-backup-recovery-checklist-for-premium-ssd-v2-and-ultra-disks.png 3x, \/wp-content\/uploads\/2026\/10\/azure-backup-recovery-checklist-for-premium-ssd-v2-and-ultra-disks.png 4x"},"classes":[]}],"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"\/wp-content\/uploads\/2026\/11\/how-azure-site-recovery-for-vmss-flex-changes-your-dr-runbook.png","_links":{"self":[{"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/posts\/59602","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/comments?post=59602"}],"version-history":[{"count":2,"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/posts\/59602\/revisions"}],"predecessor-version":[{"id":59605,"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/posts\/59602\/revisions\/59605"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/media\/59604"}],"wp:attachment":[{"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/media?parent=59602"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/categories?post=59602"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.cloudproinc.com.au\/index.php\/wp-json\/wp\/v2\/tags?post=59602"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}