Share via

Switch appliance in Azure Site recovery.

Sravani, Bodula 0 Reputation points
2026-03-17T10:24:34.36+00:00

I have a old appliance which is currently critical(all its components are critical and the machine is powered off). I have some protected servers using this old appliance. I have deployed a new appliance which is healthy. Now i would like to switch those servers(which are protected by old appliance) to be managed by new appliance. In the ASR UI with switch appliance i need to select the protected servers and give credentials to access the VM. I have added the physical server credentials to access the VM in new appliance but its now showing up in Switch appliance blade, but the credentials are appearing if go to replicate UI view.

I want to know if anyone faced the same strange issue, or Am i missing anything? Please help if you have faced similar issue and what steps taken to resolve.

Thank you

Azure Site Recovery
Azure Site Recovery

An Azure native disaster recovery service. Previously known as Microsoft Azure Hyper-V Recovery Manager.


2 answers

Sort by: Most helpful
  1. Bharath Y P 7,240 Reputation points Microsoft External Staff Moderator
    2026-03-17T10:53:17.83+00:00

    Hello Sravani, it sounds like you’ve built a fresh ASR replication appliance (RA2) and want to move your on-prem servers off the old, powered-off appliance (RA1) onto it, but you aren’t seeing any machines in the Switch appliance blade even though you’ve added credentials. A few things to check and try:

    1. Make sure you’re kicking off the Switch appliance action from RA1 (the critical/offline appliance)
      • Go to Recovery Services vault → Site Recovery infrastructure → ASR replication appliance
      • Select your old appliance (RA1) and click Switch appliance
    2. Verify the servers are actively replicating under RA1
      • Only machines in a “Replicating” state from on-prem to Azure will show up here
      • If replication is paused/failed or the machines have moved, they won’t be listed
    3. Because RA1 is “burnt down” (all components critical), you must re-enter valid OS credentials
      • Under RA2, go to Manage physical & virtual servers (or Site Recovery infrastructure → replication appliance → Configure servers)
      • Add the same credentials here—these accounts need local admin rights on those servers
      • Once RA2 can authenticate, it will install/update the mobility agent pointing at itself
    4. Check network connectivity and firewall/AV rules between RA2 and your servers
      • Required ports (443 and 9443) must be open
      • No deep packet inspection or proxy should be blocking the channel
    5. After you’ve configured the servers on RA2 and validated credentials, retry the Switch appliance workflow
      • You should now see your protected servers in the “Select machines” list

    If you still don’t see them, here are a few follow-up questions to dig deeper:

    • Are your servers showing under Configure servers on the new appliance (RA2)?

    • What OS versions/builds are the physical machines running?

    • Do those VMs/servers show as healthy or paused in the Replicated Items view?

    • Any errors in the mobility service or ASR event logs on the source machines?

    • Did you add credentials via the Portal or via PowerShell?

    Hope this helps you surface the missing piece. Let me know what you find!

    References:

    • Switch Azure Site Recovery replication appliance – https://learn.microsoft.com/azure/site-recovery/switch-replication-appliance-modernized

    • Best practices for ASR replication appliance – https://learn.microsoft.com/azure/site-recovery/replication-appliance-support-matrix

    • Deploy Azure Site Recovery replication appliance (Modernized) – https://learn.microsoft.com/azure/site-recovery/deploy-vmware-azure-replication-appliance-modernized

    0 comments No comments

  2. Q&A Assist
    2026-03-17T10:24:52.5966667+00:00
    AI generated content. This question contains an answer created with AI from Q&A Assist. Learn more

    In the modernized Azure Site Recovery model, credentials used during a Switch appliance operation are supplied in the portal at switch time and are not taken from the appliance’s stored “physical server credentials” list. This is why credentials added on the new appliance appear in the Enable/Replicate workflow but not as a selectable list in the Switch appliance blade.

    Key points from the documented behavior:

    1. Switch appliance workflow
      • In the vault, go to Site Recovery infrastructure → ASR replication appliance.
      • Select the old (critical) appliance and choose Switch appliance.
      • On Select machines, pick the protected machines that were using the old appliance.
      • On Source settings, for each selected machine, choose the new (healthy) replication appliance.
      • If the current appliance has “burnt down” (powered off/critical), the portal will require credentials to access the machines; otherwise, the credentials field is disabled.
      • These credentials are entered directly in this blade; they are not pulled from the appliance’s stored credential list.
    2. Why credentials appear in Replicate but not in Switch
      • When enabling replication for physical servers, credentials are pre‑added to the appliance (via Provide Physical server details → Add credentials) and then selected in the Enable Replication → Source settings step.
      • That credential store is used for discovery and for pushing the mobility agent during Enable Replication.
      • The Switch appliance flow is different: it only exposes a credentials input when the old appliance is unavailable, and it does not enumerate the appliance’s stored credential objects as a list. Instead, credentials are provided ad‑hoc in the switch UI.
    3. What to do in this scenario
      • Proceed with Switch appliance as designed:
        • Select the affected machines.
        • On Source settings, select the new appliance for each machine.
        • When prompted (because the old appliance is critical), manually enter valid OS credentials for each VM so the new appliance can access it.
      • There is no documented requirement or mechanism for the credentials added on the new appliance to appear as a selectable list in the Switch appliance blade, so nothing is missing in configuration.

    If scripted automation is needed instead of the portal, the Start-AzRecoveryServicesAsrSwitchAppliance cmdlet can be used and accepts a -CredentialsToAccessVm parameter, again indicating that credentials are passed at switch time rather than taken from the appliance credential store.


    References:

    0 comments No comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.