@dant123

https://github.com/vatesfr/xen-orchestra/pull/6278

fanuelsen opened this pull request in vatesfr/xen-orchestra

closed feat(xo-server-auth-saml): allow customizing callback URL #6278

@olivierlambert

I have now conluded the problem is related to MCS (Machine Creation Services). MCS does not support XenMotion Storage. When you move vms with regular XenMotion it all works as expected. We are pretty new to MCS thats why this problem shoved up.

If you migrate VMs in XenOrchestra with the default settings without touching the migration network it works fine because its using XenMotion

If you choose a migration network like the picture below it won't work because its using XenMotion Storage

Even if XenDesktop is not officially supported on XCP-ng it has been working fine for us. We have 80 terminal servers on 18 XCP-ng, been running it for years, no big problems.

Anyway thanks for your response.

@danp

Thanks for answering!

xe vm-param-list uuid=46ca24f2-f699-6d33-657d-bbb74b495c97 | grep blocked-operations
    blocked-operations (MRW): migrate_send: VM_CREATED_BY_XENDESKTOP

So I deleted the blocked operations.

xe vm-param-clear uuid=46ca24f2-f699-6d33-657d-bbb74b495c97 param-name=blocked-operations

I then rebooted the machine and checked the allowed-operation

xe vm-param-list uuid=46ca24f2-f699-6d33-657d-bbb74b495c97 | grep allowed-operations
allowed-operations (SRO): changing_dynamic_range; migrate_send; pool_migrate; changing_VCPUs_live; suspend; hard_reboot; hard_shutdown; clean_reboot; clean_shutdown; pause; checkpoint; snapshot

I can now see that migrate_send is back. But when i tried to migrate it again a new error message popped up.

vm.migrate
{
  "vm": "46ca24f2-f699-6d33-657d-bbb74b495c97",
  "migrationNetwork": "359e8975-0c69-e837-58ba-f6e8cd2e1140",
  "sr": "9290c72b-fdcd-cb03-9408-970ade25fcd0",
  "targetHost": "6a310c4f-d3b6-414e-8ca8-2b05b27b45d9"
}
{
  "code": 21,
  "data": {
    "objectId": "46ca24f2-f699-6d33-657d-bbb74b495c97",
    "code": "VDI_ON_BOOT_MODE_INCOMPATIBLE_WITH_OPERATION"
  },
  "message": "operation failed",
  "name": "XoError",
  "stack": "XoError: operation failed
    at operationFailed (/home/node/xen-orchestra/packages/xo-common/src/api-errors.js:21:32)
    at file:///home/node/xen-orchestra/packages/xo-server/src/api/vm.mjs:482:15
    at Object.migrate (file:///home/node/xen-orchestra/packages/xo-server/src/api/vm.mjs:469:3)
    at Api.callApiMethod (file:///home/node/xen-orchestra/packages/xo-server/src/xo-mixins/api.mjs:304:20)"
} 

Do you have any idea where i can poke around next?

Thanks

vm.migrate
{
  "vm": "44db041e-8855-d142-e619-2e2819245a82",
  "migrationNetwork": "359e8975-0c69-e837-58ba-f6e8cd2e1140",
  "sr": "9290c72b-fdcd-cb03-9408-970ade25fcd0",
  "targetHost": "8adfc9f3-823d-4ce8-9971-237c785481a9"
}
{
  "code": 21,
  "data": {
    "objectId": "44db041e-8855-d142-e619-2e2819245a82",
    "code": "OPERATION_BLOCKED"
  },
  "message": "operation failed",
  "name": "XoError",
  "stack": "XoError: operation failed
    at operationFailed (/home/node/xen-orchestra/packages/xo-common/src/api-errors.js:21:32)
    at file:///home/node/xen-orchestra/packages/xo-server/src/api/vm.mjs:482:15
    at Object.migrate (file:///home/node/xen-orchestra/packages/xo-server/src/api/vm.mjs:469:3)
    at Api.callApiMethod (file:///home/node/xen-orchestra/packages/xo-server/src/xo-mixins/api.mjs:304:20)"
}

After updating xcp-ng with latest patches i cant migrate my virtual machines anymore. They are all in the same pool on the same storage. Tried in both XOA and XCP-NG Center, same result.

We are using XCP-NG 8.2

Anyone have a clue where to look?