docs(test): hard-stop on migration creation failures and preserve clone

tests/cmc-upgrade-kernel-test.md

@@ -63,6 +63,7 @@ Validate CMC behavior across staged kernel upgrades on a cloned VM, including re
 - If any blocker occurs after clone creation, stop the test and leave the cloned VM powered on for manual inspection.
 - Do not delete or power off the clone on blocker-fail outcomes.
 - Cleanup actions that remove hosts from CMC must target only the cloned host used in the current test run.
+- Treat migration session creation failures (for either migration #1 or migration #2) as blocker-fail events.
 
 ## Test Procedure
 1. Remove offline hosts in `skidamarink` using MCP offline-host cleanup.
@@ -95,39 +96,41 @@ Validate CMC behavior across staged kernel upgrades on a cloned VM, including re
 24. Continue all remaining steps using DHCP IP and credentials from `/home/aw/code/cds/.env.credentials.local`.
 25. Using `cirrusdata` (`gcstage`, project `skidamarink`), reinstall CMC on clone, always adding `-no-prebuilt-mtdi-nexus`.
 26. Create local migration from 10GB source disk to 11GB destination disk using `cirrusdata`.
-27. Wait for initial sync completion.
-28. Check available kernels again using full candidate listing (not latest-only output).
-29. Select first-upgrade target from filtered candidate list (same major; same minor preferred), ensuring it is not the latest candidate.
-30. Verify matching dev/header packages for the selected first-upgrade target are available.
-31. Install selected first-upgrade kernel and matching dev/header packages, then reboot.
-32. Verify running kernel and installed dev/header packages match the selected first-upgrade version.
-33. If versions do not match exactly, stop as blocker-fail.
-34. After reboot, verify clone is online in `skidamarink` using `cirrusdata`.
-35. SSH to clone and verify MTDI, Galaxy Migrate services/driver are up.
-36. Write sample data to source 10GB disk.
-37. Trigger sync and confirm tracking status using `cirrusdata`.
-38. Uninstall CMC.
-39. Post-uninstall cleanup checkpoint:
+27. If migration session creation fails (including API/service errors such as 5xx), hard stop as blocker-fail.
+28. Wait for initial sync completion.
+29. Check available kernels again using full candidate listing (not latest-only output).
+30. Select first-upgrade target from filtered candidate list (same major; same minor preferred), ensuring it is not the latest candidate.
+31. Verify matching dev/header packages for the selected first-upgrade target are available.
+32. Install selected first-upgrade kernel and matching dev/header packages, then reboot.
+33. Verify running kernel and installed dev/header packages match the selected first-upgrade version.
+34. If versions do not match exactly, stop as blocker-fail.
+35. After reboot, verify clone is online in `skidamarink` using `cirrusdata`.
+36. SSH to clone and verify MTDI, Galaxy Migrate services/driver are up.
+37. Write sample data to source 10GB disk.
+38. Trigger sync and confirm tracking status using `cirrusdata`.
+39. Uninstall CMC.
+40. Post-uninstall cleanup checkpoint:
    - Run MCP offline-host cleanup for `skidamarink`.
    - If the cloned VM is still marked online after uninstall, remove that cloned VM host entry specifically via MCP (target only this test clone host).
    - Because CMC status can lag behind VM state, poll briefly for status transition; if still online, perform targeted MCP host removal for the tested clone.
-40. Check available kernels.
-41. Select latest-upgrade target kernel from the filtered candidate list (same major required; same minor preferred).
-42. Verify matching dev/header packages for the selected latest-upgrade target are available.
-43. Install selected latest-upgrade kernel and matching dev/header packages, then reboot.
-44. Verify running kernel and installed dev/header packages match the selected latest-upgrade version.
-45. If versions do not match exactly, stop as blocker-fail.
-46. Reinstall CMC via `cirrusdata` (`gcstage`, `skidamarink`), always adding `-no-prebuilt-mtdi-nexus`.
-47. Create a local migration (10GB -> 11GB) via `cirrusdata` and wait for initial sync completion.
-48. Confirm machine is online in `skidamarink` using `cirrusdata`.
-49. SSH and verify MTDI, Galaxy Migrate services/driver are up.
-50. Success-path cleanup only: power off cloned machine.
-51. Success-path cleanup only: delete cloned VM and its disks from vCenter inventory.
-52. Success-path final cleanup checkpoint:
+41. Check available kernels.
+42. Select latest-upgrade target kernel from the filtered candidate list (same major required; same minor preferred).
+43. Verify matching dev/header packages for the selected latest-upgrade target are available.
+44. Install selected latest-upgrade kernel and matching dev/header packages, then reboot.
+45. Verify running kernel and installed dev/header packages match the selected latest-upgrade version.
+46. If versions do not match exactly, stop as blocker-fail.
+47. Reinstall CMC via `cirrusdata` (`gcstage`, `skidamarink`), always adding `-no-prebuilt-mtdi-nexus`.
+48. Create a local migration (10GB -> 11GB) via `cirrusdata` and wait for initial sync completion.
+49. If migration session creation fails (including API/service errors such as 5xx), hard stop as blocker-fail.
+50. Confirm machine is online in `skidamarink` using `cirrusdata`.
+51. SSH and verify MTDI, Galaxy Migrate services/driver are up.
+52. Success-path cleanup only: power off cloned machine.
+53. Success-path cleanup only: delete cloned VM and its disks from vCenter inventory.
+54. Success-path final cleanup checkpoint:
    - Run MCP offline-host cleanup for `skidamarink`.
    - If the cloned VM is still marked online at the end of the test, remove that cloned VM host entry specifically via MCP (target only this test clone host).
    - Because CMC status can lag behind VM deletion/power-off, wait/poll briefly first; if still online, perform targeted MCP host removal for the tested clone.
-53. Blocker-fail path after clone creation:
+55. Blocker-fail path after clone creation:
    - Stop test immediately after recording failure details.
    - Leave cloned VM powered on and present in inventory for manual inspection.
    - Do not run clone power-off/delete steps in blocker-fail path.
@@ -138,6 +141,7 @@ Validate CMC behavior across staged kernel upgrades on a cloned VM, including re
 - Clone cannot be created on datastore `AutomatedTest-UnitTesting`.
 - DHCP transition cannot be completed (clone remains static at `<INITIAL_CLONE_HOST_OR_IP>`).
 - Kernel upgrade candidate criteria not met.
+- Migration session creation failed (including API/service errors such as HTTP 5xx or equivalent backend unavailability).
 - Any critical migration/service validation failure that blocks continuation.
 
 ## Per-Host Test Result Record