This guide explains how to identify and resolve issues where a CloudM Migrate server (Primary or Secondary) has stopped working, becomes unresponsive, or encounters connectivity issues during a migration.
Symptoms
You may be experiencing a server issue if you observe the following:
- Stalled Progress: The migration UI shows the migration as "Running," but the item count or data volume has not increased for a significant period.
- UI Errors: The CloudM Migrate user interface displays "Server Error", server connection errors, or "Service Error" messages.
- System Alerts: Monitoring tools report high resource usage (CPU/RAM) or network disconnects on the migration server.
Note: This guide points to gathering service logs, traces, and batch configuration files which are only accessible to CloudM Migrate Self-Hosted users. CloudM Migrate Hosted users must raise a ticket with CloudM Support to investigate issues on that deployment.
Step 1: Diagnose the Issue
Before applying a fix, determine if the issue is related to the CloudM software service or the local server environment.
A. Check CloudM Service Logs
If the server is accessible, check the CloudM Service logs to identify software-level errors.
- See the Logs & Diagnostics section of our KB for information on how to access the service logs and traces on the server in question.
- Look for "[WARN]", "{ERROR]", or "Exception" entries near the time the migration stalled.
B. Check Local Server Health
- Resource Exhaustion: Check Task Manager (Windows) or System Monitor. If CPU or RAM is at 100%, the server may be under-provisioned.
-
Windows Events: Check the Event Viewer, particularly under System and/or Application for more specific information.
- See Providing Other Troubleshooting Logs for more.
-
Network Connectivity: Verify the server has a stable internet connection and can reach both the Source and Destination platforms.
- Test source and destination connections from the Connections tab.
-
Server Connectivity: Verify the primary and secondary servers are connected via the CloudM Migrate Service Manager.
- For more information on multi-server connectivity troubleshooting, see Troubleshooting Multi-Server Connection Issues.
Step 2: Resolution Steps
Scenario A: Primary Server/Service Failure
The Primary server orchestrates the migration. If it fails, the migration logic pauses.
-
Option 1: Attempt Graceful Cancellation (Recommended)
- If the Web UI is still accessible, navigate to the Progress screen and click Cancel.
- Allow the system time to finish current items and generate the migration reports.
-
Option 2: Stop the CloudM Service (Force Stop)
- If the UI is unresponsive or the cancellation process hangs indefinitely ("Stopping" for too long), you must stop the service directly.
- Open Windows Services on the server and Stop the CloudM Migrate Service.
- Warning: Stopping the primary migration service this way results in an "ungraceful" stop. Migration reports will not be generated for this run.
- Restart the Server: Reboot the machine to clear any locked resources or transient network glitches.
-
Restart the Migration:
- Once the server is back online, restart the migration project from the Web UI.
- How it works: CloudM Migrate will scan the migration history to identify items already moved. It will skip these items and resume processing only the remaining data.
- For more information, see Migration History.
Scenario B: Secondary Server Failure
Secondary servers act as workers. If one fails, the impact depends on your configuration.
- If other Secondaries are running: The migration will typically continue on the remaining active servers.
- If this is the only Secondary: The migration will stall, similar to a Primary server failure.
To resolve:
- Resolve the underlying local issue (e.g., restore network connectivity or free up disk space).
- Restart the CloudM service on the Secondary server.
- The server should automatically reconnect to the Primary and pick up pending items from the queue.
Step 3: Addressing Specific Local Issues
-
Resource Exhaustion (RAM/CPU):
- If the server crashed due to lack of resources, you must increase the server specifications (add RAM/CPU) before restarting the migration to prevent a recurrence.
See Insufficient memory to continue the execution of the program.
- If the server crashed due to lack of resources, you must increase the server specifications (add RAM/CPU) before restarting the migration to prevent a recurrence.
-
Network Timeouts:
- CloudM Migrate will retry service connections up to 10 times. If the network is restored before these retries are exhausted, the migration may self-correct. If the maximum retries are reached, follow the Restart the Migration steps above.