SMS.0807 Network Error Between Source and Target Servers

Symptom

The migration task failed, and the message "SMS.0807 Network error between source and target servers" was displayed on the console.

Possible Causes

During the full replication phase, the source server establishes a connection with the target and spawns a subprocess to transfer data. If the connection is interrupted, the subprocess automatically attempts to reconnect. If reconnection fails after the maximum number of retries is exceeded, this error message is displayed. Locate and rectify the fault based on the possible causes below.

Unstable Network Connection Between the Source and Target Servers

The network connection is unstable, and the retry times exceed the upper limit. In this case, you can configure automatic recovery and adjust the configuration parameters (such as the maximum number of automatic recovery attempts and the interval) based on your requirements to avoid errors caused by network connection timeout. For details, see How Do I Configure Automatic Recovery?

After the configuration is complete, restart the migration task.

Source and Target Server Disconnected

The network is disconnected during data transfer. You can perform the following operations to locate the fault:

1. Check whether the target server is stopped.
   1. Sign in to the Huawei Cloud ECS console.
   2. In the ECS list, check the status of the target server.
      • If the ECS is stopped, choose More > Start in the Operation column. Then restart the migration task.
      • If the ECS is running, go to the next step.

2. Check whether the security group, firewall, or ACL rules of the source server are modified during the migration.

   Check whether the connection is disallowed by any security groups, firewalls, or ACL rules on the network between the source and target servers.

   • If it is, modify their settings to allow the source server to access the target server. For details about the network requirements, see How Do I Set Up a Secure Migration Network for Using SMS? Then you can use the following method to check the connectivity:
     Use SSH or Telnet to check whether the source server can connect to the target server. The commands are in the following format:

     telnet <target-server-IP-address> <port>
     ssh -p <port> <username>@<target-server-IP-address>

     For example, if the target server's IP address is 192.168.0.x and port 22 is used, run either of the following commands:

     telnet 192.168.0.x 22
     ssh -p 22 root@192.168.0.x

     After confirming that the network is connected, restart the migration task.

   • If no, go to the next step.

3. If the issue is not caused by network connectivity problems, it may stem from SSH command execution, such as misconfigured SSH files or insufficient user permissions. Such errors typically occur when Linux file-level migration progress stays at 0%. Perform the following steps:
   1. Open the log file in ../SMS-Agent/agent/Logs/f2f_migrate_error_*.log on the source server. For example: f2f_migrate_error_boot_.log.
   2. Check whether the log file contains the exception information related to "Bad owner or permissions on xxx". If yes, the user who executes the migration does not have the required permission on the xxx file. Run the ll xxx command in any path to check the permission, owner, and group of the file. If the permission is insufficient or the owner is different from the migration user, change the permission settings after carefully evaluating the impact on source services.

      Change the permission: For example, chmod 600 xxx. This command changes the permission on the xxx file to 600.

      Change the owner: For example, chown root:root xxx. This command changes the owner and owner group of the xxx file to root.

   3. Check whether the log file contains the exception information related to "Bad configuration option: xxx". If yes, the SSH configuration file xxx contains abnormal information. Check the xxx file and correct the configuration or comment out the incorrect configuration.

4. If the task is still abnormal after the preceding steps, the sparse files on the source server may have been changed. As a result, the pipe does not respond for a long time when the tar command is used to process these files, causing a timeout. Handle the problem as follows:
   1. Open the log file in ../SMS-Agent/agent/Logs/f2f_migrate_error_*.log on the source server. The file name is f2f_migrate_error_*.log, for example, f2f_migrate_error_var_.log. Check whether the log file contains the exception information related to "Broken pipe". If yes, go to the next step.
   2. Open the configuration file ../SMS-Agent/agent/config/g-property.cfg using a text editor and find the tar.exclude.dir configuration item.
   3. Add ,/var/* to the value of tar.exclude.dir.

      For example, if the original value of tar.exclude.dir is /path/to/exclude1,/path/to/exclude2, the new value must be /path/to/exclude1,/path/to/exclude2,/var/*.

   4. Save the configuration file and restart the migration task.