Diagnostics and error recovery for Azure Import/Export jobs
For each drive processed, the Azure Import/Export service creates an error log in the associated storage account. You can also enable verbose logging by setting the
LogLevel property to
Verbose when calling the Put Job or Update Job Properties operations.
By default, logs are written to a container named
waimportexport. You can specify a different name by setting the
DiagnosticsPath property when calling the
Put Job or
Update Job Properties operations. The logs are stored as block blobs with the following naming convention:
You can retrieve the URI of the logs for a job by calling the Get Job operation. The URI for the verbose log is returned in the
VerboseLogUri property for each drive, while the URI for the error log is returned in the
You can use the logging data to identify the following issues.
The following items are classified as drive errors:
Errors in accessing or reading the manifest file
Incorrect BitLocker keys
Drive read/write errors
The following items are classified as blob errors:
Incorrect or conflicting blob or names
Blob not found
Truncated files (the files on the disk are smaller than specified in the manifest)
Corrupted file content (for import jobs, detected with an MD5 checksum mismatch)
Corrupted blob metadata and property files (detected with an MD5 checksum mismatch)
Incorrect schema for the blob properties and/or metadata files
There may be cases where some parts of an import or export job do not complete successfully, while the overall job still completes. In this case, you can either upload or download the missing pieces of the data over network, or you can create a new job to transfer the data. See the Azure Import/Export Tool Reference to learn how to repair the data over network.