Troubleshooting
Problem
ACO5422E BACKUP is terminating abnormally with HRESULT:0x80040bc5 rc = 428
Cause
When the SQL Server's MemToLeave area becomes badly fragmented there is insufficient contiguous space to allocate the buffers required for the VDI (Virtual Device Interface) backups.
Resolving The Problem
ACO5422E Received the following from the MS SQL server: Microsoft][ODBC SQL Server Driver][SQL Server]BACKUP LOG is terminating abnormally. Microsoft SQL-DMO (ODBC SQLState: 42000) (HRESULT:0x80040bc5)
ACO5436E A failure occurred on stripe number (0), rc = 428
ACO5407E The SQL server aborted the operation.
The following errors are in the Windows event application log:
NT AUTHORITY\SYSTEM BackupVirtualDeviceFile::RequestDurableMedia: Flush failure on backup device 'TDPSQL-00000BD4-0000'. Operating system error 995(error not found).
NT AUTHORITY\SYSTEM BackupVirtualDeviceSet::SetBufferParms: Request large buffers failure on backup device 'TDPSQL-00000BD4-0000'. Operating system error -2147024888(error not found).
Operating system error -2147024888 may display when performing a Data Protection for SQL backup. This error indicates that there is not enough storage space to complete the operation. Although reducing the value of the Data Protection for SQL SQLBuffers and SQLBuffersize parameters often allows the backup process to complete, this can cause overall performance to decrease. The default value of the SQLBuffers parameter (0) allows the SQL Server to determine the number of buffers to allocate. Since the buffers are allocated in memory, less memory is used when the value of the SQLBuffers parameter is reduced.
This error occurs when the SQL Server's MemToLeave area becomes so badly fragmented that there is insufficient contiguous space available to allocate the number of buffers required for the VDI (Virtual Device Interface) backups. The result is that the Data Protection for SQL backup fails until the SQL Server is restarted to clean up the MemToLeave area.
Microsoft suggests increasing the MemToLeave area for the SQL Server by using the "-g" switch when starting the SQL Server. However, this may just increase the amount of time that it takes for the memory to become fragmented.
Observation indicates that the following factors seem to make this error occur more quickly:
- large amount of data backed up and/or high backup frequency
- large value for the SQLBuffersize parameter in the tdpsql.cfg file
- using more than one stripe to perform the Data Protection for SQL backup
Although Data Protection for SQL allows the number and size of the SQLbuffers to be specified, it cannot control nor determine how the SQL Server will accommodate these settings. This is a Microsoft SQL Server issue and should be addressed with Microsoft Support.
Product Synonym
TSM
Was this topic helpful?
Document Information
Modified date:
17 June 2018
UID
swg21156744