Microsoft KB Archive/931266

= Event ID 57 is logged in the System log of the destination cluster node when you move a resource group to another node of a Windows Server 2003-based server cluster =

Article ID: 931266

Article Last Modified on 2/22/2007

-

APPLIES TO


 * Microsoft Windows Server 2003, Enterprise Edition (32-bit x86)
 * Microsoft Windows Server 2003, Datacenter Edition (32-bit x86)

-



SYMPTOMS
When you move a resource group to a different cluster node on a Microsoft Windows Server 2003-based server cluster, the following event is logged many times in the System log of the destination cluster node: Event Type: Warning

Event Source: Ftdisk

Event ID: 57

Description: The system failed to flush data to the transaction log. Corruption may occur.

Additionally, many informational messages that resemble the description text in the previous event are logged in the %SYSTEMROOT%\Cluster\Cluster.log file.

You experience this problem if the following conditions are true:
 * The resource group contains many disk resources.

Note You may experience this problem when you try to move a resource group that contains as few as five Physical disk resources. However, you are more likely to experience this problem when you try to move a resource group that contains eight or more Physical disk resources.
 * You move the resource group to a cluster node that has not hosted the resource group since the cluster node has been restarted.



CAUSE
This problem occurs if a program or driver that is installed on the affected cluster node delays the Plug and Play (PnP) notification process. A delay in the PnP notification process lets the NTFS file system mount newly-arrived volumes at the same time that the NTFS file system detects other arriving volumes. In this situation, the following symptoms occur:
 * After all the volumes have arrived at the destination cluster node, the Cluster service briefly takes the volumes offline.
 * Because one or more of the volumes have already been mounted, the NTFS file system tries to flush the transaction log of each mounted volume before the volume is taken offline.
 * Because the cluster disk driver (Clusdisk) blocks I/O operations during the process to take the volumes offline, the NTFS file system is unable to flush the transaction log.
 * The volume manager disk driver (Ftdisk) logs the failure to flush the transaction log as event ID 57. This event is logged in the System log on the affected server.



RESOLUTION
To resolve this problem, try to determine which program or driver is delaying the PnP notifications. Then, remove the program or driver. Alternatively, contact the appropriate software vendor to determine whether an update is available for the program or driver.

For information about how to contact software vendors, click the appropriate article number in the following list to view the article in the Microsoft Knowledge Base:

65416 Hardware and software vendor contact information, A-K

60781 Hardware and software vendor contact information, L-P

60782 Hardware and software vendor contact information, Q-Z

Microsoft provides third-party contact information to help you find technical support. This contact information may change without notice. Microsoft does not guarantee the accuracy of this third-party contact information.

Additional query words: MSCS

Keywords: kbtshoot kbclustering kbeventlog kbprb KB931266

-

[mailto:TECHNET@MICROSOFT.COM Send feedback to Microsoft]

© Microsoft Corporation. All rights reserved.