The information in this article applies to:
SYMPTOMSWhen you use a chkdsk /f or chkdsk /f /r command on a Microsoft Cluster Server (MSCS) shared drive, chkdsk may not run after the system restarts or it may generate an error message stating that the volume is corrupt. CAUSEUnder most circumstances, running chkdsk with either the /f or /r switch will require restarting because of open handles on the shared drive. Normally, when the computer restarts, there are no services or drivers running that prevent autochk from checking the disk. When using MSCS, the file system does not mount the shared disk until the Cluster service starts because the owner of the shared disk is unknown. This causes the drive type to show up as unknown. Running chkdsk in read-only mode may work properly, but does not fix any problems. RESOLUTION
If file corruption is suspected, follow these steps to close all open
handles to the shared disk and run chkdsk /f against the drive:
To run chkdsk /f, you need to disable the clustering drivers so that, when restarting, the file system can mount. The second node in the cluster needs to be shut down so that it does not gain control of the shared disk when the primary node shuts down. To temporarily disable clustering, use the following instructions:
MORE INFORMATION
For additional information on Microsoft Cluster Server, please see the
following articles in the Microsoft Knowledge Base:
Additional query words: autochk MSCS
Keywords : kbhw kbsetup ntsetup ntdriver NTSrv |
Last Reviewed: January 22, 1999 © 2000 Microsoft Corporation. All rights reserved. Terms of Use. |