Return
-
Title
Core Server with NAS/CIFS Repository showing windows event id 50 and 139 with source Mup and/or MRxSmb "### Delayed-Write ###" -
Description
Whenever you find a repository configured on a CIFS Share (NAS) that is showing DVM Bad Blocks, Checksum Errors, Read Errors,or other similar errors associated with repository inconsistencies and the Core Server is showing these errors on the Windows System Event Logs, it is most likely the root cause of the repository inconsistencies.
-
Cause
About event ID 139 with Source Mup, this error can occur if there are pending cached writes to a file located on a remote network share and the connection to that file is unexpectedly terminated. The connection to the file could have been terminated for various reasons, including network communication failures and filter drivers on either the client or the server cancelling the write I/O. When this error occurs, the pending cached write I/O in memory cannot be written to the target file on the remote network share. The delayed write fails and the target file may now be corrupt.
There are several different sources for an event ID 50 message. On this case, an event ID 50 message logged from a MRxSmb/Mup source occurs if there is a network connectivity problem.
-
Resolution
The Repository Inconsistencies were more than likely caused by the environmental issues, so the network performance or instability issue needs to be fixed.
It is a good idea to get started by checking:
- NIC drivers/firmware/settings.
- Switch settings and firmware upgrades.
- NAS configurations and firmware upgrades.
- Also using a 10 Gig connection or at least LACP(Link Aggregation) and NIC Teaming should improve this behavior.
- Apply other networking troubleshooting that is relevant depending on your hardware.