z/OS Network File System Guide and Reference
Previous topic | Next topic | Contents | Contact z/OS | Library | PDF


NFS server cache monitoring and reporting

z/OS Network File System Guide and Reference
SC23-6883-00

The NFS Server issues a message to indicate an MVS data buffer shortage each time the server's threshold limit is reached, and when the threshold is overcome. This message implies that some of the current requests may fail. This threshold is governed by the server site attributes that specify the storage limit (in bytes) for the buffers (bufhigh attribute) and a percentage of bufhigh storage limit before BUFHIGH watermark is reached (the percentsteal attribute). See Buffer usage and caching for more information.

To mitigate this situation, a buffer alert limit can also be specified with the z/OS NFS server bufhigh site attribute, or the MODIFY operation command bufhigh operand. When an alert limit is set and that value is reached, message GFSA383I is written to the console reporting the current rounded usage percentage of the maximum allowed data buffer pool size value (bufhigh). While the data buffer usage remains above the alert limit, a new message will be written to the console every 30 seconds reporting the current usage value.

If the buffer usage drops below the alert limit, message GFSA384I is written to the console to report that the buffer shortage situation has been mitigated.

System Automation can be configured to monitor these warning messages and inform the application administrators on the remote NFS clients to react accordingly and possibly throttle the incoming workload before the z/OS NFS server reaches a critical "out of memory" situation.

Go to the previous page Go to the next page




Copyright IBM Corporation 1990, 2014