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


Using z/OS component tracing

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

To improve the recording and tracing of NFS diagnostic information, you can use component trace services. After activating component tracing and recreating the problem, the diagnostic information in the trace buffers can then be captured in a z/OS® SVC dump or written to a DASD or tape data set and then viewed using the IPCS CTRACE command. Use of component trace services offers other advantages over using the server's log data sets, including performance improvements and virtual storage constraint relief.

To activate component tracing of NFS information, an operator can issue a TRACE CT command specifying the procedure name of the z/OS NFS server or client with individual trace options or a member of SYS1.PARMLIB that contains the trace options. The operator can choose different PARMLIB members (if they were previously created) by specifying CTRACE=nn on a START mvsnfs command, where nn is the suffix of a CTINFSnn PARMLIB member. When CTRACE=nn is not specified, the default PARMLIB member, CTINFS00, is used to initialize CTRACE for the server. If the specified PARMLIB member is absent or incorrect, CTRACE is activated with the default options in effect (CTRACE buffer size of 10M and FFDC,ERROR trace options). For details on these commands and their NFS trace options, see Network File System operation. For details on the member of SYS1.PARMLIB, see Using NFS component trace PARMLIB members CTINFSnn and CTINFCnn .

The following sections describe how to use NFS component tracing in diagnosis procedures. Further information on using the IPCS CTRACE command can be found in z/OS MVS IPCS Commands. Additional information regarding the MVS™ DISPLAY, TRACE, and DUMP commands may be found in z/OS MVS System Commands.

When you use component tracing, NFS diagnosis information continues to be recorded in the NFS server log data sets, at the level specified on a MODIFY mvsnfs,LOG=xxx operator command. IBM® recommends that you adjust the level of logging to avoid duplicating diagnostic records and to maximize the performance benefits of using component trace buffers.

If additional NFS diagnosis information is required using the log data sets for the z/OS NFS server, see Capturing diagnostic information using z/OS NFS log data sets and from other components for further details.

Go to the previous page Go to the next page




Copyright IBM Corporation 1990, 2014