Highlights of updates for z/OS V1R11 include the following:
- The sysplex root file system can be dynamically replaced with the alternate sysplex root file system if the ALTROOT statement in the BPXPRMxx member of SYS1.PARMLIB was previously specified. The ALTROOT statement was introduced in z/OS V1R11.
- OEDIT can be used to edit ASCII data, if the file is tagged as ASCII.
- When dynamically replacing the sysplex root file system, you can use the new FORCE option to replace the failing or failed sysplex root file system with the user-specified new sysplex root file system.
- File systems can now be remounted without changing the mount mode.
- A new user syscall trace capability can be used to trace application activity. The trace can be activated at the start of the application or after an application is already running.
New shell commands:
- The bpxtrace command activates tracing for one or more processes. It can be run from the z/OS UNIX shell or the TSO/E environment.
- The ability to use automatically generated UIDs and GIDs instead of default OMVS segments has been added.
Previously, if the BPX.DEFAULT.USER profile in the FACILITY class was defined, users that accessed z/OS UNIX services who did not have an OMVS user or group segment were assigned the default OMVS segments for the length of the user session. All users of the default OMVS segments shared the same UID and GID. As of z/OS V1R11, if BPX.UNIQUE.USER has been defined, users that access z/OS UNIX services who do not have an OMVS user or group segment are automatically assigned an OMVS segment with a unique UID and GID. The new OMVS segments are added to the user and group profiles in the RACF database. It is strongly suggested that unique UIDs and GIDs be automatically generated instead of assigning default OMVS segments.
- A new subtype has been added to the SMF type 92 record. Subtype 15 audits the changing of a file's security attributes for APF-authorized programs, shared library programs, or programs defined to program control.
- Performance for dbx symbolic discovery and queries has been improved.
dbxdebugger can now load debug data on demand and to query symbol information from a module level mapping. Because all of the debug data files do not have to be loaded at startup, debugging of programs can start very quickly. Global module information is kept in a mapping and only a small working set of the debug data is loaded at any time. As a result, symbol queries during a debug session can occur more quickly and the storage requirement for the loaded debug data decreases dramatically.
- To improve performance for high-volume server applications such as web servers, an asynchronous I/O (BPX1AIO) interface to the accept_and_recv (BPX1ANR) callable service is provided.