![[Contents]](TOC.GIF)
![[Prev Chap]](REW.GIF)
![[*]](BLANK.GIF){kind=spacer}
![[Next Sect]](NEXT.GIF)
![[Next Chap]](FF.GIF)
![[Index]](INDEX.GIF)
![[(i)]](HELP.GIF)
The DIGITAL UNIX Version 4.0D software is a replacement for the Version 4.0A, Version 4.0B, and Version 4.0C operating system software and provides new or additional support for the following:
defragment
utility
salvage
utility (field test version)
pread
and
pwrite
For more information about some of these features, see the following sections. You should also refer to Chapter 8, which describes features and interfaces scheduled to be retired in future versions of the operating system.
DIGITAL UNIX, including all system components and packaged layered products, has gone through an extensive evaluation to ensure that dates are handled and represented properly for the Year 2000 (Y2K) and beyond. You can find out more about this evaluation and other Y2K topics in the white paper titled DIGITAL UNIX Year 2000 Readiness that is located on the DIGITAL UNIX V4.0D Documentation, Volume 1 CD-ROM in HTML format. This document contains important information about the Y2K problem's background, Y2K readiness of DIGITAL UNIX, testing methodologies and guidelines for applications, and details on DIGITAL UNIX Y2K updates and features. It also contains information about Y2K contacts within DIGITAL, plus URL pointers to additional information and the latest version of this document on the DIGITAL UNIX external web page.
Note that the License Management Facility and supporting code have also been enhanced to allow users with restricted PAKs to set the date forward for Year 2000 (Y2K) testing. For more information, see Section 4.3.3.
![[Prev Sect]](PREV.GIF)
On UNIX systems, each user is known to the system by a unique number called a user identifier (UID). The system also knows each user group by a unique number called a group identifier (GID). The system uses these numbers to track user file access permissions and group privileges and to collect user accounting statistics and information.
With DIGITAL UNIX Version 4.0D, the maximum number of UIDs and GIDs allowed increases from 65,535 (16 bits) to 2,147,483,647 (31 bits). Note that this increase does not mean that 2.14 million users can simultaneously log onto a system; rather, it means that 2,147,483,647 user and group identifiers are available. The maximum number of users that can be logged on is determined by the available system resources.
To preserve strict binary compatibility with legacy applications, the new limits are not enabled by default in Version 4.0D. See Appendix C for instructions about how to enable extended UIDs and GIDs and a list of restrictions that apply to the implementation of the new limits.
The Class Scheduler provides you with a method for controlling the execution of tasks or applications by restricting the amount of time that they can access the processor (CPU).
For example, background daemons such as the print spooler can be given
less access time. The CPU will then have more time available to
perform other tasks. To do this, you specify that the print daemon,
/usr/lbin/lpd,
is allowed to use no more than a certain percentage of the available
CPU time. You can also group resource user identifiers, such as a
user's UID, into classes and assign the required CPU access time to
each class.
Furthermore, this feature can help you to allocate system resources so that the most important work receives the required processing time. For example, you may want to run two versions of a production database on your system. One version is used as part of your business operations, while the other is a test copy, with different tuning parameters. The test database can be assigned to a different class so that your daily operations are not impacted by the testing.
For more information on class scheduling, refer to the following
reference pages:
class_scheduling(4),
class_admin(8),
runclass(1),
and
classcntl(2).
By default in DIGITAL UNIX, when a core file is written to a disk,
the system saves the file under the name
core.
Each subsequent core file overwrites its predecessor because the file
name is identical.
When the new enhanced core file-naming feature is enabled, the system will attempt to create unique names for core files in the form:
core.prog-name.host-name.tag
The uniquely named files that result will not be overwritten by subsequent core files, thereby preventing the loss of valuable debugging information when the same program or multiple programs fail multiple times (and perhaps for different reasons).
If you want your core files to be saved with unique file names, you must enable this feature by following the instructions in Appendix G.
Crash dumps are now compressed to reduce the reboot time associated
with a crash dump and to reduce the disk space required to hold the dump
file. The crash dump utility,
savecore,
automatically detects and moves dump files from the swap partition to
/var/adm/crash
and saves them under the name
vmzcore.n to distinguish them from noncompressed dump files
(vmcore.n).
This release contains a new utility named
expand_dump
to enable you to produce a conventional
noncompressed crash dump file from a compressed crash dump. Also,
the
dbx
debugging utility has been upgraded to allow you to read the
compressed crash dump files.
For more information, see
Section 4.3.5
and the
savecore(8)
and
expand_dump(8)
reference pages.
The binary event logger puts event data in a kernel buffer. Large servers that contain large amounts of physical memory and that use many disks often require a larger kernel buffer than a small system requires. The size of the kernel buffer is now determined by the amount of physical memory in the system at boot time. The size assigned to the kernel buffer now can range from 32 KB to 1 MB.
If necessary, you can override the assigned kernel buffer size by
specifying a value for the new
binlog-buffer-size
attribute in the
generic
subsystem using
sysconfigdb
or
dxkerneltuner.
The minimum size of the kernel buffer is 8 KB; the maximum size
is 1 MB. Refer to the
sysconfigdb(8)
reference page and the
System Configuration and Tuning
guide for information about modifying attributes.
You still can increase the size of the kernel buffer by editing the
/sys/data/binlog_data.c
file, as described in the
System Administration
guide; however, you can decrease the size of the kernel buffer only
by modifying the
binlog-buffer-size
attribute.
Early child exit notification allows a parent process to be notified
via a
SIGCHLD
signal as soon as the system detects that a child process is about to
be terminated by a fatal signal. This feature is needed for
responsive failover mechanisms that vital system
daemons and other similar programs can use. For example, early notification
is especially helpful when the child process will be generating a core
file as the result of its abnormal termination. Creating a core file
can take an indefinite amount of time, and without early child exit
notification it can delay the parent's failover operation.
For more information, see Appendix H.
IP switching over ATM provides a method for carrying IP traffic over ATM, based on the Ipsilon Networks, Inc. reference model.
A DIGITAL UNIX host is attached through a point-to-point physical connection to an IP switch. The IP switch is an ATM switch with added IP controller software that performs IP routing and traffic classification. The IP controller and the DIGITAL UNIX host communicate using the Ipsilon Flow Management Protocol (RFC 1953) to dynamically create ATM virtual circuits for carrying IP traffic. DIGITAL UNIX Version 4.0D provides support for IP switching on a dedicated ATM adapter.
ATM LAN Emulation (LANE) provides a means to carry Ethernet or IEEE
802.3 LAN protocols over ATM, including multicast and broadcast
protocols. In DIGITAL UNIX Version 4.0D, LANE support has been
improved to more closely emulate an Ethernet interface, including
support for
packetfilter
and
tcpdump.
In addition, use of the LAT protocol is now supported over
ATM-emulated LANs.
In this release, most of the ATM subsystem has been more closely
integrated with the rest of the DIGITAL UNIX networking support.
ATM adapters, signaling, Classical IP (over SVC's only), and LAN
Emulation can now be started automatically during system boot.
Also, you can now use the
netsetup
and
netconfig
utilities to configure ATM logical network interfaces (LIS and ELAN).
An
atmsetup
utility is provided to manage the ATM boot-time configuration
database. The
/etc/atm.conf
file is now used only for configuring Classical IP over PVCs, IP
switching, and user-written convergence modules. Further enhancements
to
atmsetup
and boot-time startup are planned for a future release.
For more information, refer to
Section 4.4.2,
Section 8.25,
the
Asynchronous Transfer Mode
guide, and the
atmsetup(8)
reference page.
Starting with Version 4.0D, DIGITAL UNIX supports
Classless Inter-Domain Routing (CIDR), which includes supernetting and
route aggregation support. Refer to the
CIDR(7)
reference page for additional details about these features.
In previous versions of the DIGITAL UNIX operating system, the number of file descriptors, or the maximum number of open files per process, was limited to 4096. For DIGITAL UNIX Version 4.0D, the number has been increased to 65,536 file descriptors (64K) per process.
The previous maximum number of file descriptors (4096) is now the default for backward compatibility reasons. However, customers whose applications need many thousands of open files can obtain them by using the procedure detailed in Appendix E.
New and changed documentation for this feature includes the following
reference pages:
setsysinfo(2),
getsysinfo(2),
setrlimit(2),
select(2),
and
poll(2).
Header files that have been changed for this feature include the
following files:
<param.h>
See the
NEW_OPEN_MAX_SYSTEM
field.
<sysinfo.h>
See the
SSI_FD_NEWMAX
and
GSI_FD_NEWMAX
fields.
Starting with DIGITAL UNIX Version 4.0D, the Compact Disk File
System (CDFS) supports the organization of multiple sessions on one
CD-ROM volume. Note that the contents of all sessions are available
as one file system and are not available separately. See the
cdfs(4)
reference page for more information.
You no longer need to select the CD-ROM File System (CDFS) as a
kernel option to access information on ISO 9660 or ISO 9660 Rock Ridge
CD-ROMs. Starting with DIGITAL UNIX Version 4.0D, the CDFS is
dynamically loaded and enabled whenever you specify the
-t cdfs
option with the
mount
command.
As in previous versions of DIGITAL UNIX, you can still select the CDFS as a kernel option. However, users may want to avoid adding the CDFS enablers to the kernel on systems with limited system resources because doing so will increase the amount of memory that the kernel will consume.
Either method will work, but not selecting the CDFS kernel option simply delays the allocation of system resources until a CDFS CD-ROM is mounted.
A new option,
-o dual,
has been added to the
mount
command, allowing AdvFS users with plexed (mirrored) LSM volumes to
use a backup procedure similar to that described in the
Logical Storage Manager
guide in Section 8.3.3, "Backup Using a Plex."
In previous releases of DIGITAL UNIX, customers using LSM with a UFS file system could perform the backup as documented. However, customers using LSM with an AdvFS file system were unable to perform this operation because the AdvFS domain ID for the unmounted (and disassociated) plex was the same as the disk volume and the plex(es) that were still on line. Attempts to mount the unmounted plex again in order to restore the plex set would fail. The I/O error reported indicated that an attempt was made to mount an AdvFS domain ID that was already mounted.
The
-o dual
mount option solves this problem for filesets that are mounted using
it. However, this practice is not recommended when LSM and AdvFS are
used together. Instead, DIGITAL recommends using the method described
in Chapter 4 of the
AdvFS Administration
guide, under "Backups,"
and summarized here:
clonefstet
command.
umount
command.
vdump
command.
mount
command.
The
/usr/bin/defragment
utility, used to clean up or eliminate fragmentation in AdvFS file
domains, was formerly provided with the POLYCENTER AdvFS Advanced
Utilities layered product in the
AFAADVANCEDxxx
subset. Starting with Version 4.0D, it is shipped with the base
operating system in the
OSFADVFSxxx
subset and no longer requires the AdvFS Advanced Utilities license.
There is also a new shell script,
/usr/sbin/defragcron,
which will run the AdvFS
defragment
utility over all file domains or a given set of file domains. This
script has been added to the
crontab
entries for root. By default, this will defragment all file domains
every night between 1 a.m. and 5 a.m., local time. DIGITAL recommends
that you change this
crontab
entry so that the
defragment
utility is run at a time that better suits the specifics of your
environment.
See the
defragment(8)
reference page for more information about defragmenting AdvFS file
domains. See the
defragcron(1)
and
crontab(1)
reference pages for information on how to change the default
defragmentation events to better suit your local environment.
An AdvFS
salvage
utility is being developed for a future version of the DIGITAL UNIX
operating system. This utility facilitates the recovery of file data
from damaged AdvFS file domains.
A field test version of the
salvage
utility is being delivered with this version of the operating system.
The utility is located in
/usr/sbin
and is documented in
salvage(8).
Note that because the
salvage
utility is still in development, it is not yet officially supported by
DIGITAL. Customers may test the utility in a controlled environment,
one that can withstand the loss of data or data corruption; however,
it should not be used in a production environment. Furthermore, the
availability of this field test version of the
salvage
utility should not be considered an alternative to regular backups
of critical production data.
If you are in doubt about whether or how to use the
salvage
utility, contact a DIGITAL service representative.
In Version 4.0D and subsequent releases of DIGITAL UNIX, the
prmetaonly
attribute for the
prestoserve
subsystem may now be used with AdvFS. This attribute controls whether
PrestoServe caches only file system metadata instead of both metadata
and synchronous write data (the default). If the attribute is set to
1 (enabled), PrestoServe caches only metadata on all file systems for
which PrestoServe is enabled.
The use of PrestoServe can improve the overall run-time performance on systems that perform large numbers of synchronous writes. Caching only metadata may improve the performance of applications that access many small files or applications that access a large amount of file-system metadata but do not re-read recently written data.
For more information, see the System Configuration and Tuning guide.
AdvFS now logs a domain panic in the binary error log as well as displaying a message at the console. When a domain failure occurs, a problem-specific message is generated, as well as the following string:
AdvFS Domain Panic; Domain <domain_name> Id <domain_id>
You can use the DIGITAL UNIX Event Manager to examine the binary
error log for domain panics. For more information, see the
binlogd(8)
reference page. Also, see
Appendix F
for information about recovering from domain panics.
The Virtual Memory subsystem now supports a new mechanism to expedite
memory region lookup operations in the kernel. Applications that use
large number of map entries (e.g. many memory mapped files or a large
number of threads) will benefit from the faster lookup. This
feature is enabled by default
(vm-map-index-enabled).
See the
System Configuration and Tuning
guide for information about additional parameters that allow fine tuning.
Use the new
PCI_Developer_Debug
attribute in the
pci
subsystem to display PCI configuration code for static drivers and for
dynamically loaded drivers. The attribute allows driver developers
to obtain PCI bus-level hardware information and the driver
configuration status as the kernel invokes the PCI bus code. Using
sysconfigdb,
you can set the
PCI_Developer_Debug
attribute to debugging values that are defined in the
/usr/sys/include/io/dec/pci/pci_debug.h
file.
For example, if you set the
PCI_Developer_Debug
attribute to 0x00000020, which is the value of the
PCID_CNFG_HDR_S
definition in the
/usr/sys/include/io/dec/pci/pci_debug.h
file, the PCI bus configuration code will display the value of every
field read from the PCI configuration header, as the bus code reads
and assigns the kernel-related data structures in the PCI
configuration header.
Refer to the
sysconfigdb(8)
reference page and the
System Configuration and Tuning
guide for information about modifying system attributes.
The
pixstats
command is now implemented as an option to the
prof
command. Both
pixstats(1)
and
prof(1)
style command syntax are supported by the new implementation. The
following changes have been made between the new and earlier
implementations of
pixstats:
prof -pixstats
instead of
pixstats.
proc_at_0x
if no symbol is available) for an initial block and
<procedure-name>+<offset>
for subsequent blocks.
jsr
or
bsr
instruction
counts.
For details on the use of the
prof -pixstats
command, see the
prof(1)
reference page.
Since the release of DIGITAL UNIX Version 4.0, the Ladebug and
TotalView debuggers have been able to determine the state of threads
in a threaded process by using the
libpthreaddebug.so
shared library. Until Version 4.0D, the header file for this library
was not available on the DIGITAL UNIX kit.
With DIGITAL UNIX Version 4.0D, the header file
pthread_debug.h
provides the DECthreads debugging interface and the sole documentation
regarding the interface and library. This header and library are
currently provided with minimal support. DIGITAL expects that the
interface will continue to evolve as we and our partners gain
experience with it. Although DIGITAL will often be able to retain binary
compatibility with earlier versions of the interface, this cannot
currently be guaranteed.
The CDE desktop now supports the proper display of multiple languages in
one login session as long as the correct fonts have been installed.
To enable this support, set the
LANG
environment variable to the desired language and start your
applications. The applications will display the characters
appropriate to the specified locale.
There are some limitations to this new feature:
LANG
environment variable to a certain language, the ASCII fonts used by
the applications you invoke may be different from the ones used when
you start the whole CDE desktop in that language.
Beginning with Version 4.0D, DIGITAL is mastering the DIGITAL UNIX Documentation CD-ROM in ISO 9660 format.
The HTML, PostScript, and text areas of the CD-ROM are ISO 9660 Level
1 compliant; any system that supports the ISO 9660 Level 1 CD-ROM
standard can mount the CD-ROM and browse the files in these areas.
If you are using DIGITAL UNIX, you can mount the CD-ROM by
specifying the
-t cdfs
option with the mount command, as in the following example:
#
mount -rd -t cdfs -o rrip /dev/rz4c /usr/share/doclib/online
The HTML library is located in the following file:
/mnt-point/DOCS/HTML/LIBRARY.HTM
It provides links to the HTML, PostScript, and text documents on the
CD-ROM. You might need to configure your browser's helper
preferences to start a PostScript viewer for viewing the PostScript
documents. See Section 6.6 of the
Installation Guide
for Version 4.0B for more information about invoking the Netscape
Navigator to view the documentation on systems running DIGITAL UNIX.
The CD-ROM also contains the complete set of DIGITAL UNIX base
system reference pages for browsing with the DIGITAL UNIX
webman
viewer. The files in this area are usable only on UNIX systems that
support the Rock Ridge Extensions to the ISO 9660 standard. See the
following file for more information about installing and using
webman:
/mnt-point/DOCS/HTML/WEBMAN/INSTALLI.TXT
1.24 DIGITAL UNIX Device Driver Kit
Starting with DIGITAL UNIX Version 4.0D, DIGITAL is shipping all
device driver-related information in the DIGITAL UNIX
Device Driver Kit
(DDK). The DDK includes the following:
-
Books
-
Writing Network Device Drivers
-
Writing a Graphics Device Driver and DDX for the DIGITAL UNIX X Server
-
Writing Device Drivers: Tutorial
-
Writing Device Drivers: Reference
-
Writing Device Drivers: Advanced Topics
-
Writing Device Drivers for the SCSI/CAM Architecture Interfaces
-
Writing EISA and ISA Bus Device Drivers
-
Writing PCI Bus Device Drivers
-
Writing TURBOchannel Device Drivers
-
Writing VMEbus Device Drivers
-
Read This First letter
-
Release Notes
-
White papers
-
Examples
-
X Consortium source code
To order a DIGITAL UNIX
Device Driver Kit
(order number QA-MT4AV-G8), contact your DIGITAL sales representative
or call 1-800-DIGITAL.
1.25 DE500-BA and DE500-FA Support
Starting with Version 4.0D, DIGITAL UNIX supports the DE500-BA and
DE500-FA network interface adapters. The DE500-BA is a
10BaseT/100BaseTX adapter that supports 802.3u standard autonegotiation,
and the DE500-FA is a 100BaseFX adapter. See the
tu(7)
reference page for more details on how to configure these adapters.
1.26 Server System MIB and Server Management MIB Subagent
The DIGITAL Server System and Server Management MIBs (Management
Information Bases) are new subagents with DIGITAL UNIX
Version 4.0D that provide the ability to manage servers locally and
remotely.
The DIGITAL Server System MIB facilitates the monitoring of the
complete state of an AlphaServer system, including hardware, firmware,
and environmental information. The DIGITAL Server Management MIB
allows an administrator to manage the attributes of any MIBs. It
monitors the attribute values, invokes action scripts if the values
exceed their predefined thresholds, and if specified, polls the
attributes on every restart of the subagent.
A Server Management MIB enabler has been provided for the ServerWORKS
and Polycenter Netview software packages, so administrators can manage
server systems with these packages.
You can find the Server System MIB
(svrSystem.mib)
and Server Management MIB
(svrMgt.mib)
definitions in the
/usr/share/sysman/mibs
directory. You can find the Server System subagent
(svrSystem_mib)
and Server Management subagent
(svrMgt_mib)
in the
/usr/sbin
directory. The subagents and the MIBs are a part of the
OSFCLINET425
mandatory subset.
1.27 VMEbus Backplane Driver Enhancements
The VMEbus backplane
(vb)
network driver, supported on DIGITAL AXPvme and Alpha VME systems, has
been enhanced to conform to the single binary module method for static
configuration. Therefore, to configure
vb
on systems running Version 4.0D, you need only adjust certain
subsystem attributes using the
sysconfigdb
utility. You no longer need to edit the
/usr/sys/data/if_vb_data.c
file, as was required in previous versions of DIGITAL UNIX.
For more information about configuring
vb,
see the
vb(7)
and
vme_manual_setup(7)
reference pages.
Note that although the instructions in the
vb(7)
reference page specify the process for editing the
/etc/sysconfigtab
file manually to specify the
vb
configuration attributes, you should use
sysconfigdb
instead. As discussed in
Section 4.3.1,
using a standard text editor to edit the file could cause a problem
during your next update installation. Refer to the
sysconfigdb(8)
reference page and the
System Configuration and Tuning
guide for information about modifying system attributes.
1.28 New Printer Support
The
lprsetup
utility has been updated to include support for the following
printers: LA30N, LA30W, LG04plus, LG08plus, LG12plus, LGL04plus,
LGL08plus LN17, and LN17ps.
1.29 New Tape Drive Support
DIGITAL UNIX Version 4.0D adds support for the following tape
drives: TZS2, TZS20, TLZ10, and TLZ1.
1.30 DEC C Compiler Version 5.6
The DEC C compiler has been upgraded to Version V5.6. This new
version contains both bug fixes and improved optimizations. It also
contains some new features, which are listed below:
-
Header file protection:
-protect_headers
and
protect_headers_setup
The
-protect_headers
feature in DEC C can solve pointer size and data alignment problems
when using
-xtaso_short,
-nomember_alignment,
or
-Zpn
(where n!=8). See the
cc(1)
and
protect_headers_setup(8)
reference pages for more information.
-
Compiler support for Thread Local Storage (TLS)
TLS allows static extent data to be allocated on a per-thread basis.
Therefore, each thread has its own copy of the data that can be
modified without affecting the values seen by the other threads in a
process. For more information about TLS support, see
Appendix I.
-
New diagnostics to detect undefined side effects
These diagnostics detect simple expressions with side effects that
are undefined in ANSI C. Statements such as
i = i++;
will now be diagnosed as relying on undefined behavior.
-
Enabled
-preempt_module
for better optimization in non-shared compilations
With this change, the
cc
driver will look for conditions when the compilation cannot produce a
shared library, and if found, it will pass the
-preempt_module
flag to DEC C, enabling DEC C to further optimize the module. The
previous default was
-preempt_symbol.
The improved optimizations that this compiler offers may expose latent
source code bugs in your application. Software tools such as
atom
and
third
may be used to help diagnose these problems.
1.31 cc.alt Compiler Updated
The cc.alt subset
(CMPDEVALT420)
has been updated. See the
/usr/lib/cmplrs/cc.alt/README
file in the subset for a description of the changes.
1.32 DEC C++ Runtime Library Updated
The DEC C++
libcxx
runtime support library has been updated for DIGITAL UNIX Version
4.0D to provide support for the DEC C++ Version 6.0 compiler.
The update also fixes a memory leak that occurred in
libcxx.so
for previous releases. C++ programs that created threads and used the
throw()
function from within those threads exposed this memory leak within the
runtime exception routines.
1.33 Ladebug Debugger Updated
DIGITAL UNIX Version 4.0D contains an updated version of the Ladebug
Debugger, Version 4.0-43. The release notes for this version of
Ladebug are located in the
OSFLDBDOC425
subset on the
DIGITAL UNIX V4.0D Operating System, Volume 1
CD-ROM and can be accessed only when Ladebug is installed. After
installation, you can find the release notes the following locations:
/usr/doc/ladebug/release-notes.txt (plain text)
/usr/doc/ladebug/release-notes.html (HTML)
The following changes have been made since the last release of
Ladebug:
-
The internationalized version of the debugger is now available in all
kits of the Version 4.0-24 and later debuggers. You no longer need to
install the special IOSLDBxxx subsets of the debugger software to
achieve localization to other language locales. The IOSLDBxxx subsets
are no longer shipped with DIGITAL UNIX Version 4.0C and later releases.
-
The debugger shipped in this kit is fully integrated under DEC FUSE
V3.1.
-
C++ Dynamic Type Evaluation has been added.
-
Watchpoint support is available on DIGITAL UNIX Version 4.0 and later.
-
The C++ Version 5.6 compiler has been enhanced to reduce drastically
the amount of debugging information included in the object files that
it produces. This will result in significant improvements in startup
time for Ladebug when debugging C++ programs compiled with this
compiler. See the release notes supplied with C++ Version 5.6 for full
details.
-
The method that the debugger uses to look up symbols (program
variables) has been enhanced.
-
The debugger now supports programs that use Thread Local Storage.
-
A "Search Source File" dialog box has been added to the GUI.
-
The debugger now fully supports the Fortran REAL*16 and C long double
data types.
-
A "Quick Help" feature has been added to the graphical user
interface. The "Quick Help" panel is located at the bottom of the
Main Window. When enabled, it displays one line of context-sensitive
help when you drag the pointer over pull-down or pop-up menus.
-
Ladebug now provides an interface for editing long command lines
similar to that provided by the Korn shell
(
ksh).
-
A number of new built-in aliases have been added.
-
Bugs have been fixed.
See the Ladebug release notes for more information on these and
other new features and fixed problems.
1.34 Profiling and Memory Analysis Added to Developers' Toolkit
The Graphical Program Analysis (GPA) tools Version 1.0 is a new set of
standalone tools available from the Developers' Toolkit, an optional
part of the DIGITAL UNIX operating system. The initial GPA toolset
consists of three standalone tools:
-
Profiler
(
dxprof)
The Profiler gathers and analyzes run-time statistics on your program,
such as performance and resource use. It displays the information in
histogram and text format. The tool's Motif-based interface for
pixie,
prof,
gprof,
and
hiprof
makes the task of analyzing performance data much more accessible to
programmers.
-
Heap Analyzer
(
dxheap)
The Heap Analyzer helps you find errors in your program's use of
memory, including memory leaks. It displays the information in
histogram and text format. The tool's Motif-based interface for the
Third Degree, an ATOM-based memory leak detection tool, makes the task
of analyzing memory use much more accessible to programmers.
-
Man Page Browser
(
dxman)
The Man Page Browser displays reference pages in a scrollable
hypertext window. Its graphical interface lets you search for strings
or expressions in reference pages, navigate to other reference pages
by clicking on hotspots, copy text to other windows, and print
reference pages directly.
The kit and instructions for installing GPA are located on the
DIGITAL UNIX V4.0D Associated Products, Volume 1
CD-ROM. You must be running DIGITAL UNIX Version 4.0 or higher to
use the tools.
Once GPA is installed, if you are running the Common Desktop
Environment (CDE), you can invoke the GPA tools in the following
manner:
-
Click on the CDE Applications Manager.
-
Click on the Developers_Toolkit icon.
-
Click on the GPA icon.
-
Click on the icon for the GPA tool you would like to use.
If you do not use CDE, you can use the executable names previously
listed to invoke the tools from a command line. For example:
#
dxprof
1.35 Netscape Navigator Gold Version 3.04
This release of DIGITAL UNIX contains Version 3.04 of the Netscape
Navigator Gold World Wide Web browser. For information about how to
install, configure, and invoke Navigator, see
Section D.1
of this book and Chapter 6 of the DIGITAL UNIX
Installation Guide.
1.36 Netscape FastTrack Server Version 2.01
This release of DIGITAL UNIX contains Version 2.01 of the Netscape
FastTrack Server, an easy-to-use entry-level Web server designed to
let you create and manage a Web site. For information about how to
install and configure FastTrack, see
Section D.2.
1.37 Java Development Kit Version 1.1.4
This release of DIGITAL UNIX includes the Java Development Kit (JDK).
The JDK is a port to DIGITAL UNIX of the Java Development Kit
delivered by JavaSoft and made available for Solaris and Windows NT.
It includes all components of that kit. It also includes a
just-in-time (JIT) compiler for improved run-time performance.
For more information, see the installed JDK documentation at:
/usr/share/doclib/java/index.html
For the most recent information, including information on upgrades
to the JDK, see the following web page:
http://www.digital.com/java/
1.38 Performance Manager Version 4.0D
A new maintenance release of Performance Manager, Version 4.0D, is
being delivered with DIGITAL UNIX Version 4.0D. A change in
version numbering protocol occurred after Version 2.1A, resulting in a
jump to Version 4.0B to reflect synchronization with the current
release of DIGITAL UNIX. This version, 4.0D, directly follows 4.0B
in the sequence.
Performance Manager Version 4.0D includes the following enhancements:
-
Dash node name support
-
FDDI network support
-
Support for TruCluster Software Products Version 1.5 configurations
-
Documentation available in Adobe Acrobat (pdf) format
If you are upgrading from any of the Version 2.x releases, you will
also benefit from the following enhancements that were introduced in
Version 4.0B:
-
More features for the graphical user interface, including a toolbar
-
Support for monitoring thresholds that have been exceeded for specified
metrics
-
Support for invoking commands and user-supplied scripts once a
threshold has been crossed
-
More system management and thresholding scripts
-
Performance analysis, system management scripts, and cluster analysis
-
Per-process and per-thread metrics
-
Oracle7 database support
The release notes for Performance Manager are included on the
DIGITAL UNIX V4.0D Associated Products, Volume 2
CD-ROM.
The PostScript file is
PMGR425_RELNOTES.ps
and the text file is
PMGR425_RELNOTES.txt.
1.39 NetWorker Save and Restore Version 4.4
DIGITAL UNIX Version 4.0D contains a new release of NetWorker Save
and Restore, Version 4.4. NetWorker Save and Restore is a graphical
utility that backs up and recovers files. You can run it in one of two
modes: the limited feature NetWorker SingleServer product or the full
feature NetWorker Save and Restore product:
-
Networker SingleServer
The NetWorker SingleServer product allows you to back up local files on
a single local system to a local tape. NetWorker SingleServer is
licensed free of charge with the DIGITAL UNIX base operating system.
It can be easily upgraded to provide backup and archive services to a
group of systems in a heterogeneous environment.
-
NetWorker Save and Restore
The full NetWorker Save and Restore product allows you to back up and
restore files in a distributed environment. NetWorker Save and
Restore consists of both server and client software. You must
register a NetWorker license PAK in the license database to use
the full product. Additional prerequisite or optional software
licenses may also be required.
NetWorker Save and Restore Version 4.4 adds support for cluster server
failover. In the Available Server Environment (ASE), you can configure
the cluster members to allow the NetWorker server to migrate between
the members by defining NetWorker as a highly available application.
By installing the NetWorker Server on each cluster member and
configuring cluster members as NetWorker servers, NetWorker will
automatically fail over as directed by the ASE manager.
The following NetWorker subsets are available on the
DIGITAL UNIX V4.0D Associated Products, Volume 2
CD-ROM. All subsets are applicable to NetWorker SingleServer except
for
BRXCKIT440:
-
Release Notes and Documentation
(
BRXRNOTES440)
-
Server for DIGITAL UNIX
(
BRXSOAKIT440)
-
DIGITAL UNIX Client
(
BRXCKIT440)
-
Reference Pages
(
BRXSMAN440)
See the
NetWorker Save and Restore for DIGITAL UNIX Software Product Description
(SPD 50.98) for a complete description of features.
1.40 SCSI CAM Layered Components
SCSI CAM Layered Components, which provides the media changer and optical
disk drivers, is newly provided with DIGITAL UNIX on the
DIGITAL UNIX V4.0D Associated Products, Volume 2
CD-ROM as a convenience to users of tape and optical libraries. It
will continue to be provided in the
Software Product Library
as well.
The minimum version of SCSI CAM Layered Components that can be used
with DIGITAL UNIX Version 4.0D is Version 3.1B. The version
provided on the
Associated Products
CD-ROM will always be appropriate for the version of DIGITAL UNIX
with which it is shipped.
1.41 Open3D for DIGITAL UNIX
The DIGITAL Open3D layered product, which provides 2D and 3D (OpenGL)
device support for graphics options on workstations, is newly provided
with DIGITAL UNIX on the
DIGITAL UNIX V4.0D Associated Products, Volume 2
CD-ROM as a convenience to workstation system users with 3D-capable
graphics. It will continue to be provided in the
Software Product Library
as well.
The minimum version of DIGITAL Open3D that can be used with
DIGITAL UNIX Version 4.0D is Version 4.6. The version provided on
the
Associated Products
CD-ROM will always be appropriate for the version of DIGITAL UNIX
with which it is shipped.
1.42 Advanced Server V4.0 for DIGITAL UNIX
The Advanced Server for DIGITAL UNIX (ASDU) product (SPD 61.56.xx)
provides seamless interoperability between DIGITAL UNIX servers,
Microsoft Windows NT servers, and Microsoft Windows clients.
The ASDU software enables a DIGITAL UNIX system to run the services
that make it appear as a Microsoft Advanced Server. Through the ASDU
software, DIGITAL UNIX resources are available to Microsoft users
without modification to their software.
Key features of the ASDU product include:
-
Windows NT Advanced Server Version 4.0 file, print, security, and
management services on a system running the DIGITAL UNIX operating
system software
-
A password utility that enables users to synchronize their Microsoft
and DIGITAL UNIX or Network Information System (NIS) passwords
-
Common Internet File System (CIFS) support
-
International language support
-
TruCluster Software Products support
-
Two complimentary licenses
The ASDU server is an evolution of the PATHWORKS Version 6.x for
DIGITAL UNIX (Advanced Server) product. The ASDU product provides
improvements to the PATHWORKS product such as support for mixed-case
and long file names and a seamless upgrade procedure. The upgrade
procedure enables you to transfer your PATHWORKS configuration, user
account, and share data files to restore your PATHWORKS (Advanced
Server) environment on the ASDU server.
The ASDU software and documentation are delivered on the
DIGITAL UNIX V4.0D Associated Products, Volume 2
CD-ROM. Two clients can use the complimentary licenses after the
installation and configuration the ASDU software. Additional ASDU
licenses can be purchased and loaded into the License Management
Facility (LMF) on the system where the ASDU software is installed.
You can learn more about the ASDU product by reviewing the ASDU
tutorial. You can view the tutorial online with a web browser by
opening the
index.html
file in the
/mnt-pnt/Advanced_Server/doc/html/tutorial directory on the
DIGITAL UNIX V4.0D Associated Products, Volume 2
CD-ROM.