OpenVMS VMS721H1_UPDATE-V0400 Alpha V7.2-1H1 UPDATE ECO Summary
TITLE: OpenVMS VMS721H1_UPDATE-V0400 Alpha V7.2-1H1 UPDATE ECO Summary
New Kit Date: 09-APR-2001
Modification Date: 22-MAY-2001
Modification Type: Updated links to ftp site. If this kit has already
been installed on your system, there is no need to
install it again.
NOTE: An OpenVMS saveset or PCSI installation file is stored
on the Internet in a self-expanding compressed file.
For OpenVMS savesets, the name of the compressed saveset
file will be kit_name.a-dcx_vaxexe for OpenVMS VAX or
kit_name.a-dcx_axpexe for OpenVMS Alpha. Once the OpenVMS
saveset is copied to your system, expand the compressed
saveset by typing RUN kitname.dcx_vaxexe or kitname.dcx_alpexe.
For PCSI files, once the PCSI file is copied to your system,
rename the PCSI file to kitname-dcx_axpexe.pcsi, then it can
be expanded by typing RUN kitname-dcx_axpexe.pcsi. The resultant
file will be the PCSI installation file which can be used to install
the ECO.
Copyright (c) Compaq Computer Corporation 2001. All rights reserved.
New Kit Date : 09-APR-2001
Modification Date: 07-MAY-2001
Modification Type: DOCUMENTATION: Technical Modification in problems
ADDRESSED IN VMS721H1_UPDATE-V0400
OP/SYS: OpenVMS Alpha
COMPONENT: System
SOURCE: Compaq Computer Corporation
ECO INFORMATION:
ECO Kit Name: VMS721H1_UPDATE-V0400
DEC-AXPVMS-VMS721H1_UPDATE-V0400--4.PCSI
ECO Kits Superseded by This ECO Kit: VMS721H1_ACRTL-V0100
VMS721H1_CLIUTL-V0100
VMS721H1_CPU2208-V0100
VMS721H1_CPU2308-V0100
VMS721H1_FIBRE_SCSI-V0300
VMS721H1_GRAPHICS-V0200
VMS721H1_LIBRTL-V0100
VMS721H1_LOADSS-V0100
VMS721H1_MANAGE-V0100
VMS721H1_MIME-V0100
VMS721H1_MOUNT96-V0100
VMS721H1_RMS-V0300
VMS721H1_SHADOWING-V0300
VMS721H1_SYS-V0200
VMS721H1_UPDATE-V0300
ECO Kit Approximate Size: 113424 Blocks
Kit Applies To: OpenVMS Alpha V7.2-1H1
System/Cluster Reboot Necessary: Yes
Rolling Re-boot Supported: Yes
Installation Rating: INSTALL_1
1 - To be installed by all customers.
Kit Dependencies:
The following remedial kit(s) must be installed BEFORE
installation of this kit:
None
In order to receive all the corrections listed in this
kit, the following remedial kits should also be installed:
None
ECO KIT SUMMARY:
An ECO kit exists for system executables on OpenVMS Alpha V7.2-1H1. This
kit addresses the following problems:
PROBLEMS ADDRESSED IN VMS721H1_UPDATE-V0400 KIT:
o The VMS721H1_UPDATE-V0400 kit is a consolidation of ECO kits
that have shipped since the release of the
VMS721H1_UPDATE-V0300 kit. There are no new changes in this
kit.
If you have installed all the kits listed under the heading
"KITS SUPERSEDED BY THIS KIT:" you also need to install the
VMS721H1_UPDATE-V0400 remedial kit.
The following sections describe the problems addressed by the
superseded kits, with the execption of the
VMS721H1_UPDATE-V0300 kit. Those problem descriptions can be
found in a separate file, [SYSHLP]VMS721H1_UPDATE-V0300.README
PROBLEMS ADDRESSED IN KIT VMS721H1_ACRTL-V0100:
o The Compaq C RTL functions which make calls to RMS have been
modified to better support ODS-5 disks. This is an extensive
set of changes throughout the RTL to support case
preservation, long file names, and non-standard file name
characters. The default behavior of the RTL has not changed.
The Compaq C RTL supports case preservation in file names on
ODS level 5 disks. This applies when creating or reporting
file names. By default, this feature is disabled. To enable
this feature define the logical name DECC$EFS_CASE_PRESERVE to
"ENABLE". If file names are all in upper case, define
DECC$EFS_CASE_SPECIAL to "ENABLE" to convert the names into
lower case when reporting the name in UNIX style. If file
names are not all in upper case, then the case is preserved.
Most C RTL functions now accept long OpenVMS-style file names
as arguments. For OpenVMS Alpha Version 7.2, some basic
Compaq C RTL I/O functions (creat, stat, and the functions
from the open family of functions) were enhanced to accept
long OpenVMS-style file names for an ODS-5 device.
In this ECO kit, all other Compaq C RTL functions, except
chdir and the functions from the exec family of functions,
were also enhanced to accept long OpenVMS-style file names for
an ODS-5 device.
All C RTL functions that accept or report full file
specifications will process file specifications up to 4095
bytes long, subject to the rules defined for the media format.
For file specifications in OpenVMS format, there are no
special restrictions. In situations where a full file
specification cannot be reported because the buffer is too
short, the function attempts to report the abbreviated name.
There are restrictions with UNIX file names. Names containing
special characters, such as multiple periods, caret, or
multinational characters, may be rejected. A function call
may report failure if the output buffer is not large enough to
receive the full name. For OpenVMS-style names, the reported
name would contain a FID-abbreviated name. There is no
representation of FID-abbreviated names defined for UNIX.
o The Compaq C RTL was enhanced to allow interpreting the
leading part of a UNIX-style file specification as either a
subdirectory name or a device name.
The translation of the UNIX-style file name "foo/bar" to the
OpenVMS-style file name "foo:bar" was the only translation in
previous versions OpenVMS. It remains the default.
To request translation of the UNIX-style name "foo/bar" to the
OpenVMS-style name "[.foo]bar", DEFINE the following logical
name to any value:
$ DEFINE DECC$DISABLE_TO_VMS_LOGNAME_TRANSLATION ENABLE
To return to the default file-name translation, DEASSIGN the
logical name. The logical name is checked only once per image
activation, not on a file-by-file basis.
o The file creation functions (fopen, open, creat) have been
corrected to not inherit file attributes if a directory by the
same name exists for the file being created. For example,
creating a file named "foo." in a directory which contained a
"foo.dir" would incorrectly inherit file attributes from this
directory file.
o The stat function no longer puts the wrong values into
st_?time members of the stat structure. More precisely,
during daylight savings time the values of st_?time are no
longer one hour ahead from file creation/modification time
reported by the $DIRECTORY command.
o Nonquoted command-line arguments passed to C and C++ programs
(argv arguments) can now optionally have their case preserved,
rather than being lowercased as in previous versions. By
default, this feature is disabled.
To enable this case preservation feature, define the logical
name DECC$ARGV_PARSE_STYLE to "ENABLE" and set the
process-level DCL parse style flag to "EXTENDED" in the
process running the program. Enabling this feature also
ensures that the image name returned in argv[0] is also
case-preserved.
o The time functions time, ftime, gettimeofday and getclock
would fail for a TDF (time differential factor) of -13 hours,
the value required in New Zealand. According to the U.S.
Naval Observatory information the valid range for the TDF is
-13.75 to +13, inclusive. The C RTL now allows this full
range.
o The _malloc64 function was corrected to allow initialization
of this function in a threadsafe manner. Prior to this
change, if the first call occurs after the program has created
multiple threads and more than one thread calls the _malloc64
routine, an access violation may occur.
o In the previous ECO kit, a change was introduced to the
function strptime() for XPG5-compatibility to handle
expressions for the 2-digit year format %y when the century
was not specified. The requirement is for 2-digit years in
the range 0-68 to be treated as 21st century and years in the
range 69-99 as 20th century. The 2-digit years in the range
69-99 were incorrectly being treated as 21st century.
o The execlp and execvp functions were unconditionally prefixing
the file specification of the image to be run in the child
process with "VAXC$PATH:" string. When passed a fully
qualified file specification, this would result in an invalid
file specification.
o The behavior of lseek() and fseek() at positions beyond the
end of data in a file were not conforming to the POSIX/ANSI
standard. The file was being physically null padded on disk
even though the standard only requires this null padding if
data is actually written at this point. In the case reported
by the user, seeking to the end of a file by using a large
offset would take a long time and may exhaust available disk
space. To get the POSIX behavior, you must define
DECC$POSIX_SEEK_STREAM_FILE as "ENABLE". The default behavior
is unchanged.
o The Compaq C RTL was enhanced to open all files for shared
access as if the "shr=del,get,put,upd" option was specified in
the open* or creat call. To enable this feature, define the
logical name DECC$FILE_SHARING to the value "ENABLE". The
value is case-insensitive and is checked only once per image
activation, not on a file-by-file basis.
o The Compaq C RTL was failing when the amount of data written
to a pipe exceeded the buffer size of that pipe. The RTL now
breaks the large write into a sequence of writes whose size is
the size of the underlying mailbox used to implement pipes.
o The opendir function now accepts ".." to be equivalent to both
"../" and "[-]".
o A problem was reported that the RTL could hang if exit were
called while threads were continuing to do I/O. Analysis
showed that the hang was caused by a thread holding the lock
necessary to close the file and the exit handler blocking on
this lock. The change implemented was to not attempt to close
files whose locks were currently in use.
o When a stream file is extended using ftruncate() it is
documented that the extended area will be filled with zeroes.
This is now done correctly. Prior to this change, the file
was not extended.
o The documentation of the select function states that it
returns either when a socket is ready to be read or written,
when the timeout period expires, or when exceptions occur. It
was not properly terminating when an exception occurred. This
problem was corrected by dynamically linking to and calling an
abort function in the socket library. A complete correction
to interrupting the select function will also require an eco
kit from the Compaq TCP/IP product which includes this new
abort function.
o The fstat function was corrected to not access violate when
called with stdin, stdout, stderr, or a directory
specification.
o The chdir function was not well behaved when the calling
program passed an argument which was a search list logical.
The chdir function was unconditionally changing the logical
sys$disk to the first member of the search list was
effectively dropping the remaining members. The change in
behavior is seen as a step in the right direction, but there
are plans for more work in this area.
o The stat functions were not correctly handling files which
were not directories but did use a .dir extension. This
problem was discovered when porting a unix application which
used a .dir extension to store application specific
information.
o The RTL functions which process file specifications were not
correctly processing files which contained a hyphen character.
The RTL was incorrectly assuming that hyphens were only
allowed to specify a relative directory path.
PROBLEMS ADDRESSED IN VMS721H1_CLIUTL-V0100 KIT:
o Issuing two different SET DEVICE commands can crash a system
with a SSRVEXCEPT bug check. An excerpt of the crash dump
follows:
Crashdump Summary Information:
------------------------------
Bugcheck Type: SSRVEXCEPT, Unexpected system service
exception
Current Process: STARTUP
Current Image: $1$DKB406:[SYS0.SYSCOMMON.][SYSEXE]SET.EXE
Failing PC: FFFFFFFF.8018C338 NSA$SIZE_NSAB_C+000D8
Failing PS: 00000000.00000203
Module: SECURITY (Link Date/Time: 29-MAR-2000
00:55:08.31)
Offset: 0000E338
Images Affected: [SYSEXE]SET.EXE
o A DCL (Digital command language) $ SHOW CLUSTER command will
hang on startup on cluster-nodes with many SCS ports, e.g.,
CIPCA, CIXCD, MC:... or Pxx0: devices, if the
virtual-circuit (PB$ path-block) count exceeds four times the
SCS-node count. An "SCS-node" is defined as any SCS-speaking
node, such as HSJxx/HSCxx Storage Controllers, RFxx DSSI disk
controllers, or VMS nodes, etc.
Images Affected: [SYSEXE]SHWCLSTR.EXE
PROBLEMS RELATED TO RMS ADDRESSED IN VMS721H1_CLIUTL-V0100:
o Fix for interaction between SET RMS_DEFAULT /CONTENTION_POLICY
and /QUERY_LOCK settings.
The process and system defaults for these two qualifiers are
correct when the system is initially booted and will remain so
if neither setting is changed. Any setting of the
/CONTENTION_POLICY works correctly for the CONTENTION_POLICY
value; however, it also changes the QUERY_LOCK setting --
generally to disable. Since the QUERY_LOCK disable setting is
ignored by RMS unless both the RAB$V_NLK and RAB$V_RRL record
unlocking options are set within the application, this is not
expected to have a negative impact. However, any users trying
to change or reset the /QUERY_LOCK setting itself will find
they are not able to.
In order to change either or both of the settings, the
workaround until this kit is applied is to always enter both
/CONTENTION_POLICY and /QUERY_LOCK in the same SET RMS_DEFAULT
command.
For example, if the user wishes to change just the
CONTENTION_POLICY system value to "always" and leave the
QUERY_LOCK default as "enable," the following command should
be entered:
$ SET RMS_DEFAULT/SYSTEM/CONTENTION=ALWAYS/QUERY=ENABLE
Or if the user wishes to change QUERY_LOCK to "disable" and
leave the default setting for CONTENTION_POLICY:
$ SET RMS_DEFAULT/SYSTEM/CONTENTION=NEVER/QUERY=DISABLE
To restore the settings to the default system values, the
following command should be entered:
$ SET RMS_DEFAULT/SYSTEM/CONTENTION=NEVER/QUERY=ENABLE
Images Affected: [SYSEXE]SET.EXE
o New qualifier /SHARE added to SET FILE.
The /SHARE qualifier allows you to enable or disable global
buffers or statistics on a file currently being accessed by
other users. Only new accessors of the file will acquire the
new settings.
This qualifier requires SYSPRV (system privilege).
The /SHARE qualifier is only valid with the following
qualifiers:
/[NO]GLOBAL_BUFFER=count/SHARE
/[NO]STATISTICS/SHARE
If other qualifiers are needed, they must be entered in a
separate SET FILE command.
/[NO]GLOBAL_BUFFER=count/SHARE
The /SHARE qualifier can be used to enable or disable global
buffers on a file currently being accessed. However, any new
global buffer settings will only be applied to new accessors
of the file. And if a file is already opened with global
buffers, any new number of global buffers will not take effect
until the file is closed by all accessors of the file.
/[NO]STATISTICS/SHARE
The /SHARE qualifier can be used to enable or disable
statistics on a file currently being accessed. However, only
statistics of new accessors of the file will be measured.
EXAMPLE:
$ SET FILE/SHARE/GLOBAL_BUFFER=5000/STATISTICS INVENTORY.IDX
This example sets 5000 global buffers on the INVENTORY.IDX
file and enables statistics. Without the /SHARE qualifier, if
the file was open when the SET FILE command was issued, the
following error would be returned: SYSTEM-W-ACCONFLICT (file
access conflict). The /SHARE qualifier allows the global
buffers and statistics to be enabled on an open file.
However, these settings only apply to new accessors of the
file.
Images Affected: [SYSEXE]SET.EXE
PROBLEMS ADDRESSED IN VMS721H1_CPU2208-V0100 KIT:
o Allow mixed CPU operation in support of field replacements.
This allows field replacement of CPUs without having to
replace all the CPUs in the system.
Images Affected: [SYS$LDR]SYS$CPU_ROUTINES_2208.EXE
PROBLEMS ADDRESSED IN VMS721H1_CPU2308-V0100 KIT:
o After installation of the VMS721H1_FIBRE_SCSI-V0200 ECO kit,
GS80, GS160 and GS320 configurations using PKQDRIVER to
support the system disk will not boot on soft partitions other
than 0. When booting you will see error messages similar to
the following:
%PKQDRIVER-W- PKB0, pkq_alloc_queues: REQ/RSPQ Ring Buffer BAP
Registration Failed
%PKQDRIVER-W- PKB0, pkq_alloc_queues: Run AUTOGEN With Feedback
To Set NPAG_BAP_* Params
Note that this problem cannot be fixed by running AUTOGEN. To
fix this issue you must install this remedial kit.
DEC-AXPVMS-VMS721H_CPU2308-V0100--4.PCSI
Images Affected: [SYS$LDR]SYS$CPU_ROUTINES_2308.EXE
[SYS$LDR]SYS$SMHANDLER_SLAVE_2308.EXE
PROBLEMS ADDRESSED IN VMS721H1_FIBRE_SCSI-V0300 KIT:
o After installation of the VMS721H1_FIBRE_SCSI-V0200 ECO kit,
GS80, GS160 and GS320 configurations using PKQDRIVER to
support the system disk will not boot on soft partitions other
than 0. When booting you will see error messages similar to
the following:
%PKQDRIVER-W- PKB0, pkq_alloc_queues: REQ/RSPQ Ring Buffer BAP
Registration Failed
%PKQDRIVER-W- PKB0, pkq_alloc_queues: Run AUTOGEN With Feedback
To Set NPAG_BAP_* Params
Note that this problem cannot be fixed by running AUTOGEN. To
fix this issue you must install the following remedial kit:
DEC-AXPVMS-VMS721H_CPU2308-V0100--4.PCSI
PROBLEMS ADDRESSED IN VMS721H1_FIBRE_SCSI_V0200 KIT:
o PKQ_WATCHDOG would lose timed out I/Os, which would cause
SHADDETINCON or SHADBOOTFAIL bugchecks or cause other system
hangs.
Images Affected: [SYS$LDR]SYS$PKQDRIVER.EXE
o The system crashes with an INCONSTATE bug check in
SYS$FGEDRIVER while performing FibreChannel I/O. See
crashdump summary below:
Crashdump Summary Information:
------------------------------
Bugcheck Type: INCONSTATE, Inconsistent I/O data base
Current Process: NULL
Current Image:
Failing PC: FFFFFFFF.9B1187DC SYS$FGEDRIVER+087DC
Failing PS: 38000000.00000804
Module: SYS$FGEDRIVER (Link Date/Time: 19-AUG-2000
02:12:10.74)
Offset: 000087DC
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o Prevent loss of access to disks during a DRM failover.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o Modify command timeout setup code to avoid setting a value
that will cause a command to be timed out prematurely.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o In KERNEL$ABORT_EXCHANGE, add several checks to determine
whether or not a hard reset is in progress to avoid issuing a
READ_XRI command after the adapter has been reset.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o In FG$WRITE_IOCB, initialize FCCD$IS_DUE_TIME to zero before
optionally setting it based on a timeout value set by the
caller. This will prevent the use of a stale timeout value on
un-timed requests.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o The Emulex LPx000 ASCIC can incorrectly determine that it has
seen a parity error on the PCI and assert SERR, which will
crash the system. This is only an issue when it is recovering
from an adapter reset.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o The system can crash with an INCONSTATE bugcheck in
SYS$PGADRIVER, if there is a change in the Fabric, such as
pulling a cable on the switch.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o During DRM testing, disks from an HSG80 became unavailable.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o The system crashes with an INVEXCEPTN bugcheck at
@SYS$FGEDRIVER+0A2A0.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o A 126-block stale read occurred on HSG80/60 disks. See "STALE
DATA' pattern below:
STALE DATA BLOCKS 1-126 (FULL XFER: 3 EVENTS):
All DGAxxx disks are HSG80 (repeat HSG80-based) !!
STATE DATA BLOCKS 17-126 (PARTIAL STALE XFER: 1
EVENT): Single DGAxxx disk is HSG60-based !!
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o The system crashes when an Emulex ASIC incorrectly detects a
PCI parity error and asserts SERR on reset recovery.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o A system can crash with an INVEXCEPTN bugcheck during error
processing.
Crashdump Summary Information:
------------------------------
Bugcheck Type: INVEXCEPTN, Exception while above ASTDEL
CPU Type: AlphaServer 8200 5/625
Current Process: BATCH_4506
Current Image: $28$DUA47:[PRODPROG.STANDARD]
DATAPROC_ALP.EXE
Failing PC: FFFFFFFF.8F8735C4 SYS$DKDRIVER+0B5C4
Failing PS: 18000000.00000804
Module: SYS$DKDRIVER (Link Date/Time: 14-SEP-2000
00:27:05.42)
Offset: 0000B5C4
Images Affected: [SYS$LDR]SYS$DKDRIVER.EXE
PROBLEMS ADDRESSED IN VMS721H1_FIBRE_SCSI-V0100 KIT:
o The SCSI port goes off-line with a FATAL controller error.
See a partial console log below:
Console log:
************
%PBA2 CPU04: 02-MAY-2000 09:00:39 SMCI channel online
%EIA0, Auto Negotiation mode set by console
%EIB0, Auto Negotiation mode set by console
%EIC0, FastFD(Full Duplex 100BaseFX) mode set by console
%EIB0, FastFD(Full Duplex 100BaseTX) Ethernet connection
selected
%EIA0, FastFD(Full Duplex 100BaseTX) Ethernet connection
selected
%STACONFIG-I-LANACP, creating the LANACP process
%STACONFIG-I-LANACPWAIT, waiting for LANACP to exit
%STACONFIG-I-LANACPEXIT, LANACP has exited
%STACONFIG-I-PE, loading PEDRIVER
%STACONFIG-I-PEDONE, return from PEDRIVER load
%STACONFIG-I-SCSI, loading SCSI class/port drivers
%CNXMAN, Sending VMScluster membership request to system
QTV9
%PBA0 CPU04: 02-MAY-2000 09:00:50 SMCI channel online
%CNXMAN, Now a VMScluster member -- system WILD4
%SYSINIT-I-LOCK, taking out lock on system device
%PKQDRIVER-W- PKD0, FATAL Controller error; port marked
off-line
DSA4095:
%SYSINIT-I-GALAXY, Initializing community lock ID
%SYSINIT-I-DEFINE, defining system logical names
%SYSINIT-I-OPEN, opening PAGEFILE.SYS
%SYSINIT-I- PAGEFILE.SYS not found - system initialization
continuing...
Images Affected: [SYS$LDR]SYS$PKQDRIVER.EXE
o Multiple errors occur when running the PKQDRIVER in the HSOF
distribution media. Re-installing a PKQDRIVER older version
of HSOF eliminates the problem(s).
Images Affected: [SYS$LDR]PKQDRIVER.EXE
o The system can hang while either crashing or booting.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
[SYS$LDR]SYS$PGADRIVER.EXE
o When running low on map registers PKQDRIVER will crash, incur
alignment faults and corrupt the erase page, which causes lock
manager crashes.
Images Affected: [SYS$LDR]SYS$PKQDRIVER.EXE
o During a boot, if the connection limit on the FibreChannel
storage unit is reached, e.g., HSG80 has a limit of 64
entries, the driver configuration will fail without any error
message. If the system is trying to boot from a FibreChannel
disk on one of these storage devices, the whole system will
hang during the boot.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
[SYSMSG]SYSMSG.EXE
o An INCONSTATE bugcheck occurred when a mailbox command
returned with a DMA error.
Images Affected: [SYS$LDR]FGEDRIVER.EXE
o The system may crash whenever MSCP multipath failover is used.
Images Affected: [SYS$LDR]DKDRIVER.EXE
[SYS$LDR]MKDRIVER.EXE
o Data corruption can occur when running four 16GB hard
partitions.
Images Affected: [SYS$LDR]SYS$DKDRIVER.EXE
[SYS$LDR]SYS$FGEDRIVER.EXE
o The system can hang when booting from a FibreChannel system
disk.
Images Affected: [SYS$LDR]SYS$DKDRIVER.EXE
[SYS$LDR]SYS$FGEDRIVER.EXE
o SYS$DKDRIVER hangs all $1DGx devices connected to it.
Images Affected: [SYS$LDR]SYS$DKDRIVER.EXE
[SYS$LDR]SYS$FGEDRIVER.EXE
o An INCONSTATE bugcheck occurs when the adapter is being reset.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
o The system crashed because the fork lock index in the TQE was
no longer valid.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
[SYS$LDR]SYS$PGADRIVER.EXE
o A hard hang was introduced with VMS721H1_FIBRE-V0300 kit.
Images Affected: [SYS$LDR]SYS$FGEDRIVER.EXE
[SYS$LDR]SYS$PGADRIVER.EXE
NEW FUNCTIONALITY ADDRESSED IN VMS721H1_GRAPHICS-V0200 KIT:
o This kit provides support for the PowerStorm 300/350 Graphics
adapter. Specific information about this support can be found in
the P300_350_README file. This file is available in the following
formats:
+ Text : P300_350_README.TXT
+ Acrobat Reader : P300_350_README.PDF
+ HTML : P300_350_README.HTML
These files can be found at SYS$COMMON:[SYSHLP]
NOTE: The PowerStorm 350 Graphics adapter is not currently
supported on GS80, GS160 and GS320 systems. This
information and functionality is included in this
kit for Cluster Compatibility purposes for other
machines that may be running VMS V7.2-1H1.
o A new device dependent X server image is provided to improve
performance on EV6 (21264) based platforms containing ELSA GLoria
Synergy graphics modules. A new procedure,
SYS$MANAGER:DECW$DEVICE_CONFIG_P2.COM, is executed automatically
during DECwindows startup. The new procedure will, on appropriate
platforms, define a logical name that will redirect the server to
use the new P2 device dependent X code, DECW$SERVER_DDX_P2.EXE.
DECW$SERVER_DDX_GZ.EXE will continue to be used on older platforms
containing ELSA GLoria Synergy graphics modules.
Image(s) Affected: [SYSLIB]DECW$SERVER_DDX_P2.EXE
P350 PROBLEMS ADDRESSED IN THE VMS721H1_GRAPHICS-V0200 KIT:
o With PowerStorm 300/350 graphics adapter(s), XGetImage on
8-plane visuals returned incorrect image data if the specified
width was not a multiple of eight.
Images Affected: [SYSLIB]DECW$SERVER_DDX_GB.EXE
o On systems containing PowerStorm 300/350 graphics adapter(s),
under certain conditions of 3D-application load, the address
space of the DECwindows server would grow without limit,
eventually exhausting the supply of page table entries (PTEs)
resulting in a "not enough core" message being displayed.
Images Affected: [SYSLIB]DECW$SERVER_DIX.EXE
[SYSLIB]DECW$OPENGLPEERRI.EXE
o Failure to create a 3D rendering context (e.g. due to
resource exhaustion) is no longer a fatal server error with
PowerStorm 300/350 adapter(s).
Images Affected: [SYSLIB]DECW$OPENGLPEERRI.EXE
[SYSLIB]DECW$SERVER_DDX_GB.EXE
o PowerStorm 300/350 multi-head initialization problems on
systems running at 667 MHz or faster have been solved.
Images Affected: [SYS$LDR]SYS$GBDRIVER.EXE
o With PowerStorm 300/350 adapter(s), the Store Colors operation
on 8-plane PseudoColor windows could result in incorrect
colors if the array of color values did not denote a
consecutive range of pixel values.
Images Affected: [SYS$LDR]SYS$GBDRIVER.EXE
o The 1600 x 1200 @ 60 Hz video mode now works with PowerStorm
300/350 adapter(s).
Images Affected: [SYS$LDR]SYS$GBDRIVER.EXE
o Fast systems with PowerStorm 300/350 adapter(s) and large
memory (667 MHz and above) sometimes crash when graphics
operations are attempted while the system is already handling
a heavy load of non-graphics I/O.
Images Affected: [SYS$LDR]SYS$GBDRIVER.EXE
o A system crash due to a page fault with the IPL set too high
has been fixed in the PowerStorm 300/350 driver.
Images Affected: [SYS$LDR]SYS$GBDRIVER.EXE
o Xserver on Elsa Gloria Synergy crashes with the following
message in DECW$SERVER_0_ERROR.LOG:
Images Affected: [SYSLIB]DECW$SERVER_DDX_P2.EXE
o Clipped text was incorrectly displayed with ELSA GLoria
Synergy graphics modules. This occurred when scrolling
horizontally within a text box (e.g., using Netscape). When
enough text is typed into a text box to cause the field to
scroll to the left, the text is corrupted or not displayed.
Images Affected: [SYSLIB]DECW$SERVER_DDX_GZ.EXE
o A 3rd party middleware product called GMS has missing text on
the ELSA Gloria Synergy card.
Images Affected: [SYSLIB]DECW$SERVER_DDX_GZ.EXE
o An invalid exception ("INVEXCEPTN, Exception while above
ASTDEL") bugcheck will sometimes occur while booting with an
ELSA GLoria Synergy card installed.
Images Affected: [SYS$LDR]SYS$GZDRIVER.EXE
[SYS$LDR]SYS$GZVDRIVER.EXE
o During startup, DECwindows prints an error saying that
decw$fs-errors.log cannot be purged from SYS$SYSTEM.
o A memory leak in the SYS$WSDRIVER (the pseudo display device
driver) has been fixed.
Images Affected: [SYS$LDR]SYS$WSDRIVER.EXE
o The server crashed when no TCP/IP license was available.
Images Affected: [SYSLIB]DECW$TRANSPORT_TCPIP.EXE
o Multiple network interfaces are now supported.
Images Affected: [SYSLIB]DECW$TRANSPORT_TCPIP.EXE
o The following font server problems have been fixed:
+ The X11 server crashed with %SYSTEM-F-ASTFLT if given an
unavailable Multiple network interfaces are now
supported.server node.
+ The X11 server would hang if the font server aborted or
restarted, or if there were significant network delays
enroute to/from the font server.
+ The X11 server crashed if a client with an active font
server request terminated.
+ Font server connection problems were not reported in the
X11 server log.
+ The font server sometimes stopped accepting connections.
+ The font server crashed when clients using DHCP made
requests.
+ Font server clients experienced problems if the font
server was idle for approximately ten minutes or longer.
+ The font server could not be restarted until five minutes
after it had terminated; it would report "address in use"
for the TCP/IP socket.
Images Affected: SYS$LIBRARY:DECW$SERVER_DIX.EXE
SYS$SYSTEM:DECW$XFS.EXE
SYS$SYSTEM:FSLSFONTS.EXE
SYS$SYSTEM:FSINFO.EXE
PROBLEMS ADDRESSED IN VMS721H1_LIBRTL-V0100 KIT:
o A memory leak occurs when using
LIB$FIND_FILE/LIB$FIND_FILE_END with long file names
(flags=LIB$M_FIL_LONG_NAMES).
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
o A synchonization failure occurs for multithread programs using
the DEC C RTL routine malloc and free with "long" pointers (P2
space heap routines). These DEC C RTL routines in turn use
the LIBRTL Routines LIB$VM_MALLOC_64 and LIB$VM_FREE_64.
Symptoms could be varied, including "%LIB-F-MEMALRFRE, memory
already free" errors.
For an application to see this problem it would have to have
the following characteristics:
1. Call the DEC C RTL routines malloc and free
2. Be multithreaded
3. Be compiled with the /POINTER_SIZE=LONG and
/REENTRANCY=MULTITHREAD switches
4. Be linked
/THREADS_ENABLE=(MULTIPLE_KERNEL_THREADS,UPCALLS)
5. Run on a multiprocessor system with kernel threads
enabled.
Note that programs using the "normal" P0 heap routines are not
affected by this problem.
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
o The LIB$CREATE_DIR RTL routine is documented to accept 64 bit
parameters. However, passing a 64 bit descriptor to the
routine fails with a LIB$_INVSTRDES error or passing a 64 bit
address may result in an SS$_ACCVIO error.
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
o A program can fail with the following error:
%LIB-W-ILLRECTYP, illegal record type (nnn)
%TRACE-W-TRACEBACK, symbolic stack dump follows
followed by:
%SYSTEM-F-ACCVIO, access violation, reason mask=00,
virtual address=0000000000000008, PC=nnnnnnnnnnnnnnnn,
PS=0000001B
if the program:
1. Is created that is directly linked with a shareable image
2. Calls LIB$FIND_IMAGE_SYMBOL to find an entry point in the
same shareable image
3. The program is executed and the user creates a new version
of the shareable image while the program is executing
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
o LIB$FIND_IMAGE_SYMBOL fails to locate symbols for images whose
record attributes have been altered.
This has been seen with DCOM as follows:
$ mcr DCOM$COMREGEDT
%SYSTEM-F-ACCVIO, access violation, reason mask=00,
virtual address=0000000000000008, PC=0000000000B43348,
PS=0000001B
Improperly handled condition, image exit forced.
Signal arguments: Number = 0000000000000005
Name = 000000000000000C
0000000000000000
0000000000000008
0000000000B43348
000000000000001B
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
o LIB$FIND_IMAGE_SYMBOL fails for images on disks whose device
names are more than 13 characters - which generally means a 3
digit unit number.
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
o In the case of a $ PRINT [X]FOO.PRF,;-1 command, LIB$FIND_FILE
may fail to find the second file (;-1). This problem may be
seen with certain DCL commands, such as PRINT, and with user
application calls to LIB$FIND FILE when LIB$M_FIL_LONG_NAMES
and the "multiple" flags are set (bits 1 and 2 of the flags
argument).
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
o A failure to unwind properly occurs when a large amount of
stack space is being used. This could result in an improperly
handled condition error, or a bad stack notification, or other
error.
Images Affected: [SYSLIB]LIBRTL.EXE
[SYSLIB]LIBRTL_INSTRUMENTED.EXE
PROBLEMS ADDRESSED IN VMS721H1_LOADSS-V0100 KIT:
o When used by a user who has more than 124 rights identifiers,
a $PERSONA_CREATE requests fail with a BADBLOADR error.
Images Affected: [SYSLIB]SECURESHR.EXE
PROBLEMS ADDRESSED IN VMS721H1_MANAGE-V0100 KIT:
o A GALAXY license server experiences an EXQUOTA error which
prevents it from starting.
Images Affected: [SYS$STARTUP]VMS$CONFIG-050_LMF.COM
PROBLEMS ADDRESSED IN VMS721H1_MIME-V0100 KIT:
o Memory corruption occurred when using the OpenVMS MIME utility
as evident when repeatedly opening and closing UUencoded
files.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o When composing a simple, single-part MIME text message with no
attachments, the READ command would display properly until the
file was saved, closed, and re-opened for read. Then, extra
headers for a text attachment would appear in the MIME header
area, which resulted in an error:
%MIME-E-OPNFILERR, error opening file
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o When using MIME to compose messages, the resulting MIME file
may not have its MIME message Boundary parameter string
properly quoted. The string is part of the 'Content-Type:'
multipart-mixed header field. Consequently, the MIME message
may not be read correctly.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o An ANALYZE/RMS error would occur if used on the zero allocated
block file that was created, but never written to, by
composing a MIME 'NEW' Message.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o Attachments can be removed, but not saved for READ only MIME
messages.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o An ACCVIO can occur when MIME attempts to display the VIEWERR
error message. This occurs when a MIME file is opened for
which no viewer exists.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o Multiple memory leaks occur when repeatably opening and
closing UUencoded files. This excess use of memory resources
may not be readily apparent to the user.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o Extraneous files can be left behind by MIME, if the user
performs a QUIT command, or after prompting and answering 'no'
to a file save operation upon EXIT.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o The MIME command OPEN with the /NEW qualifier would fail.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o The MIME command OPEN would continue to parse for up to eight
files, if none were specified with the OPEN command. The
operation would usually fail with the following errors:
%RMS-E-FNF, file not found
%MIME-W-NOMSGOPEN, No currently active message
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o Multiple error messages with the same error would be returned,
if the file sought to be opened was not found.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o Informational messages may get overwritten when MIME exits
with the DCL (DIGITAL command language) prompt.
Images Affected: [SYSEXE]MIME.EXE
[SYSHLP]MIME$HELP.HLB
o Ensure that MIME$HELP.HLP files are consistent across all
streams. Remove the fold record in KESTREL_FT2 and
KESTREL_PLUS.
Images Affected: [SYSHLP]MIME$HELP.HLB
o Initialize VCB$L_BLOCKFACT to be at least 1, so that the XQP
does not incur 'divide by zero' and UNXSIGNAL bugchecks.
Images Affected: [SYSLIB]MOUNTSHR.EXE
[SYSEXE]VMOUNT.EXE
o When mounting a CD that was created from a disk that was
larger than the CD itself, a check that compares the volume
size, cluster factor and block count fails with a
%MOUNT-F-FILESTRUCT error.
Images Affected: [SYSLIB]MOUNTSHR.EXE
[SYSEXE]VMOUNT.EXE
o On certain devices, mounting bound volumes sets which are also
shadow set members can cause MOUNT-F-INCONSET and
MOUNT-I-DOVOLSET error messages to be generated.
Images Affected: [SYSLIB]MOUNTSHR.EXE
[SYSEXE]VMOUNT.EXE
o The /DENSITY qualifier is quietly ignored when used with some
tape devices. Many customers continue to use the /DENSITY
qualifier on tape drives that do not support any specific
density. Rather, they support various forms of compaction.
This has led to confusion about what the drive should or
should not be doing.
This change simply reports the SS$_IVDENS warning message if
/DENSITY was supplied and the device driver could not process
it correctly. Note that the MOUNT will continue and the
default compaction characteristics will be used.
Images Affected: [SYSLIB]MOUNTSHR.EXE
[SYSEXE]VMOUNT.EXE
PROBLEMS ADDRESSED IN VMS721H1_RMS-V0300 KIT:
o Fix to prevent a fatal INVEXCEPTN lock manager bugcheck when
global buffers are enabled.
This problem involves a race condition with the following rare
set of conditions occurring simultaneously:
+ The last accessor of a global section cleaning up any
cached global bucket system locks on one node.
+ A lock request on another node resulting in delivery of a
system fork lock blocking interrupt to the latter node
that attempts to convert a lock after it has been dequeued
by the last accessor.
Images Affected: [SYS$LDR]RMS.EXE
o Fix for invalid input parameter in call to SCH$QAST to prevent
a potential fatal INVEXCEPTN system bugcheck when global
buffers are enabled.
Images Affected: [SYS$LDR]RMS.EXE
o Fix for potential RMS lock hang when global buffers are
enabled.
This problem involves a rare condition if the last accessor of
the global section is interrupted by an abort rundown, which
may lead to the need for the cleanup of the global section not
being properly detected. This could result in the survival of
the global section and a global bucket system lock after the
last accessor is deleted.
Since this problem requires the accessor that is aborted to
also be the last accessor, it is more apt to occur with a file
that accessors come and go -- with a file that is opened and
closed frequently and may have only one accessor at a
particular time (e.g. RIGHTSLIST.DAT) -- versus a file with
global buffers enabled that typically has several concurrent
accessors at all times until there is an application (or
system) shutdown.
Images Affected: [SYS$LDR]RMS.EXE
o Fix to prevent a nonfatal RMS bugcheck (ENQDEQFAIL) when an
enqueue request fails with a CVTUNGRANT status.
The ENQDEQFAIL bugcheck may be reported when an RMS lock
manager request to rearm the blocking AST associated with a
directory cache fails with a CVTUNGRANT status.
The process may either:
1. Terminate (SYSGEN parameter BUGCHECKFATAL not enabled)
with an RMSBUG; or
2. Crash the system (BUGCHECKFATAL enabled) with a RMSBUG.
The footprint of the crash will reflect an R2 register
value of %XFFFFFFF4 (ENQDEQFAIL) with an R0 register value
of %X0000213c (SYSTEM-F-CVTUNGRANT).
Images Affected: [SYS$LDR]RMS.EXE
o Fix for SQO error on CONVERT/NOSORT with collated key.
Producing an indexed file with a collated key using the
qualifier /NOSORT with the CONVERT may fail with the following
error:
%CONV-F-READERR, error reading
-RMS-F-SQO, operation not sequential (SQO set)
Images Affected: [SYSEXE]CONVERT.EXE
[SYSLIB]CONVSHR.EXE
o RMS support for new qualifier /SHARE added to SET FILE, which
is dependent on the VMS721H1_CLIUTL-V0100 kit being installed.
The /SHARE qualifier allows you to enable or disable global
buffers or statistics on a file currently being accessed by
other users. Only new accessors of the file will acquire the
new settings. This qualifier requires SYSPRV (system
privilege).
The /SHARE qualifier is only valid with the following
qualifiers:
/[NO]GLOBAL_BUFFER=count/SHARE
/[NO]STATISTICS/SHARE
If other qualifiers are needed, they must be entered in a
separate SET FILE command.
/[NO]GLOBAL_BUFFER=count/SHARE
The /SHARE qualifier can be used to enable or disable global
buffers on a file currently being accessed. However, any new
global buffer settings will only be applied to new accessors
of the file. And if a file is already opened with global
buffers, any new number of global buffers will not take effect
until the file is closed by all accessors of the file.
/[NO]STATISTICS/SHARE
The /SHARE qualifier can be used to enable or disable
statistics on a file currently being accessed. However, only
statistics of new accessors of the file will be measured.
EXAMPLE:
$ SET FILE/SHARE/GLOBAL_BUFFER=5000/STATISTICS INVENTORY.IDX
This example sets 5000 global buffers on the INVENTORY.IDX
file and enables statistics. Without the /SHARE qualifier, if
the file was open when the SET FILE command was issued, the
following error would be returned: SYSTEM-W-ACCONFLICT (file
access conflict). The /SHARE qualifier allows the global
buffers and statistics to be enabled on an open file.
However, these settings only apply to new accessors of the
file.
Images Affected: [SYS$LDR]RMS.EXE
[SYSEXE]SET.EXE
[SYSLIB]DCLTABLES.EXE with modified SET.CLD
The latter two images require the installation of the
VMS721H1_CLIUTL-V0100 kit.
PROBLEMS ADDRESSED IN VMS721H1_RMS-V0200 KIT:
o Autoextend problem with shared write sequential file if global
buffers enabled.
When an autoextend occurs with a shared write sequential file
with global buffers enabled, if both readers and writers are
concurrently processing data at the end-of-file, a window
exists for the readers when either an infinite process loop
may occur or an inappropriate RMS-E-EOF or RMS-F-IRC error may
be returned. The likelihood of either problem occurring
increases significantly if the deferred-write (DFW) option is
enabled or the sequential file does not have a generous
initial allocation or extend quantity.
Images Affected: [SYS$LDR]RMS.EXE
o Fix for remote file transfer performance degradation if UPI
enabled.
As a side-effect of a DAP/FAL change made to the Alpha Version
7.2 family, the SQO (sequential only) option was
inappropriately disabled in a record stream when the UPI (user
provided interlock) option was enabled if RMS detected any
form of write sharing. This resulted in reducing the
performance of a remote file transfer that had the UPI option
enabled due to the overhead of sending the file
record-by-record rather than using the more efficient block
mode transfer supported by the SQO option.
Images Affected: [SYS$LDR]RMS.EXE
PROBLEMS ADDRESSED IN VMS721H1_RMS-V0100 KIT:
o Fix to prevent ACCVIO on put or update to an indexed file.
A LDQ (load quad) instruction generated by the Bliss-64
compiler (default quadword alignment) may result in an attempt
being made to read or write 4 bytes beyond the end of an RMS
internal buffer which ends exactly on a page boundary with an
adjacent inaccessible page.
The process may either:
1. Terminate (SYSGEN parameter BUGCHECKFATAL not enabled)
with an access violation (ACCVIO); or
2. Crash the system (BUGCHECKFATAL enabled) with a SSRVEXCEPT
ACCVIO.
There are too many variables (for example, bucket size, number
of buffers, number of keys, record location) to predict in
advance that a file is vulnerable to this problem. However,
we do know that the right combination of conditions must be
rare since the problem has been dormant for several releases.
It was introduced with the conversion of several Bliss-32 RMS
modules to Bliss-64 as part of the support for 64-bit
addresses added to Alpha V7.0.
o Fix for data access synchronization.
An unusual timing condition may cause an application to get
incorrect data returned due to an interaction between caching
and read record access. Global buffers must be enabled on the
file, and there must be several processes trying to
concurrently access a data bucket that has to be brought into
the global cache (i.e., is not already in the global cache).
This was uncovered doing internal stress testing with an SMP
(8 CPU) system with caching enhancements for a future release.
It was reproduced with very heavy contention to a data bucket
with sequential read access being done. We have not been able
to reproduce the problem without the speed of the caching
enhancements. It is unknown whether this problem could occur
in V7.2-1H1, but it is best to apply this update as a
precaution.
o Modification to the global buffer hashing interlock mechanism.
The hashing interlock mechanism, which was implemented in
V7.2, has been re-designed to eliminate the timer-based
strategy that was used during periods of high contention. A
system could become overburdened by Timer Queue activity (IPL
8) when the contention for an interlock reached a threshold.
This Timer Queue activity precluded the interlock owner from
releasing the interlock and resulted in excessive Timer Queue
activity consuming the primary CPU.
Images Affected: [SYS$LDR]RMS.EXE
[SYS$LDR]DDIF$RMS_EXTENSION.EXE
[SYSLIB]SDARMS$SHARE.EXE
o Fix to prevent a process hang waiting on an RMS lock when
reverse (key-less-than) search criterion is used on a file
with global buffers enabled.
Access to an RMS indexed file may hang indefinitely waiting on
a lock request when the following three conditions are met:
- Global buffers are enabled.
- Key-Less-Than match criterion is being utilized.
- The key of reference is switched between two adjacent
lookups (e.g., key of reference is switched from key 0 to
key 1).
This problem is fixed in the release of OpenVMS Alpha to
follow V7.2-1H1.
o Fix to prevent a rare SSRVEXCPT bugcheck during process
rundown.
This is a fix to prevent a SSRVEXCPT bugcheck during process
rundown when an RMS AST is delivered to the process after the
process address space has already been released. In order for
this to occur, the process must have been utilizing global
buffers. This issue represents a very small timing window and
should be extremely rare.
o Enhance performance for RU journaled files when reading by
secondary key.
Accessing a file by secondary key that has Recovery Unit (RU)
journaling enabled in which a large number of deletes has
occurred may incur an inordinate amount of direct IO activity.
This is due to RMS having to process the RU deleted entries in
the secondary key structure. This is not an issue when access
to the file is by the primary key.
This change is a performance enhancement in which RMS will now
perform cleanup of the RU deleted Secondary Index Data Record
entries when they are initially encountered through a read
operation. This will reduce the IO required to process these
entries for all future accesses. Access by the primary key of
the file already performs this optimization.
o Performance enhancement (record locking reduction) for
CONVERT/SHARE.
CONVERT/SHARE previously read the input primary data records
(record I/O) with the RAB$V_RRL (read regardless) option
enabled. As a performance enhancement, RAB$V_RRL was replaced
with the new RAB$V_NQL (no query locking) record option if
/SHARE is used. This results in no record locking being done
for all the reads of the primary data records from the input
file.
o Support for SET RMS_DEFAULT /CONTENTION_POLICY to address
locking fairness issues.
The new Alpha global buffer read-mode lock support introduced
in V7.2-1H1 is functionally compatible with both VAX and older
Alpha releases. Operations in mixed clusters produce correct
results. However, there is a locking fairness issue that may
arise with mixed cluster operations (e.g. an Alpha V7.1-2 or
V7.2-1 version or VAX versions).
In a mixed cluster environment with very high contention for
specific buckets, it is possible for accesses to write-shared
files on nodes using read-mode bucket locking to dominate
access to a bucket. Nodes without this support might be
unable to access the bucket for a protracted period of time.
While V7.2-1H1R does include the new global buffer read-mode
lock support and so will not be affected by the latter issue,
it is also possible to observe comparable behavior on all
OpenVMS versions when dealing with accesses to write-shared
files without global buffers enabled -- even on a standalone
system. A similar fairness issue between lock conversions and
new lock requests may be observed in which the new lock
requests may remain ungranted for an extended period of time.
This kit includes support in RMS for a new option to improve
fairness under high contention conditions for write-shared
files, but selecting this option may noticeably increase
locking overhead. The option may be set at a process or
system level. Since many applications will never encounter
this issue, the default system behavior leaves this option
disabled. A future lock management enhancement should make
this fairness workaround unnecessary for later releases.
The option is controlled using the /CONTENTION_POLICY
qualifier to the DCL command SET RMS_DEFAULT. The following
are valid PROCESS keywords (/SYSTEM not specified):
NEVER Never use the higher overhead option
to improve fairness for any write-shared
files accessed by this process; minimal
overhead.
SOMETIMES Use this option for fairer bucket
access (but higher overhead) to any
write-shared files with global buffers
enabled that are accessed by this process.
ALWAYS Use this option for fairer bucket
access (but higher overhead) to all
write-shared files accessed by this
process.
SYSTEM_DEFAULT (Default) Use system setting. Note
that this keyword is disallowed with
/SYSTEM.
The following are valid SYSTEM keywords (/SYSTEM specified):
NEVER (Default) Never use the higher
overhead option to improve fairness
for any write-shared files accessed
on the system; minimal overhead.
SOMETIMES Use this option for fairer bucket
access (but higher overhead) to any
write-shared files with global
on the system.
ALWAYS Use this option for fairer bucket
access (but higher overhead) to all
write-shared files accessed on the
system.
In addition to the RMS image, modifications to the following
images contained in the VMS721H1_CLIUTL-V0100 kit are required
in order for this functionality to take affect:
- [SYSEXE]SET.EXE
- [SYSMSG]CLIUTLMSG.EXE
- replacement of revised SET.CLD in [SYSLIB]DCLTABLES.EXE
Without these modified images, RMS will use NEVER as the
default for the CONTENTION_POLICY option.
PROBLEMS ADDRESSED IN VMS721H1_SHADOWING-V0300 KIT:
o The VMS721H1_SHADOWING-V0200 images would not install over the
VMS721H1_SHADOWING-V0100 images. The shadowing images in the
VMS721H1_SHADOWING-V0200 kit should have been the same images
as those in the VMS721H1_SHADOWING-V0100 kit. Instead, they
were from an earlier build than the images in the
VMS721H1_SHADOWING-V0100 kit.
Note that this is not an image functionality issue and no new
image functionality is provided by the
VMS721H1_SHADOWING-V0300. If you have installed the
VMS721H1_SHADOWING-V0200 kit you do not need to install the
VMS721H1_SHADOWING-V0300 kit
Images Affected: [SYS$LDR]SYS$SHDRIVER.EXE (new image)
[SYS$LDR]SYS$PLUS_SHDRIVER.EXE (new image)
[SYSEXE]SHADOW_SERVER.EXE (new image)
PROBLEMS ADDRESSED IN VMS721H1_SHADOWING-V0200 KIT:
o Correction for DCL SHOW display errors.
Either a user-mode ACCVIO or formatting display errors may
occur with various DCL SHOWs (e.g., SHOW DEVICE, SHOW MEMORY,
SHOW TERMINAL, SHOW WORK, SHOW NET) after the VAXSHAD07_071
TIMA kit has been installed.
Images Affected: [SYSMSG]CLIUTLMSG.EXE
NEW FUNCTIONALITY INCLUDED IN THE VMS721H1_SHADOWING-V0100 KIT:
o This kit contains new command features intended for use in
OpenVMS clusters that span multiple sites. An example of how
to apply the new commands will be found after they are
described.
o DISMOUNT
1. /FORCE_REMOVAL ddcu:
If connectivity to a device has been lost and the
virtual unit is in mount verification, this switch
causes the member to be expelled immediately from the
shadow set. If you omit this qualifier, the device is
not dismounted until mount verification completes.
Note, that this qualifier cannot be used in
conjunction with the /POLICY=MINICOPY (=OPTIONAL)
qualifier.
The device specified must be a member of a shadow set
that is mounted on this node.
o SET DEVICE
1. /FORCE_REMOVAL ddcu:
If connectivity to a device has been lost and the
virtual unit is in mount verification, this qualifier
causes the member to be expelled from the shadow set
immediately.
If the shadow set is not currently in mount
verification, no immediate action is taken.
The device specified must be a member of a shadow set
that is mounted on this node.
2. /MEMBER_TIMEOUT=xxxxxx ddcu:
Specifies the time out value to be used for a member
of a shadow set.
The value supplied by this qualifier overrides the
SYSGEN parameter SHADOW_MBR_TMO for this specific
device. The value can be different for different
members of the same shadow set.
The valid range for xxxxxx is 1 to 16,777,215 seconds.
The device specified must be a member of a shadow set
that is mounted on this node.
3. /MVTIMEOUT=yyyyyy DSAnnnn:
Specifies the mount verification time out value to be
used for this virtual unit.
The value supplied by this qualifier overrides the
SYSGEN parameter MVTIMEOUT for this specific virtual
unit.
The valid range for yyyyyy is 1 to 16,777,215 seconds.
The device specified must be a virtual unit that is
mounted on this node.
4. /READ_COST=zzz ddcu:
The valid range for zzz is 1 to 4,294,967,295 units.
The device specified must be a member of a virtual
unit that is mounted on this node.
This qualifier allows you to modify the default 'cost'
assigned to each member of a virtual unit, so that
reads are biased or prioritized toward one member
versus another.
The shadowing driver assigns default READ_COST values
to shadow set members when each member is initially
mounted. The default value depends on the device
type, and its configuration relative to the system
mounting it. There are default values for a DECRAM
device; a directly connected device in same physical
location; a directly connected device in a remote
location; a DECram served device; and for other served
devices.
The value supplied by this qualifier overrides the
default assignment. The shadowing driver adds the
current queue depth of the shadow set member to the
READ_COST value and does the read operation to the
member with the lowest value.
Different systems in the cluster can assign different
costs to each shadow set member. Reads requested for
virtual units from a system at Site 1 are performed
from a member that is also at Site 1. Reads requested
for the same virtual unit from Site 2 can read from
the member located at Site 2.
If the /SITE command qualifier has been specified, the
shadowing driver will take site values into account,
when it assigns default READ_COST values. Note, that
to have the shadowing driver determine if a device is
in the category of "directly connected device in a
remote location," the /SITE command qualifier must be
have been applied to both the virtual unit and that
device.
5. /SITE={nnn, logical_name} {ddcu: DSAnnnn:}
Qualifier indicates to the shadowing driver the site
location of the shadow set member or of the virtual
unit. Prior to using this qualifier, you may define
the site location in the SYLOGICALS.COM command
procedure, to simplify its use.
The valid range for nnn is 1 through 255.
The following example shows the site locations
defined, followed by the use of the /SITE qualifier:
$ DEFINE/SYSTEM/EXEC ZKO 1
$ DEFINE/SYSTEM/EXEC LKG 2
$!
$! At the ZKO site ...
$ MOUNT/SYSTEM DSA0/SHAD=($1$DGA0:,$1$DGA1:) TEST
$ SET DEVICE/SITE=ZKO DSA0:
$!
$! At the LKG site ...
$ MOUNT/SYSTEM DSA0/SHAD=($1$DGA0,$1$DGA1) TEST
$ SET DEVICE/SITE=LKG DSA0:
$!
$! At both sites, the following would be used:
$ SET DEVICE/SITE=ZKO $1$DGA0:
$ SET DEVICE/SITE=LKG $1$DGA1:
6. /COPY_SOURCE {ddcu:,DSAnnnn:}
Controls whether one or both source members of a
virtual unit are used as the source for read data
during full copy operations, when a third member is
added to the virtual unit. This only affects copy
operations that do not use DCD operations.
HSG80 controllers have a read-ahead cache, which
significantly improves single disk read performance.
Copy operations, normally alternate reads from the two
source members, which effectively nullifies the
benefits of the read-ahead cache.
This qualifier lets you force all reads from a single
source member for the copy operation.
If the virtual unit is specified, then all reads for
full copy operations will be performed from whichever
disk is the current 'master' member, regardless of
physical location of the disk.
If a member of the virtual unit is specified, then
that member will be used as the source of all copy
operations. This allows you to choose a local source
member, rather than a remote master member.
7. /ABORT_VIRTUAL_UNIT DSAnnnn:
To use this qualifier, the virtual unit must be in
mount verification. When you specify this qualifier,
the virtual unit aborts mount verification immediately
on the node from which the qualifier is issued. This
qualifier is intended to abort mount verification when
it is known that the unit cannot be recovered. Note,
that after this command completes, that the virtual
unit must still be dismounted using the following
command:
DISMOUNT/ABORT DSAnnnn
Use of the new DISMOUNT and SET command qualifiers.
The following diagram depicts a typical multi-site cluster
using fibre channel, and it will be used to illustrate what
steps need to be taken to manually recover one site, when the
site-to-site storage interconnect fails. Note that with
current Fibre Channel support, neither site can use the MSCP
server to regain a path to the DGA devices.
To prevent the shadowing driver from automatically recovering
virtual units, from connection related failures, two steps
must be taken prior to any failure occurring:
1. Every device that is a member of a multi-site shadow set
must have its time out set to a very high value, using the
$ SET DEVICE /MEMBER_TIMEOUT= x ddcu: command. Use of
this command will override the SHADOW_MBR_TMO value, which
would normally be used for a shadow set member. A value
for x of 259200 would be a seventy-two hour wait time.
2. Every multi site virtual unit will also have to have its
mount verification time out raised to a high value, using
the $ SET DEVICE /MVTIMEOUT = y command. The "y" value
should always be greater than the "x" value. Use of this
command will override the MVTIMEOUT value, which would
normally be used for the virtual unit. A value for y of
259260 would be a seventy-three hour wait time.
3. Every virtual unit and every shadow set member must have a
site qualifier. As was noted above, doing so will ensure
that the read cost is correctly set. The other critical
factor in doing this involves three member shadow sets.
When they are being used, the site qualifier will ensure
that the master member of the shadow set will be properly
maintained.
In this example, DSA42 is made up of $1$DGA1000 and $1$DGA2000
<><><><><><><><><><><> LAN <><><><><><><><><><><>
Site A Site B
| |
F.C. SWITCH <><><><> XYZZY <><><><> F.C. SWITCH
| |
HSG80 <><> HSG80 HSG80 <><> HSG80
| |
$1$DGA1000 --------- DSA42 --------- $1$DGA2000
On Site A:
$ DISMOUNT /FORCE_REMOVAL= $1$DGA2000:
Once that command has completed, the virtual unit will be
available for use at site A, only.
On Site B:
$ SET DEVICE /ABORT_VIRTUAL_UNIT DSA42:
Once the command completes, the virtual unit status will be
MntVerifyTimeout. Next, issue the following command to free
up the virtual unit:
$ DISMOUNT/ABORT DSA42:
These steps need to be taken for all affected multi-site
virtual units.
PROBLEMS ADDRESSED IN VMS721H1_SHADOWING-V0100 KIT:
o Executing SET DEVICE commands can result in a crashed
system at NSA$SIZE_NSAB due to invalid argument counts in
the $CMKRNL calls.
Images Affected: [SYS$LDR]SET.EXE
o SHOW LOGICAL returns a %LIB-F-INSVIRMEM error.
Images Affected: [SYS$LDR]SHOW.EXE
PROBLEMS ADDRESSED IN VMS721H1_SYS-V0200 KIT:
o A problem in a $TRNLNM code path for INTERLOCKED
translations can result in the service exiting without
releasing the logical name mutex (mutual exclusion
semaphore). If that $TRNLNM request or any subsequent
kernel mode system service request made by that process
exits with an error status, the system will crash with a
MTXCNTNZ bugcheck.
If no kernel mode system service request made by that
process exits with an error status, the system will
eventually hang, with some processes in MUTEX wait trying
to acquire the logical name mutex. If some of those
processes have already acquired other mutexes, such as
the I/O data base mutex and GSD mutex, there may be other
processes in mutex wait trying to acquire those mutexes.
The $TRNLNM bug is exercised by a fairly unusual
combination of circumstances and is more likely to be
seen on an SMP system.
Crashdump Summary Information:
------------------------------
Bugcheck Type: MTXCNTNONZ, Mutex count nonzero at system
service exit
Current Process: ORA_PRODC0661
Current Image: $1$DGA21:[ORACLE8.RDBMS]ORACLE.EXE
Failing PC: FFFFFFFF.8008EFF4
__RELEASE_SERVICE_ERROR_EXCEPT+00094
Failing PS: 38000000.00000200
Module: EXCEPTION (Link Date/Time:
28-MAY-1999 23:22:24.23)
Offset: 00018FF4
Images Affected: [SYS$LDR]LOGICAL_NAMES.EXE
o Memory reservations can consist of multiple parts. The
code to clean up the resident sections of memory
contained several paths where the MMG spinlock was not
released. This could cause the system to crash with
CPUSPINWAIT or SPLACKERR bugchecks.
This fix is required for all customers who are using the
reserved memory feature in SYSMAN.
Crashdump Summary Information:
------------------------------
Bugcheck Type: CPUSPINWAIT, CPU spinwait timer expired
Current Process: VERIFICATION
Current Image: DSA7777:[SECTIO]SECTIO.EXE;4
Failing PC: FFFFFFFF.8007C384
SYSTEM_SYNCHRONIZATION_MIN+00384
Failing PS: 18000000.00000803
Module: SYSTEM_SYNCHRONIZATION_MIN (Link Date/Time:
2-JUN-2000 11:12:08.33)
ffset: 00000384
Images Affected: [SYS$LDR]SYSTEM_PRIMITIVES.EXE
[SYS$LDR]SYSTEM_PRIMITIVES_MIN.EXE
[SYS$LDR]SYSTEM_PRIMITIVES.STB
[SYS$LDR]SYSTEM_PRIMITIVES_MIN.STB
o The GETDVI (GET DEVICE/VOLUME INFORMATION system service)
does not take into account that a device might have
multiple UCBs (unit control blocks) associated with it.
Images Affected: [SYS$LDR]IO_ROUTINES.EXE
o Applications using the kernel threads features can get
stuck in a loop.
Images Affected: [SYS$LDR]PROCESS_MANAGEMENT.EXE
[SYS$LDR]PROCESS_MANAGEMENT_MON.EXE
o a $SHOW SYSTEM/FULL" command displays the UIC of all
processes (except interactive process) as [0,0].
$GETJPI shows the correct UIC's except swapper.
o A system can crash with a SSRVEXCEPT bugcheck in
module PROCESS_MANAGEMENT_MON (SYSTEM_CHECK=1) at
offset 0000F4B4.
Images Affected: [SYS$LDR]PROCESS_MANAGEMENT.EXE
[SYS$LDR]PROCESS_MANAGEMENT_MON.EXE
o A system can crash with a REFCNTNEG bugcheck in
MMG$FREWSLX_64_C or with a FREEPAGREF bugcheck in
MMG_STD$ALLOC_ZERO_PFN_64_C. In both crashes, one of the
process's level 2 page table PFNs (page frame numbers) is
incorrectly set to 1100.
This crash can only occur with a 64-bit program running
on an AlphaServer GS-Series 80/160/620. The program must
page out a L2 page table page, and only occurs when the
process's working set is set low. Allocating a L2 page
off RAD only occurs when the process's home RAD is
changed or under low memory conditions.
The following SDA command on a crash dump will quickly
show if you have had this problem:
SDA> show process/page/l1/p2
Images Affected: [SYS$LDR]SYS$VM.EXE
o A system can crash with a SPINWAIT bugcheck with
fragmented non-paged pool. This crash will occur only if
either of the SYSGEN parameter NPAG_GENTLE or
NPAG_AGGRESSIVE are set to a number smaller than 100 (the
default case).
Images Affected: [SYS$LDR]SYSTEM_PRIMITIVES.EXE
[SYS$LDR]SYSTEM_PRIMITIVES_MON.EXE
o Protect against corrupt PTEs that do not have the KRE
(known file entry) access bit set or have invalid PFNs
(page frame numbers) on AlphaServer GS-Series 80/160/320
systems. Concurrently, improve system diagnostics.
Images Affected: [SYS$LDR]EXCEPTION.STB
[SYS$LDR]EXCEPTION_MON.STB
[SYSLIB]SDA$SHARE.EXE
[SYS$LDR]SYS$BASE_IMAGE.EXE
o Prevent LNMB (logical name block) headers from being
partially written as a result of corrupt PTEs that do not
have the KRE (known file entry) access bit set or PTEs
that have invalid PFNs (page frame numbers).
Images Affected: [SYS$LDR]EXCEPTION.STB
[SYS$LDR]EXCEPTION_MON.STB
[SYSLIB]SDA$SHARE.EXE
[SYS$LDR]SYS$BASE_IMAGE.EXE
o The spinlock contention on MMG (Memory Management
spinlock) can become excessive when running Oracle 7 with
multiple clients connected simultaneously.
Images Affected: [SYS$LDR]SYS$VM.EXE
o The effect of a SET SECURITY/OBJECT=DEVICE command can be
propagated to the wrong device(s) in a cluster for MK,
DK, and DG devices.
Images Affected: [SYSLIB]IOGEN$SHARE.EXE
[SYS$LDR]SYS$BASE_IMAGE.EXE
[SYS$LDR]IO_ROUTINES.EXE
[SYS$LDR]IO_ROUTINES_MON.EXE
o The swapper code that controls shared page tables may not
function properly, if the same level 2 page table page
maps both a memory resident page table of PFN-mapped
pages and shared page tables. A DELCONPFN crash could
occur when the level 2 page tables are deleted.
Images Affected: [SYS$LDR]SYS$VM.EXE
o A system can crash with a KRNLSTACKNV bugcheck during
heavy I/O activity, such as BACKUP. Forcing the stack
out shows interaction between SYS$DKDRIVER and
IO_ROUTINES filling up the KPB stack, usually with some
interrupt topping off the stack.
Crashdump Summary Information:
------------------------------
Bugcheck Type: KRNLSTAKNV, Kernel stack not valid
Current Process: NULL
Current Image:
Failing PC: FFFFFFFF.825D7310
Failing PS: 00000000.00001504
Module:
Offset: 00000000
Images Affected: [SYS$LDR]IO_ROUTINES.EXE
[SYS$LDR]IO_ROUTINES_MON.EXE
o A CPU can run two processes in the current state, which
results in an INCON_SCHED, 'Inconsistent scheduling'
bugcheck. this problem occurs on a single CPU system.
Crashdump Summary Information:
------------------------------
Bugcheck Type: INVEXCEPTN, Exception while above ASTDEL
Current Process: BROKER
Current Image: DSA360:[BROKER_U.AXP.][P]BROKER_EDITOR_U.EXE
Failing PC: FFFFFFFF.800C3B98 SCH$QEND_C+00038
Failing PS: 10000000.00000704
Module: PROCESS_MANAGEMENT (Link Date/Time:
29-DEC-1999 04:09:20.9
Offset: 00007B98
Images Affected: [SYS$LDR]SYS$VM.EXE
o A system can crash with an INVEXCEPTN bugcheck at
SCH$QEND_C+38 while getting the address of the process
alignment fault reporting information,
CTL$GL_REPORT_USER_FAULTS.
Crashdump Summary Information:
-------------------------------
Bugcheck Type: INVEXCEPTN, Exception while above ASTDEL
Current Process: BROKER
Current Image: DSA360:[BROKER_U.AXP.][P]BROKER_EDITOR_U.EXE;85
Failing PC: FFFFFFFF.800C3B98 SCH$QEND_C+00038
Failing PS: 10000000.00000704
Module: PROCESS_MANAGEMENT (Link Date/Time:
29-DEC-1999 04:09:20.9
Offset: 00007B98
Images Affected: [SYS$LDR]SYS$VM.EXE
o A multi-threaded process can hang with one thread
spinning in a loop using CPU time.
Images Affected: [SYSLIB]SYS$SSISHR.EXE
o A large crash dump may fail to include the GCT (Greenwich
Civil Time).
Images Affected: [SYS$LDR]EXCEPTION.EXE
[SYS$LDR]EXCEPTION_MON.EXE
[SYS$LDR]EXCEPTION.STB
[SYS$LDR]EXCEPTION_MON.STB
o The console callback FIND_NODE was so slow it caused the
console to lose clock ticks. An earlier fix required the
system manager manually to set the SYSGEN parameter VMS8
to %x1029000, which would force the console to use the
VMS version of FIND_NODE rather than the callback
version. That effectively solved the problem.
This enhancement sets VMS8 automatically to %x1029000.
Images Affected: [SYS$LDR]SYSTEM_PRIMITIVES.EXE
[SYS$LDR]SYSTEM_PRIMITIVES_MIN.EXE
PROBLEMS ADDRESSED IN VMS721H1_SYS-V0100 KIT:
o Unnecessary and unwanted path switches can occur on
multipath devices. Under certain circumstances, if a
user executes a manual path switch of one member of a
shadow set, the requested path switch takes place.
However, the other member(s) of the shadow set switch
paths as well. Further, if the user attempts to switch
the other member(s) back, the other members will switch,
but the originally switched member will then switch back
to the unwanted path!
Another symptom of this problem is that a transient error
condition on a multipath device can cause a path switch,
even though the current path is still valid.
This problem can occur if a multipath disk device is
simultaneously online, i.e., connected, on more than one
path. This configuration is created:
o If two FibreChannel cables are attached to the two
host FibreChannel ports on an HSG80 controller.
o If two or more FibreChannel host bus adapters on the
same OpenVMS host system connect to the same fabric,
i.e. the same FibreChannel switch or into a set of
cascaded switches.
o If two parallel SCSI buses are connected to the two
host ports on an HSZ80 controller.
Images Affected: [SYS$LDR]MULTIPATH_MON.EXE
o A cross process $GETJPI request for security profile
(persona) information, which includes network privileges
and rights, can lead to a SSRVEXCPT system crash. See
crash dump summary below:
Bugcheck Type: SSRVEXCPT, Unexpected system service exception
CPU Type: COMPAQ AlphaServer DS20E 500 MHz
Current Process: NETACP
Current Image: $2$DKA100:[SYS0.SYSCOMMON.][SYSEXE]NETACP.EXE;1
Failing PC: FFFFFFFF.8011657C EXE_STD$CHECK_IMAGE_NAME_C+0033C
Failing PS: 00000000.00000000
Module: PROCESS_MANAGEMENT_MON (Link Date/Time: 13-MAR-2000
13:54:01.61)
Offset: 0001057C
Images Affected: [SYS$LDR]PROCESS_MANAGEMENT.EXE
[SYS$LDR]PROCESS_MANAGEMENT_MON.EXE
o A non-privileged user can access jobs in a batch queue,
regardless of the queue protections. See the comparative
examples below:
$ show queue/full/all unhf_sys$batch ! from privileged account
Batch queue UNHF_SYS$BATCH, idle, on UNHF::
/BASE_PRIORITY=3 /CPUMAXIMUM=00:30:00 /JOB_LIMIT=3 /OWNER=[SYSTEM]
/PROTECTION=(S:M,O:D,G,W:RS) /WSEXTENT=32768 /WSQUOTA=16384
(IDENTIFIER=[SIS_DEVEL,BANNER_SCT],ACCESS=READ+SUBMIT+MANAGE)
Entry Jobname Username Status
----- ------- -------- ------
2719 DUMMY B_JOHNSTONE Holding
Submitted 1-APR-2000 09:16:27.00 /KEEP
/LOG=$1$DUA233:[B_JOHNSTONE].LOG; /NOPRINT /PRIORITY=100
/RESTART=UNHF_SYS$BATCH
File: _$1$DUA321:[B_JOHNSTONE.COM]DUMMY.COM;10
3182 DUMMY B_JOHNSTONE Holding
Submitted 1-APR-2000 14:15:28.41 /KEEP
/LOG=$1$DUA233:[B_JOHNSTONE].LOG; /NOPRINT /PRIORITY=100
/RESTART=UNHF_SYS$BATCH
File: _$1$DUA321:[B_JOHNSTONE.COM]DUMMY.COM;10
$show queue/full/all unhf_sys$batch !from non-privileged account
Batch queue UNHF_SYS$BATCH, idle, on UNHF::
/BASE_PRIORITY=3 /CPUMAXIMUM=00:30:00 /JOB_LIMIT=3 /OWNER=[SYSTEM]
/PROTECTION=(S:M,O:D,G,W:RS) /WSEXTENT=32768 /WSQUOTA=16384
(IDENTIFIER=[SIS_DEVEL,BANNER_SCT],ACCESS=READ+SUBMIT+MANAGE)
Entry Jobname Username Status
----- ------- -------- ------
2719 no privilege Holding
3182 DUMMY B_JOHNSTONE Holding
Submitted 1-APR-2000 14:15:28.41 /KEEP
Submitted 1-APR-2000 14:15:28.41 /KEEP
/LOG=$1$DUA233:[B_JOHNSTONE].LOG; /NOPRINT /PRIORITY=100
/RESTART=UNHF_SYS$BATCH
File: _$1$DUA321:[B_JOHNSTONE.COM]DUMMY.COM;10
$
In this example, the user can see entry 3182, as well as security
information, but cannot see entry 2719.
This also generates the following security alarm:
%%%%%%%%%%% OPCOM 1-APR-2000 15:38:36.66 %%%%%%%%%%% (from node
UNHA at 1-APR-2000 15:38:36.67)
Message from user AUDIT$SERVER on UNHA
Security alarm (SECURITY) on UNHA, system id: 1028
Auditable event: Object access
Event time: 1-APR-2000 15:38:36.65
PID: 2040C464
Source PID: 21012416
Username: R_KENNEY$
Process owner: [R_KENNEY$]
Object class name: QUEUE
Object name: UNHF_SYS$BATCH
Object owner: [0,0]
Object protection: SYSTEM:M, OWNER:D, GROUP:, WORLD:RS
Access requested: READ
Status: %SYSTEM-F-NOPRIV, insufficient privilege or
object protection violation
Images Affected: [SYS$LDR]SECURITY.EXE
[SYS$LDR]SECURITY_MON.EXE
o An AST (asynchronous system trap) could cause process
header expansion while in CHECK_WINDOW_64. The routine
runs at IPL0 and could allow an AST while trying to
locate a process section table entry.
Images Affected: [SYS$LDR]SYS$VM.EXE
o A problem with the $TRNLNM code path for INTERLOCKED
translations can cause the service to exit without
releasing the logical name mutex. If the $TRNLNM request
or any subsequent kernel mode system service request made
by that process exits with an error status, the system
will crash with a MTXCNTNZ bugcheck.
If no kernel mode system service request made by that
process exits with an error status, the system will
eventually hang, with some processes in MUTEX wait trying
to acquire the logical name mutex. If some of those
processes have already acquired other mutexes, such as
the I/O data base mutex and GSD mutex, there may be other
processes in MUTEX wait trying to acquire those mutexes.
The $TRNLNM bug is exercised by a fairly unusual
combination of circumstances and is more likely to be
seen on an SMP system.
Images Affected: [SYS$LDR]LOGICAL_NAMES.EXE
o CPUSPINWAIT or SPLACKERR crashes could occur when the MMG
spinlock is not released. This fix is required for all
customers who are using the reserved memory feature in
SYSMAN.
Images Affected: [SYS$LDR]SYSTEM_PRIMITIVES.EXE
[SYS$LDR]SYSTEM_PRIMITIVES_MIN.EXE
o The system can crash with an INCONSTATE bugcheck in
CACHE$MOUNT. This occurs when a process, usually
RAID$SERVER, is attempting to mount a disk, usually a
member of a Raid set. It appears as if the volume is
being mounted twice and the INCONSTATE bugcheck occurs.
Images Affected: [SYS$LDR]SYS$VCC.EXE
o The incorrect S-float value was calculated in IEEE
multiply and divide routines.
Images Affected: [SYS$LDR]EXCEPTION.EXE
[SYS$LDR]EXCEPTION_MON.EXE
o An INCONSTATE bugcheck can occur during a RAID unbind
operation.
Images Affected: [SYS$LDR]SYS$VCC.EXE
[SYS$LDR]SYS$VC_MON.EXE
o A BLKASTCNT crash occurred with Pathworks enqueuing many
locks, all with blocking ASTs (asynchronous system
traps). The crash occurred when one of the locks was
dequeued.
Images Affected: [SYS$LDR]LOCKING.EXE
INSTALLATION NOTES:
The images in this kit will not take effect until the system is
rebooted. If there are other nodes in the VMScluster, they must
also be rebooted in order to make use of the new image(s). Compaq
strongly recommends that a reboot is performed immediately after
kit installation to avoid system instability.
If it is not possible or convenient to reboot the entire cluster at
this time, a rolling re-boot may be performed.
INSTALLATION INSTRUCTIONS:
Install this kit with the POLYCENTER Software installation
utility by logging into the SYSTEM account, and typing the
following at the DCL prompt:
PRODUCT INSTALL VMS721H1_UPDATE/SOURCE=[location of Kit]
The kit location may be a tape drive, CD, or a disk directory that
contains the kit.
Additional help on installing PCSI kits can be found by typing
HELP PRODUCT INSTALL at the system prompt
Note: This kit will make functional changes to your system.
Before installing this kit you should make a backup
copy of your system disk. If you do not make a copy
of your system disk you will not be able to restore
your system to a pre-kit installation state.
All trademarks are the property of their respective owners.
This patch can be found at any of these sites:
Colorado Site
Georgia Site
Files on this server are as follows:
dec-axpvms-vms721h1_update-v0400--4.README
dec-axpvms-vms721h1_update-v0400--4.CHKSUM
dec-axpvms-vms721h1_update-v0400--4.pcsi-dcx_axpexe
vms721h1_update-v0400.CVRLET_TXT
|
