|
Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com |
HP-UX series 800 11.X patch digest
From: IT Resource Center (support_feedback
us-support2-mail.external.hp.com)
Date: Sun May 11 2003 - 14:10:40 CDT
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
HP Support Information Digests
===============================================================================
o IT Resource Center World Wide Web Service
---------------------------------------------------
If you subscribed through the IT Resource Center and would
like to be REMOVED from this mailing list, access the
IT Resource Center on the World Wide Web at:
http://support.itrc.hp.com/
Login using your IT Resource Center User ID and Password.
Then select Support Information Digests (located under
Maintenance and Support). You may then unsubscribe from the
appropriate digest.
To download a patch referenced below, access the
IT Resource Center on the World Wide Web at:
http://support.itrc.hp.com/
Login using your IT Resource Center User ID and Password.
Then select Individual Patches (under Maintenance and Support)
to access the patch. You may also download a patch via anonymous
ftp(1) from ftp.itrc.hp.com.
===============================================================================
Digest Name: weekly HP-UX series 800 11.X patch digest
Created: Sun May 11 7:05:06 EDT 2003
Table of Contents:
Document ID Title
--------------- -----------
PHSS_28943 s700_800 11.X OV ITO7.1X Windows Agent A.07.20
PHSS_28820 s700_800 11.X OV ITO7.1X HP 10.20 Embedded Perf Agt A.07.15
PHCO_29052 s700_800 11.04 (VVOS) vvosmta(1M) enhancement
PHSS_28958 s700_800 11.X OV ITO7.1X HP-UX 11.00/11.11 Agent A.07.20
PHSS_28922 s700_800 11.X OV ECS3.20 Intermediate Patch April 2003
PHNE_28183 s700_800 11.00 R6.11.00 SNAplus2 Link/Common, TN3270 patch
PHKL_28665 s700_800 11.11 PCI-X enablement for OLA/R subsystem
PHKL_28569 s700_800 11.11 INTRMIGR enabler, WSIO support patch
PHKL_28464 s700_800 11.11 IPMI driver port, LUN change
PHCO_28769 s700_800 11.11 nPar cmds cumulative patch,IOX,dual cpu
PHCO_28666 s700_800 11.11 PCI-X enablement for rad(1M) command
PHCO_27847 s700_800 11.00 cumulative patch for syslogd(1M)
PHSS_28968 s700_800 11.X OV ITO7.1X JavaGUI client A.07.14
PHSS_28962 s700_800 11.X OV ITO7.1X server patch A.07.15
PHSS_28825 s700_800 11.X OV ITO7.1X Solaris Embedded Perf Agt A.07.15
PHSS_28821 s700_800 11.X OV ITO7.1X 11.X Embedded Perf Agent A.07.15
PHNE_28588 s700_800 11.00 EISA FDDI B.11.00.0[0-7] cumulative patch
PHNE_28143 s700_800 11.00 LAN product cumulative patch
PHCO_28998 s700_800 11.11 fbackup(1M),frecover(1M) cumulative patch
PHCO_28719 s700_800 11.11 wall(1M) patch
PHCO_28718 s700_800 11.00 wall(1M) patch
PHCO_28247 s700_800 11.11 script(1) cumulative patch
The documents are listed below.
-------------------------------------------------------------------------------
Document ID: PHSS_28943
Date Loaded: 20030508
Title: s700_800 11.X OV ITO7.1X Windows Agent A.07.20
Patch Name: PHSS_28943
Patch Description: s700_800 11.X OV ITO7.1X Windows Agent A.07.20
Creation Date: 03/04/30
Post Date: 03/05/08
Hardware Platforms - OS Releases:
s700: 11.00 11.11
s800: 11.00 11.11
Products:
OpenView Operations 7.1
Filesets:
OVOPC-CLT.OVOPC-NT-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,v=HP
OVOPC-CLT.OVOPC-NT-CLT,fr=A.07.10,fa=HP-UX_B.11.11_32/64,v=HP
Automatic Reboot?: No
Status: General Release
Critical: No
Category Tags:
defect_repair general_release
Path Name: /hp-ux_patches/s700_800/11.X/PHSS_28943
Symptoms:
PHSS_28943:
- SR: B555015785
opcecaas may report that it runs out of free process
slots and therefore can't start anymore process.
- SR: 8606294668
The trap interceptor does not use new/changed
templates/policies when doing SNMP trap interception on a
Windows system that has NNM installed.
- SR: B555015674
The Windows agent installation may abort with the error
that the node is already managed by a different
management server.
The inst.log and the command line output on the node you
are installing to may contain:
E-> NOTE: VPO software, version A.0X.X installed on
system <node name>, belongs to Management Server
<management server name>.
- SR: 8606282084
The change between multiple managements doesn't work,
because the configuration file primmgr - that has higher
priority for agent, is not updated by the
'SwitchMgmtServer.cmd' tool that uses 'mgmt_sv.vbs'
script.
- SR: B555016133
opcwbemi.exe may abort while stopping the agent on
Microsoft Windows XP.
- SR: B555015741
During the shutdown of a system, the agent processes
still try to start new processes, causing pop up windows.
- SR: B555015449
Monitor agent may abort while receiving
templates/policies.
- SR: 8606290346
After installing a new version the registry key
HKLM\SOFTWARE\Hewlett-Packard\OpenView\ITO\Agent Version
remains at A.07.10.
- SR: B555015199
If OPC_IP_ADDRESS (nodeinfo) and OPC_NAMESRV_LOCAL_NAME
(opcinfo) were used in combination the entry in
OPC_IP_ADDRESS was not used in the right way.
- SR: B555015557
The current implementation of the control agent does not
allow to restart aborted sub agent processes
automatically.
- SR: 8606303962
After time change (beginning of daylight saving time),
all logfiles are read from the beginning.
- SR: B555015700
Agent often fails to shut down from the services control
panel in time. It also fails to stop in time during the
uninstall/redeployment of the agent software.
- SR: 8606299704
You may get the error
"Cannot open file <$...>. System Error Number: ... -
(OPC20-63)"
when using environment variables in the "File to be Read"
field of a logfile policy/template.
- SR: B555015170
During distribution of templates, the agent might report
that 'The codeset "-1" is incompatible to the codeset
"acp1252"'.
- SR: B555015758
opcmsgi aborts if one of the set attributes has an
unmatched '<'.
- SR: 8606290908
The OVO Windows agent may loos it's AgentID when
doing a "Redeploy All" or installing an agent package
with a higher version. In case this happens, Service
Discovery may stop working correctly.
- SR: B555015770
Agent should report the current internal status of
policies. The Flag OPC_TEMPL_STATUS TRUE can be set now,
reporting the status for monitors and logfile policies in
the two files 'monitor.status' and 'le.status' in
'%OvAgentDir%\tmp\OpC'.
- SR: B555015771
opcmona.exe may sometimes fail to start and show up a
message box that a certain ordinal can not be found.
- SR: B555015325
opcmsga wrongly generates a message OpC30-3002 'opcctla
not running on node localhost.' when receiving a
reconfigure signal, or at exit time.
- SR: H555008529
If a process dies immediately after being started by the
Control Agent, it is possible that OpC30-1094 messages
start appearing in the error logfile.
- SR: 8606297998
The ECS engine (opcecm & opceca) might have problems
when a circuit is using global dictionaries.
(This is the case for the MessageStorm detection ECS
circuit).
- SR: B555008674
The opcagt and the opcragt commands have a new option
'-version'. In both man pages, this was not documented.
The new option was not part of the usage strings of
opcagt and opcragt.
- SR: B555015378
The agent might not be able to monitor instances of the
LogicalDisk object and returns no data for it at all.
- SR: B555014942
The opcle process loops if a logfile is removed while
it is read.
- SR: B555014851
opcmsga sends the same message operation (e.g. an
acknowledge request created by opcmack(1) )
again and again if the related message is not in the
cache and one of the target managers can not be reached.
- SR: B555016032
The process opcapm.exe may show a memory leak during
runtime.
- SR: B555015912
Agent installation fails when ntmsgm.dll is in use by
other process
- SR: B555015484
You might not be able to install the agent either through
server based deployment or manual agent installation. In
case Dr. Watson is configured you will see there, that
the opcsetup.exe aborted.
The problem can also show up during agent runtime. In
this case you may not be able to execute any commands
through the agent.
- SR: 8606293840
Monitor agent and Event interceptor grow in memory.
- SR: B555015496
Queue file handling was inefficient for larger queue
files.
- SR: B555015485
In this case the cause was a mapped network drive listed
in the system environment variable PATH.
Every time the agent tries to execute an external
command, the PATH might be scanned by the used Windows
APIs.
This is causing a login error on the PC providing the
network share and in case this PC also has the "HP ITO
account" it will lock this account.
- SR: B555015441
Action agent aborts when process table is full. This
happens when more than 52 tools/actions are called from
the console in parallel.
- SR: 8606290675
Agent should report 'agent start' into the opcerror file.
- SR: R555018185
If the agent is started in Simplified Chinese codeset,
monitor agent, logfile encapsulator and message
interceptor processes will fail.
- SR: B555015155
Errors in the opcerror file are not written to the trace
file. This makes it difficult to find the place in the
trace file where the error occured.
- SR: 8606280846
Monitor Agent often reports 'Evaluation of policy
stopped'.
- SR: B555015277
The NT monitor agent aborts in process-monitoring policy
when the 'ID Process' performance counter exists twice in
the performance registry.
- SR: B555015331
The monitor agent, opcmona, may report wrong results of
executed monitor scripts or programs when using many
'advanced monitors' such as OVPERF. In some of these
cases opcmona might even abort.
- SR: B555015712
The trap/event interceptor forwards traps with the source
address 127.0.0.1 unchanged, therefore they are discarded
by the management server due to the unknown source
address.
- SR: 8606232431
VPO tries to resolve node names that only contain blanks
because of typo in template definition or variable
assignment. This leads to a lot of unnecessary DNS
traffic.
- SR: 8606300119
The logfile encapsulator does not return the last line of
a logfile immediatelly, if there is no carriage return in
this line.
PHSS_28501:
- SR: 8606290723
You may get errors about commands or DLLs that can not be
found while executing a command or batch file using the
OVO Windows agent. This behavior has been seen with the
DBSPI very often.
- SR: B555015349
The monitor agent might abort while using
templates/policies
that have an external source specified.
- SR: B555015308
When configuring the agent with apminfo.xml
to be cluster aware, the agent may not install newly
deployed policies correctly.
PHSS_27388:
- SR: B555013936
agent installation may fail with a high number of the
following messages in the inst.log:
T-> Tracing information: Can't get status for service:
...
It can even happen that the install script reports
successfull but nothing gets installed.
This may happen when you have installed HP OpenView
Internet Services and not all MS Internet Information
Services.
- SR: B553003927
opcmsgi leaks memory when handling messages with custom
message attributes.
- SR: B555014999
The start of batch files (.bat|.cmd) with Tools/Automatic
or Operator initiated actions by using the OVO Agent on
Windows NT may fail with a an error saying, the command
is not found.
- SR: B555014328
opcevti.exe may abort when using <$x>.
- SR: B555015077
Installation server always receives version A.07.10
to be installed on the target node.
This causes problems in case the installation server just
knows A.07.12 but not A.07.10.
- SR: B555014961
The OVO agent should be able to execute tools and
scheduled commands for either the "HP ITO account" or the
SYSTEM account user. Or any other user currently running
the agent. In any case this should work without a switch
user.
- SR: B555014634
The OVO Windows agent may not pickup the SYSTEM event log
entries with ID 6008 or 6009 during the system is
rebooted.
Or in other cases the opcle may go to 100 % CPU time for
several minutes during startup.
- SR: B555013965
The pdh.dll delivered with the agent
has a memory leak as it is described
in Microsoft article: Q263221.
- SR: B555013978
In case the agent buffering and you execute the command
opcagt or "opcagt -status" you may get an output like:
"Message Agent opcmsga buffering for
testmgmtsrv.bbn.hp.com (2376) is running"
- SR: B555014558
itodiag.exe/OvNtDiag.exe may start looping/hanging
forever.
- SR: B555014093
opcmona may crash (UNIX) or doesn't process all SCHEDULE
templates (Windows) when using SCHEDULE templates.
- SR: B555014513
opcle.exe may generate a message storm with messages
saying that one of the Windows event logs is corrupt.
- SR: B555014714
You may receive a message storm with messages:
OPC INTERNAL ERROR, CANNOT GENERATE MESSAGE
1500 OF SET ZERO
Or the opcle may start taking a lot of CPU over a long
period of time.
- SR: B555014523
nt_inst.exe may abort with Dr. Watson while trying to
install the Windows agent to a node.
- SR: B555013510
Autoactions are not executed on systems in a NAT
environment.
- SR: R555017956
The monitor agent is terminated if you use a session
variable within the message text of a template followed
by any other variable (for example $INSTANCE).
- SR: R555018043
Japanese characters in Perl scripts within templates
don't get converted correctly into the server code set.
If the Japanese characters are used as message text for
example these characters are corrupted.
- SR: B555014715
The Control Agent slowly grows in memory usage.
- SR: B555014212
opcle may report:
ntle.c:3670: OpC internal error: Cannot generate message
1500 of set 0
System Error Number: 1966234 (1e009a) - (OpC0-1500)
- SR: B555012869
Instead of using the trap's UUID, the trap interceptor
created a new message id for all but the first trap
template. The original message id was not set correctly
in these cases.
- SR: B555014728
Currently OvNtKill.exe / itokill.exe requires the
libopc.dll as well as some other DLLs required by
libopc.dll.
- SR: 8606282247
Logfile Encapsulator does not perform variable
replacement
for all Message Defaults fields.
- SR: 8606267798
In SNMP trap policies when using the <$r> or <$R>
variables, on Windows systems the IP address is used in
reverse order and therefore there are incorrect results.
- SR: B555014667
The first lines of a logfile are not forwarded to the
message browser when using a command to discover
logfiles and the logfile was created after the first
polling interval.
- SR: B555014708
Memory leaks in agent process. Especially monitor and
action agent leak when executing processes.
- SR: B555014378
opcctla may crash when doing a distribution on a Windows
XP node in a NAT environment.
- SR: B555014799
NT agent reinstall might cause cmd.exe startup issues.
This can happen when scheduled actions or tools have been
executed as user opc_op without specifying a password.
- SR: B555013796
When using itomserv.exe/opcntmserv.exe to list services
or devices where the Service Name or Description contains
Japanese Multibyte characters it may happen
that the text is truncated wrongly. The leads to a wrong
last character and maybe even to problem when displaying
this text in the Java or Motif UI.
- SR: B555009284
The authorization verification for remote start and stop
requests of the agent was sometimes unreliable. Possibly
this allowed more OVO servers to start or stop the agent,
than specified in the MoM configuration.
- SR: 8606242614
Messages are incorrectly suppressed by the logfile
encapsulator if "suppress identical output messages" is
specified and the messages differ only in the values of
<$LOGFILE> and/or <$LOGPATH>.
- SR: B555014882
When monitoring Windows performance values by using the
NTPerfMon (OVO UX) / Real Time Performance (OVO Windows)
sources, there might be an error that not all instances
can be found.
- SR: B555014759
When enabling/disabling policies you might discover a
memory leak in the agent processes.
- SR: B555014319
When reading Windows Event logs the opcle.exe and
services.exe may start taking a lot of CPU time for a
longer period of time.
- SR: B555013872
Agent installation sets full access rights for "HP ITO
account" on files in %SYSTEMROOT%\system32
(opcauth.dll and psapi.dll).
- SR: 8606286316
When retrieving a list of logfiles by using <`...`>
the logfile encapsulator may not retrieve any filename.
This has been seen when using perl scripts, that
internally call other system commands through the perl
function 'system' or 'open' or just use the backticks to
execute the command.
e.g.: $output = `more c:\loglist.txt`;
- SR: B555014591
When the OPC_INT_MSG_FLT is set to TRUE then the filtered
message is received corrupted on the server in
a Japanese environment.
- SR: B555015004
After uninstalling an OV Agent in Local Security Policy
there were still some SIDs left over.
- SR: R555019153
When deplying policies from a Windows management server
and using the "StoreCollection" method within the Perl
or VBScript, the monitor agent might show the Policy name
instead of the given metricname or may abort.
- SR: 8606285428
SNMP V2 Traps cannot be received via Windows Agent, when
NNM is not installed on the Node.
- SR: B555013891
In MoM environments, opcmsga does not return action
responses to SECONDARY managers, if their name is not
resolvable.
- SR: B555015072
'opcagt -id 12 -stop' may not work when starting as
'Administrator' as long as the OVO agent runs with
'SYSTEM' (LocalSystem) account.
- SR: H555008631
Customer receives a lot of OpC20-61 and OpC20-63 messages
in the error logfile when using NCS agents.
- SR: B555014442
opcle may report system error 32 and OPC20-63 while
processing a logfile on MS Windows systems and after this
stopping the processing of this logfile at all.
- SR: B555014570
The logfile encaspulator might start consuming CPU when
doing Windows Event Log monitoring.
- SR: B555012827
Support Windows XP with the VPO Windows agent.
Please see the "Defect Description" section below for
further details.
- SR: 8606244582
The distribution/deployment of action/command/monitor
binaries might fail due to some of them being used.
- SR: B555014101
Monitor agent may report error OPC30-783 and the
initialisation of the first template/policy using Windows
Performance data fails.
- SR: B555015047
opcecaas (Annotation Server) might log error
"Illegal NULL parameter (function ) (OpC10-11)"
with many similar lines.
- SR: B555014281
When using the Windows agent SDK (opcapi.h and
opcapi.lib) you may get linker unresolved symbol errors
when using one of the following API calls:
opcdata_type, opcdata_ldel, opcdata_ladd,
opcdata_lget_len, opcdata_lget_str, opcdata_lset_str,
opcdata_generate_id, opcdata_get_error_msg,
opcdata_lget_long, opcdata_lset_long, opccdata_clear,
opc_version
- SR: B555014700
The monitor agent and action agent on NT leak handles.
- SR: B555013540
opcsetup may report:
E-> Error from opc_ipinterf_getaddrs.
E-> Error from InitSetup.
E-> Setup program aborted.
- SR: B555015050
The environment PATH set by the agent during executing
external commands may contain several duplicate entries
and get very large therefore. This may cause problems
when an executed program tries to add it's own
directories as well.
- SR: B555013719
Message agent doesn't stop message buffering when the
management server is available again after a network
outage, fixed DNS problem or similar.
This can happen when the agent restarts/the machine
reboots during the network problem occurred.
- SR: 8606187183
After deploy/undeploy of opcmsg policies/templates the
suppressing times are lost. Messages that should be
suppressed after a deploy/undeploy of policies/templates
are shown.
- SR: B555014873
The exit code of commands executed through an ECS
annotate node and the OVO annotation server is always 0.
- SR: B555013534
When the "HP ITO Installtion Server" service
gets started you may find the following event in your
Application event log:
EventID: 0xC0001B68 (7016) - The HP ITO Installation
Server service has
reported an invalid current state 0.
- SR: 8606282754
When calling the perl interpreter without fully qualified
pathname within a automatic/operator initiated action or
a tool, you may run in problems, since the wrong perl
interpreter is used.
- SR: B555012562
The control agent might crash while processing a
distribution. This only happens in very special
situations. So far it has been seen in environments
which are using firewalls.
- SR: B555014132
During a distribution the agent may report an error like:
ITO responsible manager configuration. (OpC30-1203)
Cannot open file \usr\OV\tmp\OpC\cfgchg.
System Error Number: 13 (d) - The data is invalid.
(OpC20-63)
- SR: B555013096
several itoprocs.exe might hang instead of finishing.
- SR: 8606228146
opcntmserv.exe/itomserv.exe uses fixed width to display
the service names.
The defined values are 18 characters for the service
name and 25 for the description.
Defect Description:
PHSS_28943:
- SR: B555015785
One possible cause for opcecaas to report that there are
no more free process slots is, that all slots are in use
by applications that are running very long or might even
hang. In order to have control on this, the opcecaas
retrieves the timeout set for the "Annotate Node" in the
ECS circuit and kills the process in case it is beyond
this timeout.
- SR: B555015674
The setup may report an error in case your management
server name did change from short hostname to a long
hostname. e.g.: test123 changed to test123.test.com
or even test123.test.com is different to
test123.TEST.COM.
This caused the old setup to fail, since the management
server names are different while doing a string compare.
The new version of setup does a case insensitive compare
of the names and in case they do not match. It checks
whether at least the short hostname of the new and old
server matches.
- SR: 8606282084
Add the specific code to update also the primmgr file,
not only the opcinfo file.
- SR: 8606290346
The version in the registry key
HKLM\SOFTWARE\Hewlett-Packard\OpenView\ITO\Agent Version
is not updated in the right way during an installation
of a new version.
- SR: B555015557
The control agent has been changed in order to make the
restart of an aborted sub agent process configurable.
The control agent can be configured to restart aborted
sub agent processes. Furthermore it can be defined how
often a process should be restarted in a certain time
interval.
To configure this, you can use the following variables
in the opcinfo file:
OPC_RESTART_SUBAGENT
If set to TRUE, the control agent tries
to restart aborted sub agent processes.
The restart is done a defined number of
times (OPC_RESTART_COUNT) in a specified
period of time (OPC_RESTART_MINIMUM_RUN_TIME).
In case the process aborts more often,
it wont be restarted again.
Type/Unit : TRUE|FALSE
Default : TRUE
OPC_RESTART_COUNT
Defines how often an aborted sub agent process
should be restarted within the specified
minimum runtime. In case a process stops more often
it wont be restarted.
(See OPC_RESTART_SUBAGENT)
Type/Unit : integer
Default : 5
OPC_RESTART_DELAY
Defines the time the control agent waits
before it restarts an aborted sub agent
process. The time is specified in seconds.
Type/Unit : integer
Default : 10
OPC_RESTART_MINIMUM_RUN_TIME
Defines the time frame a sub agent process
should run without being restarted more
than specified by OPC_RESTART_COUNT.
The time is specified in minutes.
Type/Unit : integer (minutes)
Default : 60
- SR: B555015700
To deploy the new agent version to a system holding an
older version of the agent (7.00, 7.10, 7.12, 7.14),
execute the following steps:
- Call a tool that executes 'opcagt -stop' as 'HP ITO
account'. this tool execution will not come back with
either 'Successful' or 'Failed' but stay in state
'Starting'.
This is to stop the subagents. This will speed up the
later redeployment, causing this not to fail. Due to the
stopped agents, this action will never be completed in
the console.
- Wait for 2 minutes to make sure the stopping of the
agents is finished.
- 'Stop' and 'Rerun' this tool execution. The status
should switch to 'Failed', reporting that the 'Action
could not be executed because action agent is not
running'.
This is just to verify that the agents are actually
stopped.
- Now either 'Redeploy All' or deploy the new version of
the 'OpenView Operations Agent' package to the node. This
will now work without failing and during this process all
agents are restarted.
- SR: 8606299704
The error OPC20-63 comes when using environment variables
in the "File to be read" field, because the logfile
encapsulator does not correctly resolve those environment
variables.
This has been changed to resolve those variables.
The logfile encapsulator also resolves the environment
variables in the "File to be executed" field now.
- SR: 8606290908
The OVO Windows agent may loos it's AgentID when
doing a "Redeploy All" or installing an agent package
with a higher version.
Root cause for this problem is the agent setup. It
overwrites the currently existing nodeinfo file without
merging the values in there.
- SR: B555015771
In order to ensure, that the opcmona.exe always uses the
most recent libopc.dll, the dll is copied to
%OvAgentDir%\bin. Since the opcmona.exe is located in the
same place, this is the first location it will search for
the dll.
- SR: B555015325
During startup a timing issue prevented a connection from
opcmsga to opcctla, which generated the message. This
message was not immediately reported, but only after
receiving a signal. Retries will now prevent the
connection failure. If it still fails because opcctla is
really not running, the error will be reported
immediately.
- SR: H555008529
This is a timing issue, where internal structures are not
updated by the signal handler in time for proper values
to be written in the PIDS file.
An additional check for process presence has been
implemented before writing the PIDS file.
- SR: B555008674
The man pages for opcagt and opcragt now document the new
option '-version'. The message catalog was updated to
show the '-version' option in the usage string of the
opcagt and opcragt commands.
- SR: B555015378
The agent might not be able to monitor instances of the
LogicalDisk object and returns no data for it at all.
The root cause for this problem is a problem in the used
Microsoft API's, it is documented that wildcard patterns
can be used to retrieve performance data anywhere. But
the implementation only allows either wildcards or names.
A workaround for this issue has been implemented. The
monitor agent retrieves all data now and filters on the
searched values afterwards.
- SR: B555014851
opcmsga maintains an internal cache to find out the
target managers per message ID. The cache expires
after 1 hour (can be changed with the opcsvinfo variable
OPC_STORE_TIME_FOR_MGR_INFO) and then it runs into a
problem in its algorithm so that a message operation on
a non-cached message is sent again and again until the
last target manager in an internal list can be reached.
- SR: B555016032
The memory leak that shows up in opcapm.exe has several
causes.
1. The application opens a connection to the Cluster
several times, but does not close it.
2. The application reconnects to the Control Agent
several times without disconnecting and
overwrites the internal connection structure.
3. The application leaks while reading the XML
configuration file due to a memory leak
in the used BBC XML parser.
- SR: B555015484
The agent might not be able to execute any external
commands. This can be caused by disabling the 8.3
filename creation in the registry as it is described in
the Microsoft article Q174456.
The agent internally tries to convert all path names of
executed commands to 8.3, because otherwise you get
problems while executing programs with quoted parameters.
With this new version the error while converting the path
is ignored and the long path name is used instead.
- SR: B555015496
So far a queue file garbage collection was done, if more
than 256 Kbyte were unused. Now the unused space has to
be more than 256 Kbyte and more than a quarter of the
queue file size. This drastically reduces file I/O when
handling large queue files.
- SR: B555015485
To avoid that the "HP ITO account" gets locked out
on a machine providing network shares a new opcinfo
variable has been introduced.
Keyword :
OPC_WIN_DONT_USE_PATH_NWDRIVE
Description :
Force the Windows agent to scan the currently used
environment PATH to remove all mapped network
paths referenced in the PATH to avoid login
errors on the machines providing the network share.
Please note, this also removes relative PATH entries
like "." or ".." .
Type/Unit :
BOOLEAN (TRUE|FALSE)
Default :
FALSE
Used In :
opcinfo
In case you want to use these new features, please edit
the %OvAgentDir%\bin\OpC\install\opcinfo file.
Add the following entry and restart the agent:
OPC_WIN_DONT_USE_PATH_NWDRIVE TRUE
- SR: R555018185
Simplified Chinese codeset is now mapped as a valid
codeset.
- SR: B555015155
Whenever an error is added to the internal error list,
a trace entry with the DEBUG area ERRLIST is written to
the trace file.
When the error list is written to opcerror, another
trace line with the DEBUG area ERROR is written.
ERROR and ERRLIST are distinguished, since some errors
are added to the error list, but then later ignored
and therefore never appear in the opcerror file.
- SR: B555015331
opcmona holds a central table for all subprocess related
information. Advanced monitors are executed in separate
threads and could access this table in parallel, thus
overwriting each other's data. The table accesses are now
serialized by a mutex.
- SR: B555015712
In order to be able to correctly handle traps that have a
source address 127.0.0.1 the trap/event interceptor is
now able to replace the localhost address (127.0.0.1)
with the IP address of the node processing the trap.
In order to enable this, you need to add the following
line to the opcinfo file on your managed node:
OPC_RESOLVE_TRAP_LOCALHOST TRUE
- SR: 8606232431
VPO now ignores node names that contain only white space
characters without contacting the name service.
PHSS_28501:
- SR: 8606290723
The agent may report errors about commands or DLLs that
can not be found while executing commands using the OVO
Windows agent.
The cause is a modification of the system environment
variable PATH during the agent runtime. The agent
modifies the PATH in order to add certain agent specific
directories and to remove duplicate entries in order to
limit the size of the PATH. The size has to be reduced
because the length of PATH is limited to 2 kByte within
command prompts.
- SR: B555015349
The monitor agent aborts when using templates/policies
with external sources as soon as it receives a value
from the external source during checking for the
threshold with a previously received value. In this case
the received value is stored temporary and as soon as the
monitor agent
tries to process this value it aborts. The root cause
is that the montior agent tries to free up already freed
memory.
- SR: B555015308
When configuring the agent with apminfo.xml
to be cluster aware, the agent may not install newly
deployed policies correctly. The root cause
for this problem is the wrong status the policy has
internally once the Application Package Monitor(APM)
wants to inform all other sub agents. All policies are
set to status UNCHANGED and therefore no information of
the sub agents is done at all. A new status has been
introduced to
allow the APM to remember that there were modified
policies.
PHSS_27388:
- SR: B555013936
agent installation may fail with many of such messages:
T-> Tracing information: Can't get status for service: ..
in the inst.log file.
The install script may even report that the installation
was successfull but it wasn't.
The cause for this problem is that the installation tries
to stop MS Internet Information Services and OpenView
Internet Services in case OVIS has been installed and
fails when not all MS Internet Information Services are
installed. It then tries to stop the missing services in
a loop and in the end just dies or even brings a Dr.
Watson error message.
This fix just ignores the missing service and goes to the
next one.
- SR: B555014999
The problem while starting batch files on Windows NT is a
result of a problem in a Windows API.
This seems to have problems handling commands that have a
space in there path in case there are also quoted
parameters used.
The behavior have been modified in the way that the path
is converted to the 8.3 filename format.
Please note, the command itself is not converted.
This is not done, because the command name might be
used by the program itself for identification or
something similar.
The consequence is that you will still see this problem
when using commands containing a space under Windows NT.
The following will work:
c:\temp\test test1\dummy.bat test "test test",
but:
c:\temp\test test1\dummy 1.bat test "test test"
doesn't work on Windows NT.
- SR: B555014328
opcevti.exe may abort when using <$x>.
This issue is caused by traps having a timestamp
not in the range between 0 and 2147483647.
In case such a trap is received, the timestamp will be
reset to 0 now.
- SR: B555015077
Installation server always receives version A.07.10
to be installed on the target node.
Therefore it never tries to install the latest agent
package, but just A.07.10.
Since the version is no longer used to distinguish
between the different agent packages on the Management
Server it has been completely removed in
\usr\OV\databases directory on the Installation Server.
on the Installation
- SR: B555014961
The OVO agent should be able to execute tools and
scheduled commands for either the "HP ITO account" or
the SYSTEM account user. Or any other user currently
running the agent.
In any case this should work without a switch user.
To allow this, a design change has been made.
As soon as the agent executes some action now, it allows
actions to be executed as "HP ITO account" even in case
the agent is running as SYSTEM and vica versa.
Both user accounts are also mapped the way, that
no 'Switch User' is performed in case an action
contains those user names.
For OVO Windows 7.x users there is an updated
version (3.0) of the
'Using the SYSTEM account for Windows agents' white paper
available. This white paper is also part of this patch
and can be found unter
<InstallDir>\NLS\1033\Manuals\SystemAccount.pdf
after installing the patch.
- SR: B555014634
The OVO Windows agent may not pickup the SYSTEM event log
entries with ID 6008 or 6009 during the system is
rebooted.
The logfile encapsulator that is used to monitor
the SYSTEM event log uses entries in the event log
itself to find the time the agent has been stopped.
In order to achieve this, the agent writes a
"agent stop" event to the event log as soon as the
opcle.exe process goes down. This fails
during system shutdown, since the event log is already
stopped.
Now the agent is using a registry value instead of the
event log to get the correct time and this can be written
even late in the shutdown process.
This fix also prevents reading the SYSTEM event log in
order to find the "Agent Stop" event and therefore makes
the agent startup less expensive.
- SR: B555013965
The memory leak which can occur as described in Microsoft
article Q263221. Has been fixed by Microsoft
by providing a new pdh.dll, which is included in the
patch.
- SR: B555013978
In case the agent buffering and you execute the command
opcagt or "opcagt -status" you may get an output like:
"Message Agent opcmsga buffering for
testmgmtsrv.bbn.hp.com (2376) is running"
The output has been correct to look like on the UNIX
agents, where the information about the management
servers the agent is buffering to comes on the bottom.
- SR: B555014558
itodiag.exe/OvNtDiag.exe may start looping/hanging
forever.
This problem occurs in case of an error returned by
PdhGetFormattedCounterArray. This brings the program in
an endless loop trying to get the correct data. Has been
fixed by breaking the loop in error case immediately.
- SR: B555014093
opcmona may crash (UNIX) or doesn't process all SCHEDULE
templates (Windows) when using SCHEDULE templates.
This can occur when there are only spaces in
one of the schedule fields (Minute, Hour, Day of the
Month, Month, Year, Day of the Week).
You can verify this by going to the conf/OpC directory on
the node and doing an opcdcode monitor. When
there are entries like WEEKDAY " " the problem can occur.
Now the monitor agent treats sequences of spaces like
an empty string that is a wildcard and uses all valid
values in the possible range.
For WEEKDAY this is 0-6.
- SR: B555014714
You may receive a message storm with messages:
OPC INTERNAL ERROR, CANNOT GENERATE MESSAGE
1500 OF SET ZERO
Or the opcle may start taking a lot of CPU over a long
period of time.
This problem is caused by a corrupt Windows event log the
Logfile Encapsulator tries to read. With the last version
the opcle tried to read on with the next event instead of
exiting, but this doesn't work since the Windows API used
to read events reports very strange error codes once the
event log is corrupt.
The behavior of the opcle has been changed.
Now it stops monitoring the particular event log that is
corrupt and sends a message (OPC30-154) that this event
log has to be cleared and the agent must be restarted.
- SR: B555014523
nt_inst.exe may abort with Dr. Watson while trying to
install the Windows agent to a node.
This can happen when the command 'nslookup <target node>'
doesn't produce valid output on your installation server.
The installation server tries to get the WINS name of the
target node and therefore tries to get the IP address
first via GetHostByName.
Internally this failure results in an invalid pointer and
accessing this causes the crash.
To fix this, the invalid pointer is catched and goes to
the next possible case that can be tried to get the WINS
name.
- SR: B555014212
opcle may report:
ntle.c:3670: OpC internal error: Cannot generate message
1500 of set 0
System Error Number: 1966234 (1e009a) - (OpC0-1500)
The cause is the a wrong error number in the logfile
encapsulator itself.
- SR: B555014728
Currently OvNtKill.exe / itokill.exe requires the
libopc.dll as well as some other DLLs required by
libopc.dll.
This new version is indepented from the other agent
libraries.
- SR: 8606282247
Variable replacement is now performed for all Message
Defaults fields.
- SR: B555014708
Documented Microsoft API memeory leak in putenv():
Q235601. Replaced putenv() by SetEnvironmentVariable()
- SR: B555013796
itomserv.exe / opcntmserv.exe do truncate display and
service names and with Japanese multibyte characters it
can happen that the text is truncated exactly in the
middle of such a character.
This wrong character or better first part of a multibyte
character confuses the Java and the Motif UI and leads to
problems when calling this application.
With this version the truncation is done by using a
function which is aware of multibyte characters and
truncates them correctly.
- SR: B555009284
There were two different authorization algorithms which
interfered with each other. This has been consolidated
and is now checked more strictly.
- SR: 8606242614
The variables <$LOGFILE> and <$LOGPATH> were replaced
after the suppression rules were evaluated. Therefore the
comparison did not use the actual logfile name or path,
but compared the string "<$LOGFILE>" or "<$LOGPATH>".
- SR: B555014882
When monitoring Windows performance values by using the
NTPerfMon (OVO UX) / Real Time Performance (OVO Windows)
sources, there might be an error that not all instances
can be found. This is caused by an incorrect parsing of
the given instance name. This splitted one instance name
into parent instance and instance name even in case there
is no parent. This has been corrected now.
- SR: B555014319
When reading Windows Event logs the opcle.exe and
services.exe may start taking a lot of CPU time for a
longer period of time.
The cause was a misleading return code from a Windows API
which says the event log is on it's end, but it wasn't so
the logfile encapsulator permanently looped on the same
event instead of proceeding with the next one.
This behavior has been fixed by doing some additional
checks whether the end of the event log has been really
reached. If not, the current event will be skipped.
- SR: B555013872
Agent installation sets unrequired file permissions
on files in %SYSTEMROOT%\system32 for "HP ITO account".
These unrequired file permissions have been removed
and should disappear when installing this patch.
- SR: 8606286316
When retrieving a list of logfiles by using <`...`>
the logfile encapsulator may not retrieve any filename.
This has been seen when using perl scripts.
The cause for this problem is the StdInput file handle
provided to the executed program. This handle is invalid,
this doesn't matter as long as the called program doesn't
use internal redirection or piping.
Now the agent provides a valid file handle and so the
internal handling of the executed program should work
correctly.
- SR: B555014591
The defect was caused by the double conversion from the
server code set to the internal code set, once on the
agents side when it sent the internal message to opcmsga
and once by opcmsgi when it forwarded the message again.
Now, the message is converted back from the internal code
set to the server code set in opcmsga before sending the
message to the opcmsgi queue. The management server will
get the message through opcmsga in the internal code set,
and it will convert it into the server code set.
The conversion is made only if the internal code set is
different from the server code set.
- SR: B555015004
Some privileges of the OV-accounts were not deleted
after uninstalling the OV Agent.
- SR: 8606285428
SNMP V2 Traps can be received by the windows agent with
the following settings now:
- disable the Windows
SNMP Service / SNMP Trap Service on the Node
- add the following line to opcinfo:
SNMP_SESSION_MODE NNM_LIBS
- restart the agent (opcagt -start)
- SR: B555013891
Even if the IP address of the management server was
specified in the mgrconf file, it was not used except for
the primary manager. This behavior was changed to give
the mgrconf file precedence over name resolution.
- SR: B555015072
'opcagt -id 12 -stop' may not work when starting as
'Administrator' as long as the OVO agent runs with
'SYSTEM' (LocalSystem) account.
The root cause for this is an 'Access Denied' error,
since the 'Administrator' by default is not allowed to
send events for processes running as 'SYSTEM'.
- SR: H555008631
NCS agent open() and stat() calls did not handle EINTR,
so a check/loop was implemented to handle it.
- SR: B555014442
opcle may report system error 32 (Sharing violation) and
OPC20-63 when trying to access a file and after such an
error the monitoring of this file is stopped.
With this fix the opcle tries to read the file
at each interval even, when there was an error 32.
- SR: B555012827
Together with the "Windows XP" support for the VPO
Windows agent, there are some changes that are results of
changes in the way "Windows XP" works, these
changes/issues are documented below.
As default templates for "Windows XP" you can
use the same than for "Windows 2000".
List of all known changes/issues:
- Users created by the agent setup (HP ITO account &
opc_op) appear on the XP Welcome Screen
- Dr. Watson shows internal program name instead of
binary name.
This may affect you when troubleshooting processes
should be necessary.
- "itouser /local" doesn't work correctly, but
returns errors.
The tool can't be used with this switch under
Windows XP.
- Mapped network drives no longer visible to all users
- Windows XP is capable to really separate several
user environments
and therefore also the network drives mapped by one
user can't be seen by the other one
- This is a new functionality of Windows XP and
may effect user who have shared tools
placed on network drives and using them in scripts
executed by the agent.
- To avoid this don't use mapped network drives
but the complete network path to access such tools.
- Bug in snmptrap.exe (Microsoft Article Q319680, see
http://support.microsoft.com)
Microsoft confirmed this to be bug in Windows XP and
will be fixed with the first service pack.
Microsoft may also provide a HotFix for users
needing this functionality.
The result is that you can't do trap interception
or MIB monitoring.
- SR: B555014101
Monitor agent may report error OPC30-783 and the
initialisation of the first template/policy using Windows
Performance data fails.
This can be caused by a none critical problem in the
Windows registry. The monitor agent queries for the
languages, for which there are counter and object names.
When accessing some languages the system returns "NOT
FOUND" and the monitor agent internally reports this
errors and this causes the first template using this
source type to fail. All subsequent call will pass, since
the query of the registry is done only once to get all
performance counter and object names.
The opcmona ignores this error completely now.
- SR: B555014281
When using the Windows agent SDK (opcapi.h and
opcapi.lib) you may get linker unresolved symbol errors.
Some of the APIs were not exported in the DLL and LIB and
therefore you get linker errors.
The missing exports have been added now.
- SR: B555013540
Cause is a problem in the registry
That a TcpIP Device is there but
doesn't contain any entry for it's IPAddress.
Devices with such behavior are skipped now.
- SR: B555015050
The environment PATH set by the agent during executing
external commands may contain several duplicate entries
and get very large therefore.
To avoid this duplicates and to make the used PATH
shorter, the PATH is parsed and all duplicates are
removed.
- SR: B555013719
Message agent remains in buffering mode even when
the management server is available again.
The reason is that the agent wasn't able to resolve
the management server name to an IP address at startup
and the agent doesn't try again during runtime.
This has been fixed by checking for a resolvable name
every time a message should be buffered till the name can
be resolved, after this the normal checkalive mechanism
which handles buffered message takes place.
- SR: 8606187183
The opcmsg interceptor restarts after a deploy/undeploy
of policies/templates. During this process all the
policy/template information is cleaned and read again
from a temporary file. Because suppressing times are not
stored in this temporary file, this times are lost.
Now the suppressing times are taken over to the new data.
- SR: B555014873
The exit code of commands executed through an ECS
annotate node and the OVO annotation server is always 0.
The reason is an hardcoded return value in the OVO
annotation server. With this patch the annotation server
passes the received exit code to ECS.
- SR: B555013534
The "HP ITO Installation Server" service doesn't
set the status correctly at startup and therefore
it reports an invalid status when the operating
system asks for it.
- SR: 8606282754
When calling the perl interpreter without fully qualified
pathname within a automatic/operator initiated action or
a tool, you may run in problems, since the wrong perl
interpreter is used.
This is caused by the OVO agent, that contains a perl
interpreter. This perl interpreter can be found through
the PATH and when executing any command with the OVO
agent, the OVO agent directories are always added to the
beginning of the PATH. Therefore it is more likely that
the perl.exe shipped with OVO will be used.
In order to avoid this, you can install the agent with a
new option. While setting this option, the perl.exe will
be renamed to ovperl.exe.
In case you want to use the perl interpreter coming with
the OVO Windows agent, please use the environment
variable OVPERLBIN to call the perl interpreter in
future. This will ensure, that you are independent
whether you selected to rename the perl interpreter or
not. This variable is always set according to the setting
you made.
The option is configured differently for agent
installations with OVO Unix server and an OVO Windows
server.
OVO Unix server:
- do a manual or ftp installation
- before calling opc_inst.bat or opc_pre.bat
- edit opcsetup.inf, add
[Rename Perl]
TRUE
- start the installation
OVO Windows server:
- the tool setmgmtserver.exe has been extended with
a new switch /renperl
- in order to use ovperl.exe instead of perl.exe, please
call "setmgmtserver.exe /renperl /on"
- redeploy the agents that the changes are made
on the agents
The setting you made will be stored to the registry. This
ensures that a patch installation will not overwrite your
settings.
- SR: B555012562
The control agent might abort with Dr. Watson message
while it processes a distribution request.
The problem might only occur from time to time because
it depends on the system, the data transfer rate
between agent and server and other factors.
In the end it's a timing issue between two threads which
should end in a specific order and especially in
environments with firewalls it seems that the threads
often end in the wrong order which leaded to the
crash because of pointers used in both threads.
The issue has been fixed by defining the pointer
differently so that they are not lost in case
a thread closes.
- SR: B555014132
During a distribution the agent may report error number
OPC30-1203/OPC20-63 when trying to access the cfgchg
file.
The cause for this problem is that there are several
processes trying to get exclusive access to this file
at the same time. The problem has been fixed by doing a
retry for 10 times in case the error should occur with a
delay of one second.
- SR: B555013096
itoprocs.exe might hang when several itoprocs.exe
are running at the same time.
The reason is a dead lock between all of them.
This is fixed by using a central mutex to protect
critical areas.
- SR: 8606228146
opcntmserv.exe/itomserv.exe is only able to display
service names with a specific length of less than
18 characters.
This patch introduces a new option for the program
which allows to set the length via command line switch.
The new switch is called /width and takes as parameters
the width of the service name and the width of the
description.
Following example list the service name with a width
of 30 characters and completely hides the description.
> itomserv.exe /list s /width 30 0
The width switch is useful together with the /list and
/status option.
Enhancement:
No
SR:
R555020192 R555020110 R555019859 R555019153 R555018267
R555018227 R555018185 R555018043 R555017956 R555014598
H555008631 H555008546 H555008529 B555016326 B555016315
B555016133 B555016113 B555016032 B555015912 B555015785
B555015771 B555015770 B555015764 B555015763 B555015758
B555015754 B555015743 B555015741 B555015720 B555015712
B555015700 B555015674 B555015557 B555015538 B555015496
B555015485 B555015484 B555015449 B555015441 B555015378
B555015349 B555015331 B555015325 B555015308 B555015277
B555015230 B555015208 B555015199 B555015170 B555015155
B555015124 B555015088 B555015077 B555015072 B555015050
B555015047 B555015004 B555014999 B555014961 B555014944
B555014942 B555014882 B555014873 B555014851 B555014799
B555014759 B555014728 B555014715 B555014714 B555014708
B555014700 B555014667 B555014634 B555014591 B555014570
B555014558 B555014523 B555014513 B555014442 B555014378
B555014328 B555014319 B555014281 B555014212 B555014132
B555014101 B555014093 B555013978 B555013977 B555013965
B555013936 B555013891 B555013872 B555013796 B555013719
B555013540 B555013534 B555013510 B555013444 B555013096
B555012869 B555012829 B555012827 B555012562 B555009284
B555008674 B553003927 B553001962 8606303962 8606300881
8606300125 8606300119 8606299704 8606299437 8606297998
8606294668 8606293840 8606293839 8606290908 8606290897
8606290723 8606290675 8606290346 8606286654 8606286316
8606285429 8606285428 8606284806 8606282754 8606282247
8606282084 8606280846 8606279910 8606277183 8606267798
8606244582 8606242614 8606232431 8606228146 8606187183
Patch Files:
OVOPC-CLT.OVOPC-NT-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
OVOPC-CLT.OVOPC-NT-CLT,fr=A.07.10,fa=HP-UX_B.11.11_32/64,
v=HP:
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/opc_pkg.Z
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/opcsetup.inf
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/monitor/itoprocs.exe
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/cmds/itomserv.exe
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/cmds/itodiag.exe
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/cmds/itokill.exe
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/opc_version
what(1) Output:
OVOPC-CLT.OVOPC-NT-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/opc_pkg.Z:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/opcsetup.inf:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/monitor/itoprocs.exe:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/cmds/itomserv.exe:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/cmds/itodiag.exe:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/cmds/itokill.exe:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/ms/intel/nt/
A.07.10/RPC_DCE_TCP/opc_version:
None
cksum(1) Output:
OVOPC-CLT.OVOPC-NT-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
3460573857 4641001 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/ms/intel/nt/A.07.10/RPC_DCE_TCP/opc_pkg.Z
3126505163 415 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/ms/intel/nt/A.07.10/RPC_DCE_TCP/opcsetup.inf
3253082938 20525 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/ms/intel/nt/A.07.10/RPC_DCE_TCP/monitor/
itoprocs.exe
416638718 36909 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/ms/intel/nt/A.07.10/RPC_DCE_TCP/cmds/
itomserv.exe
2219370585 61484 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/ms/intel/nt/A.07.10/RPC_DCE_TCP/cmds/
itodiag.exe
4013054514 28716 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/ms/intel/nt/A.07.10/RPC_DCE_TCP/cmds/
itokill.exe
4030162770 8 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/ms/intel/nt/A.07.10/RPC_DCE_TCP/opc_version
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes:
PHSS_28501 PHSS_27388
Equivalent Patches:
ITOSOL_00217:
sparcSOL: 2.7 2.8
Patch Package Size: 4650 KBytes
Installation Instructions:
Please review all instructions and the Hewlett-Packard
SupportLine User Guide or your Hewlett-Packard support terms
and conditions for precautions, scope of license,
restrictions, and, limitation of liability and warranties,
before installing this patch.
------------------------------------------------------------
1. Back up your system before installing a patch.
2. Login as root.
3. Copy the patch to the /tmp directory.
4. Move to the /tmp directory and unshar the patch:
cd /tmp
sh PHSS_28943
5. Run swinstall to install the patch:
swinstall -x autoreboot=true -x patch_match_target=true \
-s /tmp/PHSS_28943.depot
By default swinstall will archive the original software in
/var/adm/sw/save/PHSS_28943. If you do not wish to retain a
copy of the original software, include the patch_save_files
option in the swinstall command above:
-x patch_save_files=false
WARNING: If patch_save_files is false when a patch is installed,
the patch cannot be deinstalled. Please be careful
when using this feature.
For future reference, the contents of the PHSS_28943.text file is
available in the product readme:
swlist -l product -a readme -d /tmp/PHSS_28943.depot
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHSS_28943.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions:
BEFORE LOADING THIS PATCH...
(A) Patch Installation Instructions
-------------------------------
(A1) Install the patch, following the standard
installation instructions.
For backing up the system before installing
a patch, you may use opc_backup(1m)
NOTE: MAKE SURE THAT NO AGENT OF THE PLATFORM
ADDRESSED BY THIS PATCH IS DISTRIBUTED
(either from the VPO Administrator's GUI
or from command line using inst.sh) WHILE
RUNNING SWINSTALL.
NOTE: This patch must be installed on the VPO Management
Server system, NOT on an VPO Managed Node directly.
Changes will take effect on managed nodes by means of
VPO Software Distribution (using 'Force Update' if
there is already an agent installed on the managed
node). See chapter 2 of the VPO Administrator's
Reference manual for more information.
(B) Patch Deinstallation Instructions
---------------------------------
(B1) To deinstall the patch PHSS_28943 run swremove:
NOTE: MAKE SURE THAT NO AGENT OF THE PLATFORM
ADDRESSED BY THIS PATCH IS DISTRIBUTED (either
from the ITO Administrator's GUI or from
command line using inst.sh) WHILE RUNNING
SWREMOVE.
# swremove PHSS_28943
-----End of Document ID: PHSS_28943------------------------------------------
Document ID: PHSS_28820
Date Loaded: 20030508
Title: s700_800 11.X OV ITO7.1X HP 10.20 Embedded Perf Agt A.07.15
Patch Name: PHSS_28820
Patch Description: s700_800 11.X OV ITO7.1X HP 10.20 Embedded Perf Agt A.07.15
Creation Date: 03/05/06
Post Date: 03/05/08
Hardware Platforms - OS Releases:
s700: 11.00 11.11
s800: 11.00 11.11
Products:
OpenView Operations 7.1
Filesets:
OVOPC-CLT.OVOPC-UX10-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,v=HP
OVOPC-CLT.OVOPC-UX10-CLT,fr=A.07.10,fa=HP-UX_B.11.11_32/64,v=HP
Automatic Reboot?: No
Status: General Release
Critical: No
Category Tags:
defect_repair general_release
Path Name: /hp-ux_patches/s700_800/11.X/PHSS_28820
Symptoms:
PHSS_28820:
B552002462: - Errors in EXSPI Mailbox Reports due to
improper handling of the DDF data.
B555015875: - coda_SubmitObservation fails after a new
metric was added to an object.
R555017222: - 64 bit counters have negative values if
populated from 32 bit fields that wrap.
R555019316: - Enhance codautil to allow users to view the
last set of logged values.
R555019330: - Gbl_Mem_Phys should be MBytes, not KBytes.
R555020973: - Provide a faster mechanism for creating a
partial list of instances.
R555020974: - Improve KMdataMatrix memory management for
sparse data.
Defect Description:
PHSS_28820:
Resolution:
B552002462: - The errors in the Exchange SPI Mailbox Reports
were due to improper handling of the DDF data
in Coda. This has since been corrected so
that only the valid data is returned to
Reporter.
B555015875: - In DefineMetric() call EnableMetric() if a
metric is defined after data has been
submitted. EnableMetric() will set the
pendingConfig flag so that the next time data
is submitted a new mesaArea will be created
with the additional metric definition(s).
R555017222: - Coda now handles a 32 bit wrap in a 64 bit
field, thereby showing the correct values for
any affected metric.
R555019316: - The codautil tool has been enhanced to allow
any user to view the object model
(through the use of the -obj option) for all
of the SPIs that that have been deployed and
are running on their nodes. In addition, the
-dumpds option (the datasource name is
required) will report all of the data for the
specified datasource for the last logged
interval. The command formats are:
codautil -obj
codautil -dumpds <datasource name>
R555019330: - GBL_MEM_PHYS is now reported in MBytes.
R555020973: - Added attributes to keep track of last
instance visited in the instance list. These
attributes are used to speed sequential access
to list elements.
R555020974: - Extensive modifications to minimize memory
used to store dataMatrix on server and client.
Enhancement:
No
SR:
B555015875 R555017222 B552002462 R555020973 R555020974
R555019330 R555019316
Patch Files:
OVOPC-CLT.OVOPC-UX10-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
OVOPC-CLT.OVOPC-UX10-CLT,fr=A.07.10,fa=HP-UX_B.11.11_32/64,
v=HP:
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/s700/
hp-ux10/A.07.10/RPC_DCE_TCP/perf_pkg.Z
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/s700/
hp-ux10/A.07.10/RPC_DCE_TCP/perf_version
what(1) Output:
OVOPC-CLT.OVOPC-UX10-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/s700/
hp-ux10/A.07.10/RPC_DCE_TCP/perf_pkg.Z:
HP OpenView Operations A.07.15 (03/13/03)
coda A.07.10.09 03/19/03 hpux.10_20 =*=
codaaccess A.07.10.09 03/19/03 hpux.10_20 =*=
codautil A.07.10.09 03/19/03 hpux.10_20 =*=
codaaccess A.07.10.09 03/19/03 hpux.10_20 =*=
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/s700/
hp-ux10/A.07.10/RPC_DCE_TCP/perf_version:
None
cksum(1) Output:
OVOPC-CLT.OVOPC-UX10-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
3241145509 1351251 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/hp/s700/hp-ux10/A.07.10/RPC_DCE_TCP/
perf_pkg.Z
2533018535 8 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/hp/s700/hp-ux10/A.07.10/RPC_DCE_TCP/
perf_version
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes: None
Equivalent Patches:
ITOSOL_00204:
sparcSOL: 2.7 2.8
OVOW_00026:
intelNT: 4.00 4.01
intelWIN2000: 1.00
Patch Package Size: 1280 KBytes
Installation Instructions:
Please review all instructions and the Hewlett-Packard
SupportLine User Guide or your Hewlett-Packard support terms
and conditions for precautions, scope of license,
restrictions, and, limitation of liability and warranties,
before installing this patch.
------------------------------------------------------------
1. Back up your system before installing a patch.
2. Login as root.
3. Copy the patch to the /tmp directory.
4. Move to the /tmp directory and unshar the patch:
cd /tmp
sh PHSS_28820
5. Run swinstall to install the patch:
swinstall -x autoreboot=true -x patch_match_target=true \
-s /tmp/PHSS_28820.depot
By default swinstall will archive the original software in
/var/adm/sw/save/PHSS_28820. If you do not wish to retain a
copy of the original software, include the patch_save_files
option in the swinstall command above:
-x patch_save_files=false
WARNING: If patch_save_files is false when a patch is installed,
the patch cannot be deinstalled. Please be careful
when using this feature.
For future reference, the contents of the PHSS_28820.text file is
available in the product readme:
swlist -l product -a readme -d /tmp/PHSS_28820.depot
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHSS_28820.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions:
BEFORE LOADING THIS PATCH...
(A) Patch Installation Instructions
-------------------------------
(A1) Install the patch, following the standard
installation instructions.
For backing up the system before installing
a patch, you may use opc_backup(1m)
NOTE: MAKE SURE THAT NO AGENT OF THE PLATFORM
ADDRESSED BY THIS PATCH IS DISTRIBUTED
(either from the VPO Administrator's GUI
or from command line using inst.sh) WHILE
RUNNING SWINSTALL.
NOTE: This patch must be installed on the VPO Management
Server system, NOT on an VPO Managed Node directly.
Changes will take effect on managed nodes by means of
VPO Software Distribution (using 'Force Update' if
there is already an agent installed on the managed
node). See chapter 2 of the VPO Administrator's
Reference manual for more information.
(B) Patch Deinstallation Instructions
---------------------------------
(B1) To deinstall the patch PHSS_28820 run swremove:
NOTE: MAKE SURE THAT NO AGENT OF THE PLATFORM
ADDRESSED BY THIS PATCH IS DISTRIBUTED (either
from the ITO Administrator's GUI or from
command line using inst.sh) WHILE RUNNING
SWREMOVE.
# swremove PHSS_28820
-----End of Document ID: PHSS_28820------------------------------------------
Document ID: PHCO_29052
Date Loaded: 20030507
Title: s700_800 11.04 (VVOS) vvosmta(1M) enhancement
Patch Name: PHCO_29052
Patch Description: s700_800 11.04 (VVOS) vvosmta(1M) enhancement
Creation Date: 03/04/30
Post Date: 03/05/07
Hardware Platforms - OS Releases:
s700: 11.04
s800: 11.04
Products: N/A
Filesets:
VirtualVaultOS.VVOS-AUX-AUDIT,fr=B.11.04,fa=HP-UX_B.11.04_32/64,v=HP
Automatic Reboot?: No
Status: General Release
Critical: No
Category Tags:
defect_repair enhancement general_release
Path Name: /hp-ux_patches/s700_800/11.X/PHCO_29052
Symptoms:
PHCO_29052:
vvosmta does not read from stdin. So input/output
redirection and pipes does not work with this command.
Defect Description:
PHCO_29052:
If the filename is not mentioned in the command line,
vvosmta does not try to read the message from stdin.
Resolution:
If the filename is not provided, try to read the message
from stdin.
Enhancement:
Yes
PHCO_29052:
Enable vvosmta to read from stdin.
SR:
8606308914
Patch Files:
VirtualVaultOS.VVOS-AUX-AUDIT,fr=B.11.04,
fa=HP-UX_B.11.04_32/64,v=HP:
/usr/sbin/vvosmta
what(1) Output:
VirtualVaultOS.VVOS-AUX-AUDIT,fr=B.11.04,
fa=HP-UX_B.11.04_32/64,v=HP:
/usr/sbin/vvosmta:
$Revision: Hewlett-Packard ISSL Level vvos_rose42 $
$Header: Hewlett-Packard ISSL Release vvos_r
ose $ $Date: Tue Apr 29 13:03:28 EDT 2003 $
$Source: seccmd/vvosmta.c, libsecalarm, vvos_rose, r
ose0304 $ $Date: 03/04/29 09:56:14 $ $Revisi
on: 1.1.1.3 PATCH_11.04 (PHCO_29052) $
cksum(1) Output:
VirtualVaultOS.VVOS-AUX-AUDIT,fr=B.11.04,
fa=HP-UX_B.11.04_32/64,v=HP:
1556693077 20480 /usr/sbin/vvosmta
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes: None
Equivalent Patches: None
Patch Package Size: 30 KBytes
Installation Instructions:
Please review all instructions and the Hewlett-Packard
SupportLine User Guide or your Hewlett-Packard support terms
and conditions for precautions, scope of license,
restrictions, and, limitation of liability and warranties,
before installing this patch.
------------------------------------------------------------
1. Back up your system before installing a patch.
2. Login as root.
3. Copy the patch to the /tmp directory.
4. Move to the /tmp directory and unshar the patch:
cd /tmp
sh PHCO_29052
5. Run swinstall to install the patch:
swinstall -x autoreboot=true -x patch_match_target=true \
-s /tmp/PHCO_29052.depot
By default swinstall will archive the original software in
/var/adm/sw/save/PHCO_29052. If you do not wish to retain a
copy of the original software, include the patch_save_files
option in the swinstall command above:
-x patch_save_files=false
WARNING: If patch_save_files is false when a patch is installed,
the patch cannot be deinstalled. Please be careful
when using this feature.
For future reference, the contents of the PHCO_29052.text file is
available in the product readme:
swlist -l product -a readme -d /tmp/PHCO_29052.depot
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHCO_29052.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions: None
-----End of Document ID: PHCO_29052------------------------------------------
Document ID: PHSS_28958
Date Loaded: 20030506
Title: s700_800 11.X OV ITO7.1X HP-UX 11.00/11.11 Agent A.07.20
Patch Name: PHSS_28958
Patch Description: s700_800 11.X OV ITO7.1X HP-UX 11.00/11.11 Agent A.07.20
Creation Date: 03/04/30
Post Date: 03/05/06
Hardware Platforms - OS Releases:
s700: 11.00 11.11
s800: 11.00 11.11
Products:
OpenView Operations 7.1
Filesets:
OVOPC-CLT.OVOPC-UX11-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,v=HP
OVOPC-CLT.OVOPC-UX11-CLT,fr=A.07.10,fa=HP-UX_B.11.11_32/64,v=HP
Automatic Reboot?: No
Status: General Release
Critical: No
Category Tags:
defect_repair general_release
Path Name: /hp-ux_patches/s700_800/11.X/PHSS_28958
Symptoms:
PHSS_28958:
- SR: B555015785
opcecaas may report that it runs out of free process
slots and therefore can't start anymore process.
- SR: B555008674
The opcagt and the opcragt commands have a new option
'-version'. In both man pages, this was not documented.
The new option was not part of the usage strings of
opcagt and opcragt.
- SR: H555009399
opcmona cores if invalid string is tried to be converted.
- SR: B555015479
When executing opcagt -kill there sometimes still
remains an entry for the Control Agent (opcctla) in the
pids file. This should not happen because the opcctla
isn't running anymore.
- SR: B555015626
opcif_read() doesn't return data if the signal pipe is
empty, but there is still more data in the queue file.
This can for example happen if the maximum pipe size of
8192 bytes was reached and therefore no more signal
bytes could be written into the signal pipe.
- SR: B555015496
Queue file handling was inefficient for larger queue
files.
- SR: 8606290675
Agent should report 'agent start' into the opcerror file.
- SR: R555018185
If the agent is started in Simplified Chinese codeset,
monitor agent, logfile encapsulator and message
interceptor processes will fail.
- SR: B555015449
Monitor agent may abort while receiving
templates/policies.
- SR: B555015155
Errors in the opcerror file are not written to the trace
file. This makes it difficult to find the place in the
trace file where the error occured.
- SR: B555015047
opcecaas (Annotation Server) might log error
"Illegal NULL parameter (function ) (OpC10-11)"
with many similar lines.
- SR: B555015557
The current implementation of the control agent does not
allow to restart aborted sub agent processes
automatically.
- SR: B555015199
If OPC_IP_ADDRESS (nodeinfo) and OPC_NAMESRV_LOCAL_NAME
(opcinfo) were used in combination the entry in
OPC_IP_ADDRESS was not used in the right way.
- SR: 8606282247
Logfile Encapsulator does not perform variable
replacement for all Message Defaults fields.
- SR: B555015331
The monitor agent, opcmona, may report wrong results of
executed monitor scripts or programs when using many
'advanced monitors' such as OVPERF. In some of these
cases opcmona might even abort.
- SR: B555015712
The trap/event interceptor forwards traps with the source
address 127.0.0.1 unchanged, therefore they are discarded
by the management server due to the unknown source
address.
- SR: B555015758
opcmsgi aborts if one of the set attributes has an
unmatched '<'.
- SR: 8606232431
VPO tries to resolve node names that only contain blanks
because of typo in template definition or variable
assignment. This leads to a lot of unnecessary DNS
traffic.
- SR: B555013121
When distributing agent software after the installation
of an agent patch, all agent components (opc_pkg.Z,
comm_pkg.Z and perf_pkg.Z) are distributed and installed,
although only one of the components was changed.
- SR: B555015349
The monitor agent might abort while using
templates/policies that have an external source
specified.
- SR: B555015325
opcmsga wrongly generates a message OpC30-3002 'opcctla
not running on node localhost.' when receiving a
reconfigure signal, or at exit time.
- SR: 8606300119
The logfile encapsulator does not return the last line of
a logfile immediately, if there is no carriage return
in this line.
- SR: 8606297998
The ECS engine (opcecm & opceca) might have problems
when a circuit is using global dictionaries.
(This is the case for the MessageStorm detection ECS
circuit).
PHSS_27387:
- SR: 8606189243
Utility for checking basic types of communication
between OVO server and agents is missing.
- SR: H555008602
If setting OPC_RPC_ONLY to TRUE in opcinfo, after a
while the message agent core dumps.
- SR: R555019153
When deplying policies from a Windows management server
and using the "StoreCollection" method within the Perl
or VBScript, the monitor agent might show the Policy name
instead of the given metricname or may abort.
- SR: B555014591
When the OPC_INT_MSG_FLT is set to TRUE then the filtered
message is received corrupted on the server in
a Japanese environment.
- SR: 8606275496
When installing an OVO 7.10 HP-UX Agent on a Managed
Node which already has an OVO 7.0 or 6.X agent on it,
the following Error occurs:
ERROR: Unexpected swinstall problem on system xyz. refer
to the logfile /var/adm/sw/swagent.log and
/var/adm/sw/swinstall.log on the managed node xyz for
further information on the problem.
- SR: B553003927
opcmsgi leaks memory when handling messages with custom
message attributes.
- SR: B555014942
The opcle process loops if a logfile is removed while
it is read.
- SR: H555006719
If the agent is running as a non-root user and the
management server processes are restarted, the agent
does not resume sending messages but continues to buffer
the messages.
- SR: B555014851
opcmsga sends the same message operation (e.g. an
acknowledge request created by opcmack(1) )
again and again if the related message is not in the
cache and one of the target managers can not be reached.
- SR: B555013891
In MoM environments, opcmsga does not return action
responses to SECONDARY managers, if their name is not
resolvable.
- SR: H555008631
Customer receives a lot of OpC20-61 and OpC20-63 messages
in the error logfile when using NCS agents.
- SR: B555014574
opcagt -start/-stop/-status doesn't work correctly, if
the currently running agent can't be reached over RPC.
- SR: B555014093
opcmona may crash (UNIX) or doesn't process all SCHEDULE
templates (Windows) when using SCHEDULE templates.
- SR: 8606262299
The logfile encapsulator reports that the File to be
executed for preprocessing of a logfile template failed.
This error occurs randomly and only from time to time.
You will get an error message similar to the following:
Command 'opcfwtmp /tmp/wtmp.stat /var/adm/wtmp
/tmp/wtmp.out' configured in source 'Logins (10.x/11.x
HP-UX)' returns 1. Ignoring this logfile. (OpC30-107)
- SR: R555017956
The monitor agent is terminated if you use a session
variable within the message text of a template followed
by any other variable (for example $INSTANCE).
- SR: R555018043
Japanese characters in Perl scripts within templates
don't get converted correctly into the server code set.
If the Japanese characters are used as message text for
example these characters are corrupted.
- SR: B555014715
The Control Agent slowly grows in memory usage.
- SR: B555012869
Instead of using the trap's UUID, the trap interceptor
created a new message id for all but the first trap
template. The original message id was not set correctly
in these cases.
- SR: B555014215
The port should be configurable where opctrapi listens
for incoming traps.
- SR: B555013719
Message agent doesn't stop message buffering when the
management server is available again after a network
outage, fixed DNS problem or similar.
This can happen when the agent restarts/the machine
reboots during the network problem occurred.
- SR: 8606187183
After deploy/undeploy of opcmsg policies/templates the
suppressing times are lost. Messages that should be
suppressed after a deploy/undeploy of policies/templates
are shown.
- SR: H555008275
Message Agent can hang for no apparent reason and stop
sending all messages to the Management Server regardless
of its state.
- SR: B555014667
The first lines of a logfile are not forwarded to the
message browser when using a command to discover
logfiles and the logfile was created after the first
polling interval.
- SR: B555014873
The exit code of commands executed through an ECS
annotate node and the OVO annotation server is always 0.
- SR: B555014132
During a distribution the agent may report an error like:
ITO responsible manager configuration. (OpC30-1203)
Cannot open file \usr\OV\tmp\OpC\cfgchg.
System Error Number: 13 (d) - The data is invalid.
(OpC20-63)
- SR: B555014771
The opcqchk support utility dumps message operations
(e.g. acknowledge requests from opcmack) only as hex dump
so it would be nice to have some readable output.
- SR: B555013548
The manual agent installation script opc_inst expects
compressed packages. So if you run it a second time
nothing happens because the packages are already
uncompressed.
- SR: B555009284
The authorization verification for remote start and stop
requests of the agent was sometimes unreliable. Possibly
this allowed more OVO servers to start or stop the agent,
than specified in the MoM configuration.
- SR: 8606242614
Messages are incorrectly suppressed by the logfile
encapsulator if "suppress identical output messages" is
specified and the messages differ only in the values of
<$LOGFILE> and/or <$LOGPATH>.
- SR: B555013620
Support for pmd's "u" option needed in opctrapi: use the
UDP packet's address as source of the trap.
- SR: B555014759
When enabling/disabling policies you might discover a
memory leak in the agent processes.
- SR: H555008529
If a process dies immediately after being started by the
Control Agent, it is possible that OpC30-1094 messages
start appearing in the error logfile.
Defect Description:
PHSS_28958:
- SR: B555015785
One possible cause for opcecaas to report that there are
no more free process slots is, that all slots are in use
by applications that are running very long or might even
hang. In order to have control on this, the opcecaas
retrieves the timeout set for the "Annotate Node" in the
ECS circuit and kills the process in case it is beyond
this timeout.
- SR: B555008674
The man pages for opcagt and opcragt now document the new
option '-version'. The message catalog was updated to
show the '-version' option in the usage string of the
opcagt and opcragt commands.
- SR: B555015496
So far a queue file garbage collection was done, if more
than 256 Kbyte were unused. Now the unused space has to
be more than 256 Kbyte and more than a quarter of the
queue file size. This drastically reduces file I/O when
handling large queue files.
- SR: R555018185
Simplified Chinese codeset is now mapped as a valid
codeset.
- SR: B555015155
Whenever an error is added to the internal error list,
a trace line with the DEBUG area ERRLIST is written to
the trace file.
When the error list is written to opcerror, another
trace line with the DEBUG area ERROR is written.
ERROR and ERRLIST are distinguished, since some errors
are added to the error list, but then later ignored
and therefore never appear in the opcerror file.
- SR: B555015557
The control agent has been changed in order to make the
restart of an aborted sub agent process configurable.
The control agent can be configured to restart aborted
sub agent processes. Furthermore it can be defined how
often a process should be restarted in a certain time
interval.
To configure this, you can use the following variables
in the opcinfo file:
OPC_RESTART_SUBAGENT
If set to TRUE, the control agent tries
to restart aborted sub agent processes.
The restart is done a defined number of
times (OPC_RESTART_COUNT) in a specified
period of time (OPC_RESTART_MINIMUM_RUN_TIME).
In case the process aborts more often,
it wont be restarted again.
Type/Unit : TRUE|FALSE
Default : TRUE
OPC_RESTART_COUNT
Defines how often an aborted sub agent process
should be restarted within the specified
minimum runtime. In case a process stops more often
it wont be restarted.
(See OPC_RESTART_SUBAGENT)
Type/Unit : integer
Default : 5
OPC_RESTART_DELAY
Defines the time the control agent waits
before it restarts an aborted sub agent
process. The time is specified in seconds.
Type/Unit : integer
Default : 10
OPC_RESTART_MINIMUM_RUN_TIME
Defines the time frame a sub agent process
should run without being restarted more
than specified by OPC_RESTART_COUNT.
The time is specified in minutes.
Type/Unit : integer (minutes)
Default : 60
- SR: 8606282247
Variable replacement is now performed for all Message
Defaults fields.
- SR: B555015331
opcmona holds a central table for all subprocess related
information. Advanced monitors are executed in separate
threads and could access this table in parallel, thus
overwriting each other's data. The table accesses are now
serialized by a mutex.
- SR: B555015712
In order to be able to correctly handle traps that have a
source address 127.0.0.1 the trap/event interceptor is
now able to replace the localhost address (127.0.0.1)
with the IP address of the node processing the trap.
In order to enable this, you need to add the following
line to the opcinfo file on your managed node:
OPC_RESOLVE_TRAP_LOCALHOST TRUE
- SR: 8606232431
VPO now ignores node names that contain only white space
characters without contacting the name service.
- SR: B555013121
A new tool has been introduced, that will be called
when an agent patch is installed to set the software
flag in the database for all nodes of that platform to
MODIFIED (node needs new agent software). Thus, it is
no longer necessary to use force update to install the
agent software to nodes of that platform.
The second change will check what component versions are
already installed on the agent and only distribute and
install the newer agent packages, if force update is NOT
used.
- SR: B555015349
The monitor agent aborts when using templates/policies
with external sources as soon as it receives a value
from the external source during checking for the
threshold with a previously received value. In this case
the received value is stored temporary and as soon as
the monitor agent tries to process this value it aborts.
The root cause is that the monitor agent tries to free
up already freed memory.
- SR: B555015325
During startup a timing issue prevented a connection from
opcmsga to opcctla, which generated the message. This
message was not immediately reported, but only after
receiving a signal. Retries will now prevent the
connection failure. If it still fails because opcctla is
really not running, the error will be reported
immediately.
PHSS_27387:
- SR: 8606189243
The support tool /opt/OV/contrib/OpC/opcnetchk was
introduced to allow a basic ICMP check, TCP check and
SNMP check.
- SR: H555008602
When using OPC_RPC_ONLY, ICMP handling is not
initialized, but the message agent will call
opc_pb_ping_reset() after a successful server checkalive
cycle. This causes an invalid (NULL) pointer to be
dereferenced and causes a core dump.
opc_pb_ping_reset() now has a check to see if ICMP
handling has been initialized and if not, immediately
returns from the function.
- SR: B555014591
The defect was caused by the double conversion from the
server code set to the internal code set, once on the
agents side when it sent the internal message to opcmsga
and once by opcmsgi when it forwarded the message again.
Now, the message is converted back from the internal code
set to the server code set in opcmsga before sending the
message to the opcmsgi queue. The management server will
get the message through opcmsga in the internal code set,
and it will convert it into the server code set.
The conversion is made only if the internal code set is
different from the server code set.
- SR: H555006719
When a communication to a message receiver fails, the
message agent starts buffering messages. It periodically
checks if a server is alive by sending it ICMP packets.
If the server cannot be reached with ICMP packets, no RPC
communication is attempted. Sending ICMP packets is not
possible when the agent is running as a non-root user, so
the sending function cannot actually send anything.
Therefore we also never receive any replies and the
message agent will buffer messages forever. To fix this,
the internal state of the message agent is updated after
we tried to send an ICMP packet if the agent is running
as a non-root user.
- SR: B555014851
opcmsga maintains an internal cache to find out the
target managers per message ID. The cache expires
after 1 hour (can be changed with the opcsvinfo variable
OPC_STORE_TIME_FOR_MGR_INFO) and then it runs into a
problem in its algorithm so that a message operation on
a non-cached message is sent again and again until the
last target manager in an internal list can be reached.
- SR: B555013891
Even if the IP address of the management server was
specified in the mgrconf file, it was not used except for
the primary manager. This behavior was changed to give
the mgrconf file precedence over name resolution.
- SR: H555008631
NCS agent open() and stat() calls did not handle EINTR,
so a check/loop was implemented to handle it.
- SR: B555014574
With this changes the opcctla is now be able to deal with
a running opcctla that is not reachable via RPC:
opcagt -status will display a warning if the currently
running opcctla is not reachable over RPC, but then it
will display the status according to the pids file.
opcagt -stop will also kill the unresponsive opcctla and
try to start a new one.
If opcctla is not reachable over RPC, opcagt -start will
kill all running agent processes and then start a new
opcctla which starts the agent processes.
But of course, the agent won't be able to start if RPC
is still not available at that time.
- SR: B555014093
opcmona may crash (UNIX) or doesn't process all SCHEDULE
templates (Windows) when using SCHEDULE templates.
This can occur when there are only spaces in
one of the schedule fields (Minute, Hour, Day of the
Month, Month, Year, Day of the Week).
You can verify this by going to the conf/OpC directory on
the node and doing an opcdcode monitor. When
there are entries like WEEKDAY " " the problem can occur.
Now the monitor agent treats sequences of spaces like
an empty string that is a wildcard and uses all valid
values in the possible range.
For WEEKDAY this is 0-6.
- SR: B555014215
Using the new opcinfo variable SNMP_TRAP_PORT <port_num>
opctrapi can now be configured to listen on another port
than 162. This is only effective, if traps are not
received through the NNM pmd.
- SR: B555013719
Message agent remains in buffering mode even when
the management server is available again.
The reason is that the agent wasn't able to resolve
the management server name to an IP address at startup
and the agent doesn't try again during runtime.
This has been fixed by checking for a resolvable name
every time a message should be buffered till the name can
be resolved, after this the normal checkalive mechanism
which handles buffered message takes place.
- SR: 8606187183
The opcmsg interceptor restarts after a deploy/undeploy
of policies/templates. During this process all the
policy/template information is cleaned and read again
from a temporary file. Because suppressing times are not
stored in this temporary file, this times are lost.
Now the suppressing times are taken over to the new data.
- SR: H555008275
Signal handler for SIGIO was installed before the socket
on which we receive ICMP replies was set to non-blocking
mode. An unsolicited SIGIO would trigger the signal
handler which would wait indefinitely on the socket for
data, which would never arrive. Since NCS agent is
single-threaded, all communication would stop.
The fix is in setting the non-blocking mode before
installing the signal handler, so it would not wait
forever.
- SR: B555014873
The exit code of commands executed through an ECS
annotate node and the OVO annotation server is always 0.
The reason is an hardcoded return value in the OVO
annotation server. With this patch the annotation server
passes the received exit code to ECS.
- SR: B555014132
During a distribution the agent may report error number
OPC30-1203/OPC20-63 when trying to access the cfgchg
file.
The cause for this problem is that there are several
processes trying to get exclusive access to this file
at the same time. The problem has been fixed by doing a
retry for 10 times in case the error should occur with a
delay of one second.
- SR: B555014771
The opcqchk support utility dumps message operations
(Tag: 43, like acknowledge requests from opcmack).
This tag type was not implemented in opcqchk so you only
got the hex dump output. Now you get a more readable
output like:
Size of item 1: 76 bytes. Tag: 43
Data:
Message operation = acknowledge request
Msg id = >2878c8b8-d45e-71d6-00d3-c0a8f4220000<
- SR: B555009284
There were two different authorization algorithms which
interfered with each other. This has been consolidated
and is now checked more strictly.
- SR: 8606242614
The variables <$LOGFILE> and <$LOGPATH> were replaced
after the suppression rules were evaluated. Therefore the
comparison did not use the actual logfile name or path,
but compared the string "<$LOGFILE>" or "<$LOGPATH>".
- SR: B555013620
NNM 6.2 introduced an event option to pmd - "u". This
option specifies to prefer the IP address in an SNMPv1
trap's UDP header over the contents of the SNMPv1 trap
PDU's agent_addr field.
A new opcinfo variable OPC_USE_UDP_AS_TRAP_SOURCE was
added for opctrapi. If set to TRUE, opctrapi will use the
UDP address instead of the agent_addr.
- SR: H555008529
This is a timing issue, where internal structures are not
updated by the signal handler in time for proper values
to be written in the PIDS file.
An additional check for process presence has been
implemented before writing the PIDS file.
Enhancement:
No
SR:
R555019153 R555018267 R555018227 R555018185 R555018043
R555017956 H555009399 H555008631 H555008602 H555008546
H555008529 H555008275 H555006719 B555016326 B555015785
B555015758 B555015712 B555015626 B555015557 B555015496
B555015479 B555015449 B555015349 B555015331 B555015325
B555015230 B555015199 B555015155 B555015047 B555014942
B555014873 B555014851 B555014771 B555014759 B555014715
B555014667 B555014591 B555014574 B555014215 B555014189
B555014132 B555014093 B555014014 B555013891 B555013719
B555013620 B555013548 B555013474 B555013121 B555012869
B555009284 B555009183 B555008674 B553003927 8606300119
8606297998 8606290675 8606286654 8606284806 8606282247
8606277183 8606275496 8606273258 8606262299 8606247907
8606242614 8606232431 8606189243 8606187183
Patch Files:
OVOPC-CLT.OVOPC-UX11-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
OVOPC-CLT.OVOPC-UX11-CLT,fr=A.07.10,fa=HP-UX_B.11.11_32/64,
v=HP:
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/opc_pkg.Z
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/opc_version
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/install/opcrinst
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/opc_inst
what(1) Output:
OVOPC-CLT.OVOPC-UX11-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/opc_pkg.Z:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/opc_version:
None
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/install/opcrinst:
HP OpenView Operations A.07.20 (04/14/03)
/var/opt/OV/share/databases/OpC/mgd_node/vendor/hp/pa-risc/
hp-ux11/A.07.10/RPC_DCE_TCP/opc_inst:
HP OpenView Operations A.07.20 (04/14/03)
cksum(1) Output:
OVOPC-CLT.OVOPC-UX11-CLT,fr=A.07.10,fa=HP-UX_B.11.00_32/64,
v=HP:
3997566606 15813835 /var/opt/OV/share/databases/OpC/
mgd_node/vendor/hp/pa-risc/hp-ux11/A.07.10/
RPC_DCE_TCP/opc_pkg.Z
4030162770 8 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/hp/pa-risc/hp-ux11/A.07.10/RPC_DCE_TCP/
opc_version
1993453616 120149 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/hp/pa-risc/hp-ux11/A.07.10/RPC_DCE_TCP/
install/opcrinst
2490562665 1293 /var/opt/OV/share/databases/OpC/mgd_node/
vendor/hp/pa-risc/hp-ux11/A.07.10/RPC_DCE_TCP/
opc_inst
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes:
PHSS_27387
Equivalent Patches:
ITOSOL_00223:
sparcSOL: 2.7 2.8
Patch Package Size: 14660 KBytes
Installation Instructions:
Please review all instructions and the Hewlett-Packard
SupportLine User Guide or your Hewlett-Packard support terms
and conditions for precautions, scope of license,
restrictions, and, limitation of liability and warranties,
before installing this patch.
------------------------------------------------------------
1. Back up your system before installing a patch.
2. Login as root.
3. Copy the patch to the /tmp directory.
4. Move to the /tmp directory and unshar the patch:
cd /tmp
sh PHSS_28958
5. Run swinstall to install the patch:
swinstall -x autoreboot=true -x patch_match_target=true \
-s /tmp/PHSS_28958.depot
By default swinstall will archive the original software in
/var/adm/sw/save/PHSS_28958. If you do not wish to retain a
copy of the original software, include the patch_save_files
option in the swinstall command above:
-x patch_save_files=false
WARNING: If patch_save_files is false when a patch is installed,
the patch cannot be deinstalled. Please be careful
when using this feature.
For future reference, the contents of the PHSS_28958.text file is
available in the product readme:
swlist -l product -a readme -d /tmp/PHSS_28958.depot
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHSS_28958.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions:
BEFORE LOADING THIS PATCH...
(A) Patch Installation Instructions
-------------------------------
(A1) Install the patch, following the standard
installation instructions.
For backing up the system before installing
a patch, you may use opc_backup(1m)
NOTE: MAKE SURE THAT NO AGENT OF THE PLATFORM
ADDRESSED BY THIS PATCH IS DISTRIBUTED
(either from the VPO Administrator's GUI
or from command line using inst.sh) WHILE
RUNNING SWINSTALL.
NOTE: This patch must be installed on the VPO Management
Server system, NOT on an VPO Managed Node directly.
Changes will take effect on managed nodes by means of
VPO Software Distribution. See chapter 2 of the VPO
Administrator's Reference manual for more information.
NOTE: The VPO Agent consists of several components that
are patched individually. This patch updates only the
Event/Action component. Therefore the software
distribution will tell, that the agent software is
updated to A.07.10, not to A.07.20. Also the version
of the ITOAgent bundle will still be A.07.10. You can
verify the installed version of the components on the
agent system using opcragt -agent_version.
For example:
opcragt -agent_version hpbbln8
Node hpbbln8.bbn.hp.com:
OPC_INSTALLED_VERSION = A.07.20
PERF_INSTALLED_VERSION = A.07.10
COMM_INSTALLED_VERSION = 2.5.3.9
Done.
(B) Patch Deinstallation Instructions
---------------------------------
(B1) To deinstall the patch PHSS_28958 run swremove:
NOTE: MAKE SURE THAT NO AGENT OF THE PLATFORM
ADDRESSED BY THIS PATCH IS DISTRIBUTED (either
from the ITO Administrator's GUI or from
command line using inst.sh) WHILE RUNNING
SWREMOVE.
# swremove PHSS_28958
-----End of Document ID: PHSS_28958------------------------------------------
Document ID: PHSS_28922
Date Loaded: 20030506
Title: s700_800 11.X OV ECS3.20 Intermediate Patch April 2003
Patch Name: PHSS_28922
Patch Description: s700_800 11.X OV ECS3.20 Intermediate Patch April 2003
Creation Date: 03/05/02
Post Date: 03/05/06
Hardware Platforms - OS Releases:
s700: 11.00 11.11
s800: 11.00 11.11
Products:
OpenView Event Correlation Services 3.2x
Filesets:
OVECS.OVECS-RUN,fr=A.03.20,fa=HP-UX_B.11.00_32/64,v=HP
Automatic Reboot?: No
Status: General Release
Critical: No
Category Tags:
defect_repair general_release
Path Name: /hp-ux_patches/s700_800/11.X/PHSS_28922
Symptoms:
PHSS_28922:
PTSfc03108:
csmerge.pl only works with the correct perl
versions (not the version installed out of the
box with HP-UX 11.11)
PTSfc03112:
Many asynchronous perl call fails in case of event
burst with error alarm having text
"Asynchronous function call/s had errors or timed out"
Defect Description:
PHSS_28922:
PTSfc03108:
sheBang path in csmerge.pl points to
/usr/contrib/bin/perl ,which is not the
supported version[5.005_03 version]. This
will result in an error when csmerge.pl is
executed.
Resolution:
sheBang path is changed to
/opt/OV/bin/Perl/bin/perl in csmerge.pl
PTSfc03112:
In case of event burst perl call fails with an
error alarm stating
"Asynchronous function call/s had errors or timed out"
Resolution:
The timeout of the annotation node used in composer
circuit is increased from 30s to 5m to accomodate
the network delays
Enhancement:
No
SR:
F333003108 F333003112
Patch Files:
OVECS.OVECS-RUN,fr=A.03.20,fa=HP-UX_B.11.00_32/64,v=HP:
/opt/OV/contrib/ecs/csmerge.pl
/opt/OV/bin/genannosrvr
/etc/opt/OV/share/conf/ecs/circuits/Composer.eco
what(1) Output:
OVECS.OVECS-RUN,fr=A.03.20,fa=HP-UX_B.11.00_32/64,v=HP:
/opt/OV/contrib/ecs/csmerge.pl:
csmerge.pl $Revision: /main/VINDHYAS/6 $ $Date: 2003
/03/28 10:44 UTC $
ECS 3.2 (A.03.02)
/opt/OV/bin/genannosrvr:
ECS.c $Revision: /main/VINDHYAS/1 $ $Date: 2002/10/2
3 06:39 UTC $
ECS 3.2 (A.03.20)
ECS REV: A.03.20
annoio.c $Revision: /main/NILGIRIS/1 $ $Date: 2000/0
1/18 07:31 UTC $
sockstack.c $Revision: /main/VINDHYAS/2 $ $Date: 200
2/09/30 18:50 UTC $
DFIX12001:sockstack.c 01Sept99
value.c $Revision: /main/VINDHYAS/6 $ $Date: 2003/03
/27 04:55 UTC $
Intermediate Patch April 2003
trace.c $Revision: /main/VINDHYAS/1 $ $Date: 2002/10
/28 12:29 UTC $
fatal.c $Revision: 1.13 $ $Date: 1998/05/07 02:26:33
$
log.c $Revision: /main/VINDHYAS/1 $ $Date: 2002/10/2
8 12:28 UTC $
mesgbuf.c $Revision: 1.18 $ $Date: 1998/05/07 00:28:
35 $
sequence.c $Revision: 1.17 $ $Date: 1998/01/28 03:39
:17 $
vstring.c $Revision: /main/2 $ $Date: 1998/08/21 10:
54 UTC $
perl_interp.c $Revision: /main/VINDHYAS/19 $ $Date:
2003/03/21 05:55 UTC $
yaccpar 1.8 (Berkeley) 01/20/91
/etc/opt/OV/share/conf/ecs/circuits/Composer.eco:
ECS 3.2 (RELEASE)
cksum(1) Output:
OVECS.OVECS-RUN,fr=A.03.20,fa=HP-UX_B.11.00_32/64,v=HP:
2392483613 16756 /opt/OV/contrib/ecs/csmerge.pl
521329944 1585152 /opt/OV/bin/genannosrvr
1214773388 1407172 /etc/opt/OV/share/conf/ecs/circuits/
Composer.eco
Patch Conflicts: None
Patch Dependencies: None
Hardware Dependencies: None
Other Dependencies: None
Supersedes: None
Equivalent Patches:
PHSS_28921:
s700: 10.20
s800: 10.20
PSOV_03254:
sparcSOL: 2.8 2.9
Patch Package Size: 800 KBytes
Installation Instructions:
Please review all instructions and the Hewlett-Packard
SupportLine User Guide or your Hewlett-Packard support terms
and conditions for precautions, scope of license,
restrictions, and, limitation of liability and warranties,
before installing this patch.
------------------------------------------------------------
1. Back up your system before installing a patch.
2. Login as root.
3. Copy the patch to the /tmp directory.
4. Move to the /tmp directory and unshar the patch:
cd /tmp
sh PHSS_28922
5. Run swinstall to install the patch:
swinstall -x autoreboot=true -x patch_match_target=true \
-s /tmp/PHSS_28922.depot
By default swinstall will archive the original software in
/var/adm/sw/save/PHSS_28922. If you do not wish to retain a
copy of the original software, include the patch_save_files
option in the swinstall command above:
-x patch_save_files=false
WARNING: If patch_save_files is false when a patch is installed,
the patch cannot be deinstalled. Please be careful
when using this feature.
For future reference, the contents of the PHSS_28922.text file is
available in the product readme:
swlist -l product -a readme -d /tmp/PHSS_28922.depot
To put this patch on a magnetic tape and install from the
tape drive, use the command:
dd if=/tmp/PHSS_28922.depot of=/dev/rmt/0m bs=2k
Special Installation Instructions: None
-----End of Document ID: PHSS_28922------------------------------------------
Document ID: PHNE_28183
Date Loaded: 20030506
Title: s700_800 11.00 R6.11.00 SNAplus2 Link/Common, TN3270 patch
Patch Name: PHNE_28183
Patch Description: s700_800 11.00 R6.11.00 SNAplus2 Link/Common, TN3270 patch
Creation Date: 03/03/16
Post Date: 03/05/06
Hardware Platforms - OS Releases:
s700: 11.00
s800: 11.00
Products:
SNAplus2-Link R6.11.00
SNAplus2-Common R6.11.00
3270-Common R6.11.00
Filesets:
SNAplus2-Link.SNAP2-LINK,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Link.SNAP2-LINK,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Link.LINK-64ALIB,fr=R6.11.00.200,fa=HP-UX_B.11.00_64,v=HP
SNAplus2-Link.LINK-64ALIB,fr=R6.11.00.010,fa=HP-UX_B.11.00_64,v=HP
SNAplus2-Link.LINK-32ALIB,fr=R6.11.00.200,fa=HP-UX_B.11.00_32,v=HP
SNAplus2-Link.LINK-32ALIB,fr=R6.11.00.010,fa=HP-UX_B.11.00_32,v=HP
SNAplus2-Common.SNAP2-SAM,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-SAM,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-NOTES,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-NOTES,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-CORE,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-CORE,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-ASCIIADM,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-ASCIIADM,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-ADMIN,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.SNAP2-ADMIN,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
SNAplus2-Common.CORE-64SLIB,fr=R6.11.00.200,fa=HP-UX_B.11.00_64,v=HP
SNAplus2-Common.CORE-64SLIB,fr=R6.11.00.010,fa=HP-UX_B.11.00_64,v=HP
SNAplus2-Common.CORE-32SLIB,fr=R6.11.00.200,fa=HP-UX_B.11.00_32,v=HP
SNAplus2-Common.CORE-32SLIB,fr=R6.11.00.010,fa=HP-UX_B.11.00_32,v=HP
3270-Common.MANAGERS,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
3270-Common.MANAGERS,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
3270-Common.MANAGER-64SLIB,fr=R6.11.00.200,fa=HP-UX_B.11.00_64,v=HP
3270-Common.MANAGER-64SLIB,fr=R6.11.00.010,fa=HP-UX_B.11.00_64,v=HP
3270-Common.MANAGER-32SLIB,fr=R6.11.00.200,fa=HP-UX_B.11.00_32,v=HP
3270-Common.MANAGER-32SLIB,fr=R6.11.00.010,fa=HP-UX_B.11.00_32,v=HP
3270-Common.3270API,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,v=HP
3270-Common.3270API,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,v=HP
Automatic Reboot?: Yes
Status: General Release
Critical:
Yes
PHNE_28183: PANIC HANG
PHNE_27404: HANG
PHNE_25181: PANIC HANG
PHNE_24341: PANIC HANG
PHNE_24093: PANIC
PHNE_22721: PANIC
PHNE_22492: PANIC
PHNE_21336: PANIC
PHNE_21001: HANG ABORT
PHNE_19527: PANIC HANG CORRUPTION
PHNE_18000: HANG ABORT
PHNE_17229: PANIC HANG CORRUPTION
Category Tags:
defect_repair enhancement general_release critical panic
halts_system corruption
Path Name: /hp-ux_patches/s700_800/11.X/PHNE_28183
Symptoms:
PHNE_28183:
(1) JAGae32590/8606268352
SNA error log reports buffer shortage conditions, while
query_buffer_availability shows that the total used was
modest but the total available varied wildly. There is a
loss of sessions under moderate stress.
(2) JAGae48710/8606284767
System Panic from nba_mm_free/nlu_remove_partner_lu. Crash
during a mainframe reset, stack shows nba_mm_free called
from nlu_remove_partner_lu and internally there are two
ASSERTs from../../p/vappn/ntlavl.c Line: 203.
(3) JAGae53557/8606289627
An ASSERT is logged in the SNA log file as follows -
LOG Message 4096 - 13, Subcode: 0 - 10
Log category: EXCEPTION Cause Type: Internal
System: gepcpro6
ASSERT:
File name = ../../p/sappn/nlu.c
Line number = 677
Expression = FALSE
this Assert is logged when -RSP is received on INITSELF
(dep LU6.2)
ASSERTs in nlu.c (lines 412 and 677) under stress.
Subsequent conversations cannot be started.
(4) JAGae55276/8606291438
After setting the SNAPLUS2 buffer availability to 64 Mb with
snapadmin set_buffer_availability,buf_avail=64000000, the
system experienced a hang (subsequently TOC'd by Service
Guard). TOS on CPU's was :-
wait_for_lock
sl_retry
vba_account_buffer_out or vba_account_buffer_in
(5) JAGae64021/8606300552
Service Guard issued a TOC due to SNA deadlock on one
processor.
Stack for the processor was :-
sna_pcisdlc_v0_get_rw_lock+0xf4
vwan_r_put+0x2c4
putnext+0xcc
streams_put+0xf8
wan_connect+0xc0
J3525_F_Level1Up+0x108
J3525_F_VHSI_IntHandler+0x904
J3525_F_InterruptHandler+0x1e0
sapic_interrupt+0x2c
mp_ext_interrupt+0x318
ihandler+0x904
vwan_write_fsm+0x68
vwan_w_put+0x2ac
putnext+0xcc
sna_sdlc_vsi_stream_lw_service+0x4c
sq_wrapper+0x94
str_sched_mp_daemon+0x310
str_sched_daemon+0x4bc
main+0xbd8
PHNE_27404:
(1) JAGae09395/8606242140
Multi-threaded APPC program hangs up under stress. There
are a variety of ASSERTs logged that indicate a confusion
between the threads.
(2) JAGae31243/8606266999
There is an unsolicited RTM sent to host when LUA session
UNBINDs. The LUA application does not provide response
times and the content of the RTM is junk.
(3) JAGae31416/8606267173
LUA application hangs up with an RUI_WRITE not responded to.
Trace shows that messages are not received at the DLC having
been delivered from Host.
PHNE_25181:
(1) JAGad77708/8606208520
System panic due to data page fault.
Data Page Faults: libsixs.a nms_nah_process_mu_rsp+0x100
Crash in nms_nah_process_mu_rsp
Stack is:-
trap 0xa48
thandler 0xb7c
nms_nah_process_mu_rsp 0x100
nms_msm_queue_handler 0x694
nba_dispatch_process 0xd0
nba_scheduler 0x208
vpr_stream_lr_svc 0x160
sq_wrapper 0xb8
str_sched_up_daemon 0x2b0
str_sched_daemon 0xf4
main 0x94c
$vstart 0x34
(2) JAGad87318/8606218168
System panic in nrm_bld_and_send_deact_sess.
Stack trace:-
nrm_bld_and_send_deact_sess
nrm_send_deactivate_session
nrm_ps_to_rm_rec
nrm_queue_handler
nba_dispatch_input
nba_dispatch_process
nba_scheduler
vpr_stream_uw_drive_scheduler
vpr_stream_uw_svc
sq_wrapper
(3) JAGad91086/8606221968
Query_mode fails after doing initialize_session_limitstrace.
(4) JAGad96946/8606227887
SNA hang up on large system (32 CPUs), stack shows deadlock
in APPN code with re-entrancy due to Streams running service
procedure in interrupt context (on the ICS stack).
PHNE_24341:
(1) JAGad47872/8606178646
Delays in startup script due to ioscan.
(2) JAGad46888/8606177656
SAM install of SNAplus2 can be very slow and appear to be
hung on large systems.
(3) JAGad71274/8606202100
SNAplus2 hangs up under stress (one CPU locked)and attempts
to stop SNAplus2 fail. A system reboot is required to
recover. Stack trace shows sna_ndg_v0_get_rw_lock .
(4) JAGad69070/8606199884
There are software verification errors on an OS-Core fileset
when SNA is also installed.
(5) JAGad76402/8606207227
Panic in SNAPlus2 Router, nba_mm_free, while snapstop is in
progress.
Stack trace :-
trap 0xd9c
thandler 0xd24
nba_mm_free 0x68
vpr_create_entity 0x1f8
0x0afvpr_stream_open 0x3a8
open_wrapper 0x98
csq_protect 0x120
osr_open 0xb44
pse_open 0xf8
hpstreams_open 0x50
hpstreams_clone_open 0x214
call_open_close 0x144
opend 0x128
spec_open 0x78
vns_copen 0x4c
vn_open 0xa8
copen 0xd0
open 0x3c
syscall 0x6f8
syscallinit 0x54c
(6) JAGad71590/8606202416
T1 timer settings in xsnapadmin do not work. The t1_timeout
parameter in define_tr_ls (or Ethernet or FDDI) is ignored
by SNAplus2. It is exposed in xsnapadmin in the advanced
dialog for the LS. The T1 timer is fixed at 1 second.
PHNE_24093:
(1) JAGad65199/8606195995
Panic in wait_for_lock called from nps_free_rcb_mus.
Crash occurs just after ASSERT in nbmmgr.c line 589.
Stack is:-
wait_for_lock+0x2cc
sl_retry+0x1c
v0_get_rw_lock+0x38
vpr_stream_close+0x38
close_wrapper+0x6c
csq_protect+0x120
osr_pop_subr+0x1fc
osr_close_subr+0x324
hpstreams_close_int+0xd40
hpstreams_close+0x2c
call_open_close+0x1f8
closed+0xb0
spec_close+0x54
vn_close+0x48
vno_close+0x20
closef+0x68
exit+0x1108
rexit+0x28
syscall+0x200
$syscallrtn+0x0
WARNING: SNA ASSERT:
09:30:06 30 APR 2001
File: ../../c/cappn/nbmmgr.c
Line: 589
Condition: FALSE
PHNE_22721:
(1) JAGad35609/8606166322
RUI does not support segmented delivery from the lower
layers. Incorrect data is returned on RUI read call when
RU is segmented. Note that similarly, incorrect data could
be passed through on a PU concentrated connection.
(2) JAGad34156/8606164859
Autodial does not work with SNAPlus2 using EISA PSI card.
(3) JAGad41008/8606171744
Incorrect usage of incoming Attach is not being rejected.
(4) JAGad42004/8606172744
snapadmin command query_available_tp sometimes reports that
a TP is still available even after each of its multiple
copies have ended.
(5) JAGad43064/8606173811
Panic in qenable when stopping sna (snap stop).
qenable is called from vpr_timer_tick.
panic+0x14
report_trap_or_int_and_panic+0x80
interrupt+0x1d4
$ihndlr_rtn+0x0
qenable+0xc
vpr_timer_tick+0x30
str_timeout+0xc4
invoke_callouts_for_self+0xc0
sw_service+0xb0
mp_ext_interrupt+0x144
ivti_patch_to_nop3+0x0
sth_get_sth+0xc8
hpstreams_write+0x18
spec_rdwr+0x18c
vno_rw+0x1d4
write+0x104
(6) JAGad34698/8606165404
snapfilter does not work on 64-bit machines
(7) JAGad40346/8606171082
A MultiThreaded HLLAPI application on 11.0 will hang
up on the first verb call as the signal from the 3270
emulator does not work correctly. This problem applies
to both TN3270 and SNAP3270 emulators
(8) JAGad42503/8606173243
There is a changed behaviour of HA_WAIT verb call. The WAIT
verb can return early eg. when PS or OIA is updated.
(9) JAGad49181/8606179959
HLLAPI Query Host Update verb returns code of 65535 .
After applying fix for JAGad42479 (signals broken with
Multi Threaded HLLAPI library) the PAUSE verb works
incorrectly. Subsequent verbs timed out after 30 seconds
with a return code of -1.
(10) JAGad54056/8606184850
HLLAPI application used with TN3270 is intermittently
receiving an RC=9 when issuing Connect_PS verb. If a
HLLAPI application is started at the same time as
3270 (typically with a shell script) then the first
HLLAPI verb (typically connect) can fail with a return
code of 9. If so all subsequent HLLAPI verbs are rejected
with a return code of 1.
PHNE_22492:
(1) JAGad21193/8606151854
System panic from nrm_process_cnos_reply.
Panic with following stack:
panic+54:
report_trap_or_int_and_panic+80:
trap+0xDB8:
$RDB_trap_patch+40:
nrm_process_cnos_reply+270:
nrm_ps_to_rm_rec+780:
nrm_queue_handler+94:
nba_dispatch_input+2D0:
nba_dispatch_process+0xB4:
nba_scheduler+21C:
vpr_stream_lr_svc+154:
sq_wrapper+94:
str_sched_up_daemon+1C4:
str_sched_daemon+390:
main+7A0:
$vstart+48:
$locore+94:
Also an assert logged at same time as panic:
WARNING: SNA ASSERT:
07:25:11 20 JUL 2000
File: ../../p/sappn/nrmcnos.c
Line: 292
Condition: wait_req != NULL
(2) JAGad13997/8606144657
Panic in SNAPlus2 requiring further analysis
Crash with stack as follows
panic+0x10
report_trap_or_int_and_panic+0xe8
trap+0xa48
thandler+0xb7c
ncs_destroy_an_cb_if_required+0x1cc
ncs_reset_xid_info+0xa0
ncs_xid_processing+0x600
ncs_reset_ls+0x738
ncs_fsm_ls_int+0x1ed4
ncs_fsm_ls_ext+0x138c
ncs_pc_to_cs_signals+0x88
ncs_queue_handler+0xd0
nba_dispatch_input+0x298
nba_dispatch_process+0xa4
nba_scheduler+0x1b0
vpr_stream_lr_svc+0x134
sq_wrapper+0xb8
str_sched_up_daemon+0x1f8
str_sched_daemon+0xf4
main+0x958
$vstart+0x34
$locore+0x74
Also following asserts logged:-
WARNING: SNA ASSERT:
22:46:03 23 JUN 2000
File: ../../p/vappn/ncsfsmln.c
Line: 104
Condition: !an_cb->num_active_conns
WARNING: SNA ASSERT:
15:20:08 24 JUN 2000
File: ../../p/sappn/ncsxidtg.c
Line: 1272
Condition: tg_number == 0
(3) JAGab57822/8606101599
Customer has migrated a configuration to R6. However they
have taken the default TN Server port number during the
migration.
When they start the node they get a BIND failure message.
(4) JAGad03916/8606134781
Error in SAM when attempting to install PCI card in
SNAPlus2 :-Internal
Error: Cannot find tunable parameter 'num_boards_3525' in
the
internal kernel configuration data structure.
(5) JAGad12558/8606143222
NOF applications which reference AP_GLOBAL_TIMEOUT fail to
compile:
e.g.
cc: 'nofsample1.c', line 238: error 1000: Unexpected
symbol: '0x01'.
cc: 'nofsample1.c', line 238: error 1588: 'NB_BYTE'
undefined.
(6) JAGad25288/8606155977
snapadmin status_node command incorrectly causes error in
logfile.
With SNAplus2 R6 ,when issuing the snapadmin status_node
command it causes an error to be logged in the sna.err
file as follows:-
------ 10:22:24 BST 15 Aug 2000 -------------------
NOF Message 1 - 8, Subcode: 1 - 10
Log category: EXCEPTION Cause Type:
External
System:
hppimd54
Lost connection to NOF application. Application handle =
0x010000DE
Cause: The Node has lost its connection to a NOF
application
or an administration tool. This can occur if a NOF
application
ends without terminating its NOF connection.
Action: Check that the NOF application issues
DISCONNECT_NODE or CLOSE_FILE before terminating.
PHNE_21336:
(1) JAGad01270/8606132121
Customer is using the T(nn) option on snapsend and
snapreceive. However, the file transfer program is
returning with an error code saying the file transfer
was canceled.
File transfer works OK without the T(nn) option.
(2) JAGac95609/8606130719
System panic after call to npo_sess_limit_data_lock_mgr().
(3) JAGac86152/8606128605
LU Type 2 is inactive on the HP side but active on the IBM
side. No DACTLU or segmentation error is seen in the DLC
trace, but sna.err file shows the following:
APPN Message 512 - 171, Subcode: 0 - 10
Session segmentation error.
....
RU : (3 bytes)
(4) JAGac78841/8606128040
System panics after receiving an UNBIND during initialise
session limits processing. The stack shows that
nba_mm_free was called from nlu_remove_mode. A similar
problem was also reported in JAGac95609.
(5) JAGab74691/8606105945
3270 user gets PROG checks 432 (bracket race) when typing
ahead together with ASSERT in vbaccess line 324 and log 52
(internal error copying buffer). User needs to do reset on
3270 to recover.
(6) JAGab71415/8606104367
Data page fault panic when attempting to start sdlc
link station on EISA 8 port ACC card on 11.00.
Stack tracestack trace for event 0
crash event was a panic
panic+0x14
report_trap_or_int_and_panic+0x4c
trap+0x7a4
$call_trap+0x38
svphpoll+0x44
vhs_timer_proc+0x58
sna_sdlc_nba_dispatch_input+0x8c
sna_sdlc_nba_dispatch_process+0xa4
sna_sdlc_nba_scheduler+0x11c
sna_sdlc_vsi_stream_uw_service+0x3e0
sq_wrapper+0x90str_sched_mp_daemon+0x150
str_sched_daemon+0x3a8main+0x544
$vstart+0x34
$locore+0x90
PHNE_21001:
(1) JAGac78479/8606127677
Cannot reliably stop SDLC link station or port.
Once the port gets stuck in the stopping state
the sna has to be recycled.
(2) JAGab78721/8606108322
snap2adm coredumps when more than 250 dependent LU62s are
defined.
PHNE_20734:
(1) JAGab84956/8606112613
A customer application linked with libmgrdce.1 library
stopped working with no errors reported.
(2) JAGab83814/8606111714
The snapmsgf program fails to run with error message
'unable to open the catalog file ./vsnamsg.cat'.
(3) JAGab75474/8606106417
With release R6.11.00.200 of SNAPlus2 various user
application errors can be reported if the user is not root
or a member of the sna group. Example errors are:
snap3720 error msg 'Unable to open the configuration file'
CPIC application fails in CMINIT with RC=20
CPIC_PRODUCT_SPECIFIC_ERROR.
(4) JAGab65304/1653307843
When using snaptrcfmt -D to decode a DLC trace that
includes a BIND, the hex information of the BIND is
modified slightly. The SLU Name length is decremented by
1. There are also many minor problems with the details of
the BIND.
PHNE_19527:
(1) JAGab83920/8606111815
Customer's QLLC link fails to come up due to error in the
XID exchange. The sna.err log shows XID 3 error during
negotiation.
(2) JAGab83645/8606110894
Application cannot reconnect to HLLAPI session after
killing 3270 program. The hllapi verb connect_ps() fails
with HARC_SYSTEM_ERROR, and extended error code1 = 2000 and
extended error code2 = 0000.
(3) JAGab79003/8606108556
K580 system panics on reboot with the following error
messages:
System Panic:
9245XB HP-UX (B.10.20) #1: Sun Jun 9 06:31:19 PDT 1996
panic: (display==0xb800, flags==0x0) nio_initialize :
l_io_vec.iov_base == NULL
PC-Offset Stack Trace (read across, most recent is 1st):
0x00276f44 0x003fdab0 0x00404968 0x001297c4 0x00240ae8
0x002a7ea8 0x002a6ae4 0x00215088 0x002147e0 0x00295368
0x00295444 0x00295444 0x00295444 0x002951c0 0x00295d20
0x002f316c 0x000c7014 0x00183960
End Of Stack
It was not possible for the kernel to find a process
that caused this crash.
jCj1D
Dumpsys() called
(4) JAGab76276/8606106834
When the 'SNAplus2 Installation' icon is double clicked in
11.00 SAM, the main menu screen takes too long (30-45
seconds on a T500 machine) to come up.
(5) JAGab75469/8606106415
When starting an SDLC link the system panics with
following stack trace:
stack trace for event 0
crash event was a panic
FUNC
panic+0x10
report_trap_or_int_and_panic+0xe8
trap+0xa48
$RDB_trap_patch+0x20
sna_sdlc_vba_ips_putb+0x18
sdl_send_set_mode_frame+0x138
sdl_psecs+0xf88
sdl_prtmg+0x106c
sdl_wrxfr+0x5a8
sdl_receive_proc+0x80
sna_sdlc_nba_dispatch_input+0x254
sna_sdlc_nba_dispatch_process+0xa4
sna_sdlc_nba_scheduler+0x11c
vsi_stream_uw_service+0x3e0
sq_wrapper+0xb8
str_sched_mp_daemon+0x114
str_sched_daemon+0x1f4
main+0x97c
$vstart+0x34
(6) JAGab75335/8606106333
SAM SDLC/PSI configuration info. inconsistent when
user cancels SDLC changes. Exact sequence leading
up to this condition is shown below:
The following SNAplus2 configuration was done on a
K-class machine with two PSI cards configured for SDLC:
A. User selects 'remove component' to de-configure all
psi cards.
B. The SDLC configuration screen pops up. Now clicks on
the 'View Entire Card' button.
C. The 'View Entire Card' screen pops up. On the SDLC
configuration screen, the user selects all the cards
displayed and clicks on 'Remove'.
D. Now the user presses OK on the 'View Entire Card'
screen. It disappears.
E. Then the user clicks on 'Cancel' in the SDLC
configuration screen to abort any changes. The main menu
screen now displays the SDLC PSI as 'Not Added'. This is
incorrect since the user aborted SDLC configuration.
As a workaround, the user can get the actual configuration
state of the SDLC /PSI by doing a 'Options->Refresh List'.
(7) JAGab75114/8606106196
When adding SDLC(PSI) on T500 system using SAM, SAM builds
vmunix_test but then exits with a core dump and the error
message:
UNEXPECTED EXIT: process LANG=C LC_ALL=C
/usr/sam/lbin/samx -C -p 18985 /opt/sna/sam/C/mainMenu.ui
exited with a non-zero exit status. sam: Warning:
selector, SLIST_LABEL, is wider than the max window width.
sam: Warning: selector, SLIST, is wider than the max
window width.
sh: 20111 Memory fault(coredump)
(8) JAGab74185/8606105839
An LUA application receives an UNBIND indication from an
established LU-LU session. It responds with RSP.UNBIND and
encounters the following error:
primary return code: LUA_STATE_CHECK (0x0200)
secondary return code: LUA_NO_RUI_SESSION(0x81000000)
(9) JAGab73055/8606105164
While running SNAplus2 R6 on HPUX 11 a link failure
occurred. The sna.err log file recorded message 4096 - 13
(ASSERT: File name = ../../p/vhlwr/vhshmod.c ). This
caused the link to go down for one minute disrupting the
current file transfer. Following the ASSERT, the sdlc
driver executed ten consecutive svphrx() calls. After nine
of these failed the SDLC driver called svphclos() to stop
the link.
(10) JAGab71689/8606105222
The following message is noted periodically in sna.err
file:
10:07:29 EDT 09 Aug 1999 4096-5(0-0) P (UH2010D3)
PID 3733 (snaperrlog)Log parameter mismatch.
Message number = 512 - 137
The message 512 - 137 then follows with a question mark
for the sense code.
(11) JAGab71519/8606104163
This system panic occurred on a H20 running 10.20 with
128MB.
panic+0x10
report_trap_or_int_and_panic+0xe8
trap+0xa48
$call_trap+0x20
nsm_delete_session_id+0x104
nsm_cleanup_lu_lu_session+0x7b8
nsm_fsm_status+0x1d4c
nsm_process_deactivate_session+0x140
nsm_process_record_from_rm+0x1a8
nsm_queue_handler+0x68
nba_dispatch_process+0x114
nba_scheduler+0x208
vpr_stream_uw_svc+0x58c
sq_wrapper+0xb8
str_sched_up_daemon+0x2b0
str_sched_daemon+0xf4
main+0x974
$vstart+0x34
(12) JAGab70700/8606106276
If you are using 3270 with LANG set to de_DE.roman8 on a
character terminal, then the terminal goes into file
transfer mode whenever the host locks the 3270 keyboard.
This is a revisit of SR 165300756 - the fix for this SR
should have also been applied to the MOTIF emulators, the
CSV library and the style file printers.
Other symptoms that have been seen include the sending of
incorrect control fields to the host during file transfer
- i.e. additions of P() and H() parameters to the
IND$FILE command.
(13) JAGab68385/8606102608
Code inspection has shown that the SNAplus2 Kernel drivers
do not have the MGR_IS_MP flag set on the streams_info
structure. This has not caused problems but may affect
performance on MP systems.
(14) JAGab65516/5003455253
When a SIGTERM signal is sent to HLLAPI program it core
dumps and cma_dump.log is produced.
(15) JAGab65399/4701429621
Dependent APPC LU does not get BIND from VTAM after the
link is re-established following an outage.
(16) JAGab65397/4701428094
SNAplus2 R6.10.20 SDLC link fails to start if DSR is not
present when link station is started. Only recovery is to
reboot the system and ensure that modem signals are present
before starting the link. Loss of DSR causes the same
problem.
The following errors are logged:
SDLC Message 768 - 17, Subcode: 1 - 11
Log category: EXCEPTION Cause Type: External
System: dqserv1
DSR was not active when activating port.
Return code = 0x0003
Cause: An error occurred on a port. The port is configured
as Non-switched but DSR was not present.
Action: Check whether the configuration setting of
non-switched is correct.
If the port is correctly configured, check the modem and
link hardware.
Check other messages for further diagnostics.
SDLC Message 768 - 94, Subcode: 0 - 11
Log category: EXCEPTION Cause Type: External
System: dqserv1
SDLC port device driver reported an error.
DLC name = SDLC0
Port name = mapport
Port number = 0x00000000
Return code = 0x0003
Detailed return code = 0x0020
Cause: An error occurred on an SDLC port. The detailed
return code provides more information on the error,
as follows:
0x0020 DSR failure
0x0021 General hardware failure
0x0022 Modem power off
0x0023 CTS failure between frames (4 wire)
0x0024 CD failure between frames (4 wire)
Action: Check the detailed return code shown; check for
previous exception messages providing more information
about the failure.
(17) JAGab65393/4701424812
Using snap2adm ascii admin tool on SNAPLUS2 R5.1 to view
the config, and then selecting OK and exiting the program
causes serious modifications to the configuration to be
made. The scenarios where this happens are as follows:
A)
Enter snap2adm and go to CONNECTIVITY -> DLC
select OK and press return, press exit
Changes are made to:
mu_credit = 4 from original 8 - in define_sdlc_dlc
link_deact_timer = 0 from original 30 - in define_sdlc_ls
use_default_tg_chars = YES from original 'NO' - in
define_sdlc_ls
B)
Enter snap2adm and go to CONNECTIVITY -> DLC -> PORT
select OK and press return, press exit
Changes are made to:
mu_credit = 4 from original 8 - in define_sdlc_dlc
address = 0x00 from original 0xc1 - in define_sdlc_port
opt1 = NONE from original NRZI - in define_sdlc_port
link_deact_timer = 0 from original 30 - in define_sdlc_ls
use_default_tg_chars = YES from original 'NO' - in
define_sdlc_ls
The setting for encoding sometimes shows as NRZ in
snap2adm, even if NRZI is in the config.
C)
Enter snap2adm, go to CONNECTIVITY -> DLC -> PORT -> LS
select OK and press return, press exit
Changes are made to:
link_deact_timer = 0 from original 30 - in define_sdlc_ls
use_default_tg_chars = YES from original 'NO' - in
define_sdlc_ls
Each iteration was done on a new copy of the configuration.
The 'in define_sdlc....' statements on the right hand side
indicate the area of the configuration file that changed.
(18) JAGab65392/4701424069
This is a potential problem that applies to any SNAPlus2
multi-threaded application that requires to use signals and
can cause the application to fail because of incorrect
signal handling behaviour.
(19) JAGab25510/5003467290
Token Ring connection hangs when frames exceeding the
largest frame size returned in RIF are dropped by the
network.
PHNE_18000:
(1) JAGab71162
Symptoms of problem: After installing patch, PHNE_19527,
and selecting Diagnostics, node tracing, sdlc level 2
tracing in xsnapadmin, activated the node. Then, ls went
into starting status, then retry pending. Several seconds
later, the system panicked. The following stack was
produced when the machine crashed:
panic+0x14
report_trap_or_int_and_panic+0x4c
trap+0xef4
$RDB_trap_patch+0x38
sdl_reset_port+0x13c
sdl_poll_port+0x3ec
sdl_hms_ctl_proc+0xb4
sdl_receive_proc+0x14c
sna_sdlc_nba_dispatch_input+0x254
sna_sdlc_nba_dispatch_process+0xa4
sna_sdlc_nba_scheduler+0x11c
sna_sdlc_vsi_stream_uw_service+0x3e0
sq_wrapper+0x90
str_sched_up_daemon+0x440
str_sched_daemon+0x1f0
main+0x6e0
$vstart+0x34
$locore+0x90
The following asserts were also seen in the console log:
WARNING: SNA ASSERT:
17:10:45 25 AUG 1999
File: ../../p/vsdlc/sdlcsigi.c
Line: 1495
Condition: pcb->resetting == FALSE
WARNING: SNA ASSERT:
17:11:15 25 AUG 1999
File: ../../p/vsdlc/sdlcsigi.c
Line: 1375
Condition: pcb->alert != NULL
(2) 5003466367
The customer has been running SNAplus2 version R6.11.00
with PHNE_17229 for about a month with no problems. Today,
after about 20 3270 users become active the system hangs.
Needed to do a TOC to recover. The following stack trace
yielded from TOC:
nba_get_q_head+0x0
nch_sscp_receive+0x9b8
nba_dispatch_input+0x298
nba_dispatch_process+0xa4
nba_scheduler+0x1b0
vpr_stream_lr_svc+0x134
sq_wrapper+0x90
str_sched_up_daemon+0x440
str_sched_daemon+0x1f0
main+0x6e4
(3) 5003455659
The customer has migrated an R4 config file to R5.1. Under
R5.1 the migrated token ring link station remains inactive
after a host IPL. This differs from the R4 behaviour.
(4) 5003424085
There appears to be no way within the snaplus2 management
utilities to list unassigned lu_0_to_3 LU's or assigned
lu_0_to_3 LUs.
Under SNAplus if you were assigning sessions to a 3270
user, and the LU you assign was already assigned to another
user, a warning message was printed advising this LU was
already assigned to another user.
Under SNAplus2 there is no warning nor message of any kind
giving this indication. In the case of this customer, there
are 200+ 3270 users and there is no way to determine which
3270 LU's have already been assigned to users without
manually parsing the sna_node.cfg and sna_domn.cfg file.
This is an enhancement request to add a command to
snapadmin and/or xsnapadmin to enable a listing of all 3270
unused LU's, and a list of assigned 3270 LU's and the users
and/or pools they are assigned to.
(5) 5003392456
With SNAPlus2 there is no way to deduce the number of
sessions that were concurrently busy on a particular LU and
what remote LU and effective TP these sessions are actually
connected to.
This is an enhancement request to add this functionality.
(6) 4701430017
A user edited the /stand/system file to configure SNAplus2
SDLC on a NIO machine with no PCI bus. He included the
psi0 and sixl but no sixw driver which is not required on a
non-PCI machine. After re-generating the kernel the SDLC
worked. Afterwards, he used SAM to check the SNAplus2
configuration and found that SDLC was marked as 'not added'
and at this point SDLC stopped working.
(7) 4701427021
Multiversion patch required for versions R6.11.00.010 and
R6.11.00.200 of SNAplus2.
(8) 4701425561
R6.11.00 on a V-Class system: After several hours of APPC
activity, (about 10 incoming allocates per second), APPC
TP's fail to load, with error messages 512-257(0-10)
logged.
In addition, a system panic has occurred while the user
APPC application was terminated.
Although these two problems are very different by nature,
it has been determined that they are closely related due
to internal mechanisms in SNAPlus2 in its communication
via Streams putq messages.
The stack trace for the panic was as follows:
panic+0x14
report_trap_or_int_and_panic+0x80
trap+0xa8c
nokgdb+0x8
putq_owned+0x2a0
putq+0x1c
vba_track_putq+0x4c
vpr_stream_output_msg+0x40c
vpr_delete_entity+0x43c
vpr_stream_close+0x1a8
close_wrapper+0x6c
csq_protect+0x120
osr_pop_subr+0x220
osr_close_subr+0x324
hpstreams_close_int+0x314
hpstreams_close+0x2c
call_open_close+0x1f8
closed+0xb0
spec_close+0x54
vn_close+0x48
vno_close+0x20
closef+0x64
exit+0x324
rexit+0x28
syscall+0x480
$syscallrtn+0x0
(9) 4701425355
While running reliability tests with new ACC driver and
SNAplus2 the lab hit a system hang a couple of times when
using qllc over X25.
(NB: ACC use their own X.25 stack which uses nli2zcom
module.)
When examining the TC with q4 we found the system stuck at
same routine, sna_q_v0_get_rw_lock, in libsixp.a. Here is
how the q4 stack looks.
sna_q_v0_get_rw_lock+0xc8
vql_stream_read_input+0xdc
putnext+0x50
N2Z_F_data_ind+0x38
N2z_iev_pass_data_up+0x114
N2z_ReadEvent_Recvd+0x209c
Zc_putq+0x5c
nacc0_receive_data+0x140
(10) 4701424069
This is a potential problem that applies to any SNAPlus2
multi-threaded application that requires to use signals and
can cause the application to fail because of incorrect
signal handling behaviour.
(11) 4701421081
Running a SNAPlus2 API using Java Native Interface (JNI)
within a Java Virtual Machine (JVM) will hang, calling
getmsg() in the path manager.
The problem is seen when using CPI-C within HP JVM.
The problem is that even though the path manager calls
fcntl to make the stream to the router non-blocking, this
does not affect getmsg() and so the path manager still
hangs.
(12) 4701418707
When using CPI-C without side information, outgoing
attaches sometimes fail because validation has been
unintentionally turned on.
(13) 1653307231
The customer has problems printing a long file from a host.
Gets the following error:
---------------- 19:34:03 SAT 21 May 1999------------------
APPN Message 512 - 452, Subcode: 0 - 10
Log category: EXCEPTION Cause Type: SNA
System: cnbv
LU type 0, 1, 2, or 3 session ended abnormally - protocol
error.
Sense code = 0x10020000
LFSID = 018E0000
Cause: An LU type 0, 1, 2, or 3 session ended abnormally
(with the sense code shown) because of a protocol error.
Action: Contact network support personnel with details of
the problem.
The sense code 0x10020000 means 'RU Length Error'.
The problem was not seen when similar files were printed
with SNAplus.
(14) 1653305805
One processor on a two processor box running R6 over 10.20
hangs which then causes cmcld to TOC the box to preserve
system integrity.
Top of stack for hanging process is:
FUNC PC
v0_get_rw_lock+0xb8 0.0x3cc4a8
vpr_route_ips_on_route+0x40 0.0x4094e0
vds_rcv_buffers_available+0x1a0 0.0x3e1720
vds_receive_proc+0x674 0.0x3e47fc
nba_dispatch_input+0x298 0.0x5af050
nba_dispatch_process+0xa4 0.0x5af184
nba_schedule_process+0x134 0.0x5af5ec
nba_send_ips+0x308 0.0x5afd3c
(15) 1653305284
After installing 11.0 PHCO_17631 SAM patch, SAM failed
to install SNAplus2. The problem appeared when the user
started 'SNAplus2 Installation' from SAM. After attempting
to add SNAplus2, SAM internal error occurred. The SAM log
file /var/sam/log/samlog showed:
Assigning driver/pseudo driver 'sixt' the planned value of
1. The current pointer undefined but is needed for this
operation.
^A^ALocation: da_list_gets():305^AError Code: 1302
This made it impossible to install SNAplus2.
Also, SAM SNAplus2 Installation may show the 'SNAplus2
Environment' as 'Not Added', even though this is known
not to be the case.
(16) 1653301382
A customer defines more than 16 LU0-3 type LUs, then stops
and starts SNAplus2 but does not activate the node.
Subsequent attempts to start xsnapadmin cause the config
daemon to terminate.
(17) 1653301333
On some European machines, the following assert is given
when node tracing is turned on during daylight saving time:
LOG Message 4096 - 13, Subcode: 1 - 0
Log category: EXCEPTION Cause Type: Internal
ASSERT:
File name = ../../cdart41/cdiag/vtdfile.c
Line number = 241
Expression = 0
Then when trying to format the resulting sna1.trc file with
snaptrcfmt, another assert appears:
LOG Message 4096 - 13, Subcode: 1 - 1
Log category: EXCEPTION Cause Type: Internal
ASSERT:
File name = ../../p/vsig/vtfread.c
Line number = 76
The trace file cannot be formatted.
(18) 1653299685
An LS is started following node startup but not following a
link failure. The customer tries to enable the reactivation
on link failure feature via snap2adm:
->>> CONNECTIVITY >>> LS >>> ADVANCED >>> Reactivate LS
after failure and then successfully saves/exits the
program.
The LS now fails to startup and the sna.err log shows:
APPN Message 512 - 91, Subcode: 0 - 10
Log category: PROBLEM Cause Type: Config
System: mbnahp4
CP name on an XID3 received from adjacent node does not
match expected CP name.
Sense code = 0x0806002C
Port name = SDLC1
LS name = SDLC
Received CP name = CCNNET01.NOTGLO
(19) 1653299073
After upgrading from R4.4 to R6.10.20 on a T600 system,
the SDLC link could no longer be activated and the lan
performance is severely degraded.
When the SNA resources are activated, the following error
messages are logged:
----------------------- 15:32:40 WET 19 mars 1999
SDLC Message 768 - 107, Subcode: 0 - 11
Log category: EXCEPTION Cause Type: External
System: centurix
SDLC write timer retry limit has been exceeded.
DLC name = SDLC0
Port name = SDLCP0
Port number = 0x00000000
Cause: An attempt to transmit a frame using an SDLC port
has timed out.
This may indicate a problem with the SDLC adapter or with
the modem and cabling. The port is stopped.
Action: Check the modem and communications link.
----------------------- 15:32:40 WET 19 mars 1999
APPN Message 512 - 60, Subcode: 0 - 10
Log category: PROBLEM Cause Type: SNA
System: centurix
An active link station has failed.
Port name = SDLCP0
LS name = SDLCL0
Adjacent CP name = 0000000000000000000000000000000000
Cause: An active link station has failed
----------------------- 15:33:29 WET 19 mars 1999
SDLC Message 768 - 17, Subcode: 1 - 11
Log category: EXCEPTION Cause Type: External
System: centurix
DSR was not active when activating port.
Return code = 0x0003
Cause: An error occurred on a port. The port is configured
as Non-switched but DSR was not present.
Also the syslog.log is filled up with the messages
'lan3_process_read_completion: Received out of sequence'
The impact of all the above is that the LAN card becomes
very slow to the point where the system becomes unusable.
The only way to recover LAN traffic is to reboot the system
without starting SNA at all.
(20) 1653293878
Invokable TP failing to start with following error messages
logged.
------------- 10:52:14 GMT 10 Feb 1999 ----------------
NODE Message 16384 - 0, Subcode: 10 - 10
Log category: EXCEPTION Cause Type: Internal
System: LR1875
Internal system error. Errno = 7
Action: Provide support services with the audit and error
log files, and trace files if available.
------------- 10:52:14 GMT 10 Feb 1999 ----------------
APPN Message 512 - 257, Subcode: 0 - 10
Log category: PROBLEM Cause Type: Config
System: LR1875
Dynamic load of TP failed.
Sense code = 0x07000000
LU alias = DFKC
TP name = lr229bci
(21) 1653245936
In contrast to SNAplus, SNAplus2 doesn't generate trace for
SDLC level 2.
PHNE_17229:
(1) 5003450536
Snapmigrate does not migrate partner (remote) LUs correctly
if they have a different network ID than adjacent CP.
In this scenario there are partner LUs accessible via cross
domain routing that have a different network than the
network ID if the adjacent node. The netid for these LUs
appears to be ignored by snapmigrate, and they are added to
the same define_adjacent_len_node resource, as the LUs
that actually reside on the adjacent system.
In the course of testing the fix to this bug with the
customer config files, we discovered another problem --
that duplicate partner LU definitions are not handled
correctly by the migration program. In R4 it is possible
(though deprecated) to define separate aliases that refer
to the same partner LU. In R5 this is not allowed --
duplicate define_partner_lu entries in the node
configuration file are logged as an error and prevent the
product from starting.
(2) 5003446971
Data page fault panic in nbm_free_buffer while running
simple SNA tests over two LAN interfaces between the three
machines running SNAplus2.
(3) 5003446278
RJE received files from host showed incorrect
user/group names.
(4) 5003442426
If the domain config file > 64K, then corruption can occur
when the file is updated or sent to other servers in CS
domain.
(5) 5003441717
The snaperrlog process can be left lying around when the
SNAplus2 daemon is not started. Attempting to restart the
SNAplus2 software using 'snap start' will fail (because the
snaperrlog process is still there from a previous run).
(6) 5003434860
Problems with xsnapadmin configuration/administration when
used from client system.
(7) 5003433862
If you migrate a SDLC leased connection from SNAPlus to
SNAPlus2, the migrated configuration will no longer work
(LS will not become 'Active'). The snapmigrate utility
produces an incorrect configuration.
(8) 5003427518
Deletion of TP definition in sna_tps file gives
ASSERTs when the original file contained duplicate entries
for that TP.
(9) 4701413054
System panic - Data Page Fault at
nsm_process_record_from_ss+130
(10) 4701407528
Possible memory leak problem in PCI SDLC Glue found by code
review.
(11) 4701400929
snapmigrate uses default X25 adapter number without any
warning messages.
(12) 4701399527
Assert errors are produced when the host sends a USSMSG10
screen to a LU configured for LU6.2. The ASSERTS are in
fact benign, and will cause no problems with the integrity
of the system.
The ASSERTs only occur when the USSMSG10 screen is
segmented, and greater than around 500 bytes in size.
(13) 4701399279
The PSI firmware header is not recognized by the snapwhat
command.
(14) 4701392670
When in client /server configuration, various ASSERT
messages recorded in the sna.err log file from the SLIM
component, followed often by a crash of the SLIM.
This causes general unpredictable behaviour of the system
when the master server goes down and is restarted.
(15) 1653289686
If using a TN3270 (not E) client and hit the clear
key while TN Server is presenting an SSCP screen,
then the client will lock up.
The host may respond with an error message.
(16) 1653289603
If using a TN3270 (not E) client and hit the clear key
while TN Server is presenting an SSCP screen, TN Server
forwards the clear key to the host(sends an empty RU
on SSCP-LU session).
The host may respond with an error message.
(17) 1653285478
Unable to start SDLC link after the SDLC PSI driver has
error to close the card .
sna.err logged : Message 32769 - 66
Cause: The SDLC driver's state is not valid to process
svphclos()
(18) 1653280123
snap2adm displays port settings incorrectly, and
occasionally does not allow the user to modify ports
and link stations.
(19) 1653279703
Assert errors logged when RJE workstation is started
due to RTM request being received.
This has no affect on operation other than bad
entries in the error log.
(20) 1653276543
3270 session can hang, even if you stop and restart the
snap3270 emulator.
If you take a trace of the problem, you will see that a
NOTIFY is not sent when the 3270 emulator is stopped or
started.
(21) 1653273979
Enhancement to allow multiple PUs to be used on a secondary
leased link. This means that if an SDLC port is connected
to a leased line, you can have multiple LS's active over
the port at the same time.
(22) 1653267179
An Application can fail to start a remote LU62 transaction,
because an invalid user ID is specified on the Attach,
when AP_SAME is specified on the ALLOCATE verb.
(23) 1653264531
Any API call will fail after a fork is invoked as follows:-
- issue a call to any API (the most common is likely to
be CSV)
- fork()
- issue a call to another API
If the call to the 'other API' is the first call to that
API made by the application, then the API call will fail.
Defect Description:
PHNE_28183:
(1) JAGae32590/8606268352
The algorithm to look at the kernel variable `freemem` to
provide an indication of what extra buffers could be
acquired is not accurate in HP-UX 11.x where `freemem` is
clearly measuring something more volatile in the virtual
memory manager.
Resolution:
Code is changed to set the maximum buffer usage to
one-fourth of the total physical memory which is given by
the kernel variable phys_mem_pages.
(2) JAGae48710/8606284767
If we receive an UNBIND on a CNOS session after we have
received an ATTACH for a CNOS conversation but not yet
received the CNOS GDS then we try to release the remote LU
control block twice in error.
Resolution:
Code is modified to ensure that the remote LU control block
is not released twice in this race condition.
(3) JAGae53557/8606289627
Trying to process two conversations simultaneously causes
an error in accessing common data.
Resolution:
Code is modified to ensure that the PS instance is locked
correctly.
(4) JAGae55276/8606291438
The Streams handling code did not break out of getq loop
after using putbq to put a message back on the queue.
Resolution:
The code is modified to handle the condition.
(5) JAGae64021/8606300552
Stack shows nested locking calls because the put procedure
is run in interrupt context. The lock was already taken on
the write side.
Resolution:
Code is modified to use a lower service procedure in the
SDLC PCI glue to handle all messages other than M_FLUSH.
PHNE_27404:
(1) JAGae09395/8606242140
There is a narrow window where one thread is unlocking a
mutex and clearing a variable using a non-atomic write
causing corruption of the variable. Another thread can read
the corrupt variable and match the thread id in error.
Resolution:
Closed the window by adding another interlocked test.
(2) JAGae31243/8606266999
The LUA layer does not zero the parameter that is used on
status messages to report response times to the lower layer
of APPN. RTM times are thus being accumulated in error.
Resolution:
Code has been modified to clear the uninitialised field to
prevent RTM times being taken on LUA sessions.
(3) JAGae31416/8606267173
The LUA layer in SNAP APPN used too small an initial credit
for the session so that when back pressure was applied by
the application not issuing an RUI_READ, the incoming queue
to the DFC layer was closed. This then prevented any
messages being received, including IPRs, thus hanging up the
send side as well.
Resolution:
The correct initial credit (2xpacing window+1) was used to
queue the messages in the correct layer (the RUI layer) thus
freeing the DFC layer to handle the send direction
correctly.
PHNE_25181:
(1) JAGad77708/8606208520
System panics due to uncorrelated NMVT RSP received.
Resolution:
The code has been changed to ignore uncorrelated NMVT RSP.
(2) JAGad87318/8606218168
Reuse of internal session control block handle occurred when
TP_ENDED was received a long time after a session has
terminated.
Resolution:
Clear cross references from RCB control blocks to SCB
control blocks when a session is deactivated.
(3) JAGad91086/8606221968
Rework of mode control block handling released the mode
control block in error after the CNOS command.
Resolution:
The mode control block is recreated when the CNOS command
completes.
(4) JAGad96946/8606227887
There was a misunderstanding of HP-UX multi-threaded Streams
implementation.
Resolution:
The code was modified to avoid running Streams service
procedure on interrupt stack. It was instead deferred to a
later thread.
PHNE_24341:
(1) JAGad47872/8606178646
Call to 'ioscan' in the init script 'snaplus2' takes time to
gather hardware info.
Resolution:
Instead of getting hardware info by actually scanning the
hardware, hardware information is retrieved from kernel data
structures.
(2) JAGad46888/8606177656
Call to 'ioscan' in SNA SAM takes time to gather hardware
info on large systems.
Resolution:
Instead of getting hardware info by actually scanning the
hardware, hardware information is retrieved from kernel data
structures.
(3) JAGad71274/8606202100
MP locking problem with interrupt driving message into read
stream of NDG while processing a message on write stream for
the same link station.
Resolution:
Change Streams Synchronization to use the QUEUE-PAIR scheme,
instead of QUEUE synchronization level, to allow Streams to
protect the NDG DLPI Glue from this parallel processing.
(4) JAGad69070/8606199884
Due to duplicate sna man pages, there are software
verification errors on a Core fileset when SNA is also
installed.
Resolution:
The duplicate man pages in the Core fileset have now been
removed.
(5) JAGad76402/8606207227
A window existed in the SNAPlus2 Node termination code that
did not protect against an open from an application.
Resolution:
Changed code to implement correct locking of SNA router
resources when opening a Stream.
(6) JAGad71590/8606202416
SNAplus2 does not program the LLC2 with the configured value
using the appropriate DLPI extension message.
Resolution:
The code was changed to issue a DLPI message from SNAplus2
to LLC2 when initialising the connection. This allows the
T1 timer value to be taken from the SNAplus2 configuration
file. Note that the default is 5 seconds. Added support
for new message in tracing(snaptrcfmt -i -d output).
PHNE_24093:
(1) JAGad65199/8606195995
Reuse of a control block handle when conversation
terminated a long time after session has been terminated
caused the problem. This may probably follow an
overnight reset at the mainframe.
Resolution:
When the session is terminated, the coupling between
conversation and session control blocks is cleaned up
correctly.
PHNE_22721:
(1) JAGad35609/8606166322
RUI does not support segmented RU delivery from the lower
layers within the APPN node. However, it was not setting
a flag to tell PC to re-segment RU's before passing them
upwards.
Resolution:
RUI was getting incorrect data from lower layers when
Ru was segmented. Code change made to request unsegmented
delivery on PLU session at OPEN_PLU time.
(2) JAGad34156/8606164859
The RTS line was not getting raised during autodial. This
prevented the card from sending the dial string to the
modem.
Resolution:
The firmware code is changed to raise RTS during autodial.
(3) JAGad41008/8606171744
There was an incorrect implementation of define statements
in the code.
Resolution:
Correct macro processing of the define statements used in
the code.
(4) JAGad42004/8606172744
Count of TPs were not kept correctly.
Resolution:
Corrected count of TPs maintained between service manager
and NOF stub.
(5) JAGad43064/8606173811
There is a narrow time window where timeout expires while
we are stopping SNA.
Resolution:
Code changed to protect against the narrow time window,
during time out expiry while stopping SNA
(6) JAGad34698/8606165404
snapfilter does not work on 64 bit systems because 32-bit
version is supplied incorrectly.
Resolution:
A 64-bit version of snapfilter has been built.
(7) JAGad40346/8606171082
The hllapi use of Signals between the emulator and the
hllapi library was not compatible with kernel threads.
Resolution:
A FIFO is used instead of signals, to send a message from
3270 to the HLLAPI library.
(8) JAGad42503/8606173243
Though JAGad01270 resolved this, the libhapi.1 version was
omitted from the build.
Resolution:
libhapi.1 library has been rebuilt to include the fix .
(9) JAGad49181/8606179959
Pause Verb was not handled correctly in the fix for
JAGad42479.
Resolution:
Leave FIFO open between verbs and catch application
termination at exit.
(10) JAGad54056/8606184850
The cause of the problem is the timing window in
initialisation. The problem only shows up when fix
for JAGad42479 (remove signals due to clash with
MultiThreaded) is applied.
Resolution:
Distinguish between initialisation and termination
window with new flag in code and return 1 in case of
initialisation window.
PHNE_22492:
(1) JAGad21193/8606151854
If a TP terminates at the same time as a CNOS race
condition is occurring some bugged processing
occurs which causes a panic.
The last error log before the panic shows that a TP had
indeed just ended - the assert shows that
we were in a CNOS race condition.
Resolution:
Modify the code in nrm_process_cnos_reply so that a check
is made to see whether wait_req is valid.
If not then we know the pending list has already been
cleared up so we don't need to do anything.
(2) JAGad13997/8606144657
System panic occurs because a second attempt is made to
free
an AN (Adjacent Node) control block , which has already
been
successfully freed.
Resolution:
The exact reason that the code is attempting to free the
control block twice is unknown despite significant
code-reading
and diagnosis of trace.
However we have put a fire-wall into the code to ensure
that it will not attempt a double-free in future.
(3) JAGab57822/8606101599
The customer is trying to use port 23 for TN Server even
though it is still configured for normal Telnet
in /etc/inetd.conf.
Resolution:
The customer needs to set up snapinetd (as described in
the section on define_tn3270_access in the
Administration Command Reference). The Migration guide
does mention this in the section on Preparing Configuration
for migration. There is also a message output by
snapmigrate telling the user to configure snapinetd:
To use the Telnet port number 23 for a TN3270 server
access record you will need to run snapinetd on this node.
However, to make things clearer the message from
snapmigrate has now been enhanced:
WARNING: To use the Telnet port number 23 for a TN3270
server access record you will need to run
snapinetd on this node.
Please see the define_tn3270_access section in
the Administration Command Reference for
more information.
(4) JAGad03916/8606134781
The problem was in SNA SAM. In the file 'sna.tm' the task
'sna_get_sync_wan_version' was getting a wrong version
number once the
patch PHNE_20195 installed.
Resolution:
The task, sna_get_sync_wan_version, in the SNA SAM file
'sna.tm' has
been changed to collect the version number correctly.
(5) JAGad12558/8606143222
The definition of AP_GLOBAL_TIMEOUT in values_c.h
includes a cast to an NB_BYTE. NB_BYTE which
is an internal type not exposed to end users.
Resolution:
Remove the NB_BYTE cast in values_c.h.
(6) JAGad25288/8606155977
This is caused by the snapadmin program issuing a
query_node_limits verb prior to issuing the actual
status verbs. Before issuing the query_node_limits
snapadmin issues a connect_node. However, no
disconnect_node verb is issued before issuing all
the status verbs (which in turn require a connect_node
to the appropriate node).
Consequently, having finished the status verbs, there is
still an outstanding NOF connection (from the
query_node_limits).
Resolution:
Before we issue our status verbs, check to see whether we
are already connected to a node. If so, issue a
disconnect_node.
PHNE_21336:
(1) JAGad01270/8606132121
The snapsend and snareceive programs issue a HLLAPI file
transfer command. The HLLAPI library issues this to the
emulator and waits for it to issue a sigusr1, implying the
file transfer is complete. The emulator sends sigusr1 to
any HLLAPI application whenever the presentation space or
OIA is updated. Unfortunately this means that the HLLAPI
library picks up one of these automatic signals and
interprets it as FT complete. It then checks the shared
memory return code and finds HARC_SUCCESS (0) as opposed to
HARCFT_COMPLETE (3). This then causes it to issue the
error code. The file transfer continues and may appear to
complete successfully.
Resolution:
Add in a flag to the shared memory that tells us whether
we still have a verb in progress. Ensure that the FT code
in HLLAPI sets this flag before issuing the FT and
that it checks the flag before assuming FT is complete.
Also ensure that emulator clears the flag once it has
really completed the verb.
(2) JAGac95609/8606130719
The analysis of the partial dump points to the
following sequence of events that led to the panic:
- STOP_NODE quiesce comes into the NOF - goes to TRS,
then the first PS
- PS COPR builds and sends RESET_SESSION_LIMITS(all) to
itself
- PS COPR locks all the modes, and issues ALLOCATE to PS
(not traced)
- PS sends ALLOCATE_RCB to RM
- we cannot contact remote, so RM cannot activate a
session
- ALLOCATE_RCB(failure) comes back to PS
- PS completes its processing of ALLOCATE
- PS calls nlu_tidy_up_plu_and_mode, which removes a
mode CB ***bug***
(note that the assert on mode state in nlu_remove_mode
would get hit here).
- PS returns the ALLOCATE verb (not traced)
- PS COPR detects the failure of the ALLOCATE
- PS COPR tries to unlock all modes but none exist.
Asserts on mode_state and mode locks seen in the dump
also support this explanation.
Resolution:
A check is made on the mode lock before freeing the mode.
The mode CB is freed by npo_reset_sess_limit_proc
instead of npb_allocate_proc.
(3) JAGac86152/8606128605
The router kept the LU 'offline' until +RSP to a NOTIFY was
received. While the LU was 'offline' no segmentation
checking was done and an incoming EBIU (from the end of the
VTAM Screen) was ignored. This kept us 'in segment' and
caused subsequent NOTIFY +RSP and a BBIU to be processed
with an error. The IBM side meanwhile considers the LU to
be online since it received no indication of a problem.
Resolution:
The fix is to ensure that even when we are offline we
still clear the in_biu flag when we see an EBIU flag.
(4) JAGac78841/8606128040
In certain race cases the mode control block was being
released to memory twice leading to the panic.
Resolution:
A major rework of the mode control block handling code was
performed to carefully check and remove mode control blocks
at the correct time. The mode lock is checked before
freeing the mode, so that npb_allocate_proc does not free
the mode CB, but npo_reset_sess_limit_proc does.
(5) JAGab74691/8606105945
The internal error (bracket race) was caused by
bad SNA data coming from the host (EB BC followed by
EC CD) and the user (or possibly HLLAPI) typing ahead.
The ASSERT that gets generated is actually harmless.
Resolution:
The bracket race is avoided by working around the bad SNA
data sent by the host (removed CD from EC). The ASSERT was
corrected by zeroing a floating length field.
(6) JAGab71415/8606104367
The psi1 driver used the svphpoll call to handle both
ACC and PSI cards.
Resolution:
Code added to the svphpoll function to check card type. If
the card type is PSI then execute the psisvphpoll function
else if type is ACC then exit.
PHNE_21001:
(1) JAGac78479/8606127677
An Internal message is being discarded during the close
processing of a link. The problem is intermittent due
to a race condition.
Resolution:
Do not discard the internal messages during close
processing.
(2) JAGab78721/8606108322
There is a hardcoded limit that means only 50 Local LUs
are properly supported by ASCII Admin.
The fact that the customer could get up to 250
dependent LU 6.2s defined was simply luck - we were
actually overwriting space allocated for remote node
definitions with the local LUs. In this case this was OK
as there were none. Once we started impinging on Mode
definitions, though, we dumped.
Resolution:
Increase the amount of space allocated for Local LUs from
50 to 300. The ideal fix would be to re-engineer ASCII
admin so that it didn't use static arrays for it's data
structures. However, in this case it seems more
efficient to simply increase the size of the array for
Local LUs.
PHNE_20734:
(1) JAGab84956/8606112613
A call to sigprocmask was added to the libmgrdce.1 library
( see SR4701424069 ). This fix was put in to stop signals
flowing up the path manager or HLLAPI threads. This was to
ensure that signals would only be processed by the user's
application not by SNAplus2 libraries. However, the
behaviour of sigprocmask under DCE threads is different to
that under kernel threads. With kernel threads the call
blocks only signals on the particular thread. Under DCE
threads, the call blocks signals for the entire process
resulting in the customer's signal catcher never being
called.
Resolution:
Modify the previous fix to take into account situations
where we are using DCE threads. On 10.20 this is straight
forward as we know at compile time whether a library is to
be used with DCE threads or not. On 11.0 use the
_CMATHREADS macro to make a run-time decision on whether we
are using DCE threads or not. If using DCE, don't do any
signal blocking, otherwise block signals as in the previous
fix.
(2) JAGab83814/8606111714
As part of the R6 development it was decided to ensure that
no text strings were hard coded into applications.
Instead, any text string that might appear to an end user
was put into a proper message catalog. Unfortunately, the
snapmsgf program is also used within the SNAplus2 build
system to build the main SNAplus2 log file. When the
change to put text messages into message catalogs was done,
it was incorrectly thought that snapmsgf was only used by
the build system. Consequently it's messages were put into
a private message catalog rather than the main catalog.
Hence, when snapmsgf is used on an end user system it is
trying to pick up a message catalog from the SNAplus2 build
system that doesn't exist.
Resolution:
The snapmsgf message catalog is added into the main sna.cat
file. Modified snapmsgf so that when used on an installed
system it picks it's messages up from the main message
catalog instead of the catalog in the build system.
(3) JAGab75474/8606106417
The permission change to 664 has been made to the device
files in the R6.11.00.200 release to allow customers a
tighter control for system auditing purposes as the sna
device files allowed unrestricted access. However, this
change cannot apply to /dev/sna_v5access file which
requires permissions of 666 to allow other unix users to
use the SNAPlus2 API's.
Resolution:
The snaplus2 script had a function which changed the device
file permissions to 666. This has been modified to set
the correct permissions at create time as follows:
create_devfile sixt /dev/sna_trace 0664
create_devfile sixr /dev/sna_v5router 0664
create_devfile clone /dev/sna_v5access 0666 0x6b
create_devfile clone /dev/sna_NDG 0664 0x6c
create_devfile clone /dev/sna_NMA 0664 0x76
create_devfile clone /dev/sna_QLLC 0664 0x75
create_devfile clone /dev/sna_SDLC 0664 0x6d
(4) JAGab65304/1653307843
The code formats a '--' in error when interpreting the
traced data. There also many minor problems with 0 and 1
being mixed up and bits not being right shifted correctly.
Resolution:
Correct code to format BIND correctly.
PHNE_19527:
(1) JAGab83920/8606111815
The SNAplus2 node carefully checks the ABM support fields
on an XID. Unlike previous versions of SNAplus, we are
now being very restrictive regarding the particular
combination of ABM support flags we let through. Whilst
we are following the SNA Formats manual correctly, this
does cause interoperability problems with some hosts.
In this circumstance we have decided to relax the
checks on the ABM support flags, to allow
interoperability with other hosts.
Resolution:
Remove test that compares ABM support flags on XID.
(2) JAGab83645/8606110894
Shared memory not correctly detached from library
following unload/load of 3270 program.
Resolution:
Correctly detach shared memory from library and then
connect to fresh shared memory.
(3) JAGab79003/8606108556
During nio_initialize, the driver code checks for NULL
IOVAs returned from sio_map() and will panic if the
returned an IOVA is NULL. However, a NULL IOVA is still
valid and no panic should occur.
Resolution:
Fix is to remove the panic on NULL IOVAs after sio_map()
calls. Also in the step data structure, invalid IOVAs are
redefined to be -1 (void* 0xffffffff) instead of 0. Also,
change all checks for 0 IOVAs to be -1 IOVAs.
(4) JAGab76276/8606106834
The ObAM definition for the SNASAM main menu window is
done in such a way that it prevents the window from
popping up immediately. The execute_me_first callback in
mainMenu.ui is utilized to construct all the SNA SAM
objects. This step can be done after SAM creates the main
menu window.
Resolution:
The solution is to change the definition of the SNA SAM
main menu dialog. The current execute_me_first callback
is changed to be the entry callback for the mainMenu
screen. A new execute_me_first callback is defined and
made into a stub.
The same solution is to be applied to 10.20 version
as well.
(5) JAGab75469/8606106415
When a SNRM retry is received very quickly after the
initial SNRM such that it crosses with the outgoing UA,
SNAP-LINK may crash because it does not have a valid
frame buffer.
Resolution:
Before sending a UA for a SNRM retry, check to see if the
frame_buffer used for UA's is present or not. If not, it
must be in the stub, so ignore the incoming SNRM retry.
(6) JAGab75335/8606106333
The problem is caused by the fact that SNA SAM is always
using the 'current' number of configured SDLC interfaces
for its display of the number of SDLC cards/ports
configured. This means it is tracking the activity of the
SDLC configuration screen. If the user moves a card from
the configured side to the unconfigured side, the number of
configured sdlc cards/ports decreases by 1. This is
incorrect. The current number of configured cards/ports is
only valid after the user committed the new configuration
(i.e., by pressing OK on the sdlc port config. screen).
Resolution:
The behavior of the callbacks for the SDLC configuration
dialog has been changed to track the number of SDLC
ports/cards correctly. The number was re-defined to be the
quantity of cards/ports previously configured and the
changes committed. Thus, when the sdlc configuration screen
is active (on display or icon-ized), the number of SDLC
cards/ports configured is the PREVIOUS number of SDLC
cards/ports configured successfully. The change was made
to both 10.20 and 11.00 versions.
(7) JAGab75114/8606106196
The cause of CR can be traced to the design of
SNA SAM. The code assumed that only D class
machines have model numbers with format of 8x1,
but the T500 has the 891 model number.
This caused the psi1 driver to be loaded into
the kernel by SNA SAM. The psi1 driver should
only be loaded into machines with EISA bus.
The T500 is an NIO bus machine and requires
the psi0 driver.
Resolution:
The function host::isUltralight() has been modified and its
dictionary of 800 numbers of D-class machines updated.
Only 8xx numbers of D-class machines should be used.
Furthermore, isUltralight() was to return true for all 8xx
machines using the EISA bus. Consequently, the host class
has to be extended to include a function isEISA800() that
will return TRUE if the host machine running SNASAM is EISA
based. This will ensure this problem will never occur
again.
For all SNASAM C++ classes remove all calls to
host::isUltralight( void ). In its place use the call
host::is800EISA( void ). The is800EISA(), as the name
indicates, returns TRUE for the the host machine is EISA
based. The determination of bus type is done using ioscan,
not uname (2).
The host class used is from R6.11.00.200 release of SNASAM
for HPUX 11. The inclusion of the new host class requires
some C++ function prototypes to change (in sdlc.cpp,
sdlcScreen.cpp, snaplus2Screen.cpp, and mainMenu.cpp). The
TMPipe class, sna.tm, and sna.da from R6.11.00.200 release
of SNASAM has to be included here as well.
(8) JAGab74185/8606105839
The problem is caused by a STATUS_SESSION(NO_SESSION,
LU_INACTIVE) followed by an CLOSE_PLU_SLU_SEC_RQ.
The CLOSE causes a dummy UNBIND to be built and
queued, but the CLOSE kills the session between
the RUI_BID returning and the subsequent RUI_READ.
Resolution:
Check the reason for the CLOSE - if it is due to a LINK
error or because the PU or LU are inactive, then simply
kill the session - don't send the UNBIND down to the
application.
(9) JAGab73055/8606105164
Problem is that we are accessing data shared between normal
context and call back context (hence typically interrupt
context) outside the correct locking. We need to keep the
lock on vhs_data_lock whenever we update data in the
pcb_shared_data area of the port control block. We had a
window where a TX was issued at the same time as a RX
completed and so both bits of code tried to set up the next
frame to RX and got confused.
Resolution:
The fix is to re-order the code to make all the changes to
pcb_shared_data area inside the locked section of code.
(10) JAGab71689/8606105222
From the code it appears that there is not really a
parameter mismatch, but it looks as though the problem is
due to us trying to log more text than we actually allow.
We are limited by a 600 byte array of log text in
nba_pd_print_var.
Resolution:
The log parameters are limited to 600 bytes to allow all of
the log datagram to fit in a 1k buffer for use by the SLIM.
Consequently, we can't change this 600 byte limit.
Instead, for the particular log mentioned in this defect,
we should ensure that the sense code gets put into the
buffer before the error data (that way we will ensure we
get all of the sense code). This will also stop the
erroneous messages in the log file as we will be logging
all of the parameters.
(11) JAGab71519/8606104163
In certain circumstances RM will be asked to delete an
SCB that was created by another instance of RM. If
this happens then we can assert or crash when
trying to delete the session.
Resolution:
Fix is to store the process ID of the RM instance that
creates the SCB on the SCB, then when RM receives a
DEACTIVATE_CONV_GROUP (which is routed from the NOF using
the LU name/alias field) it can check that the SCB was
created by this instance of RM and reject the signal
(with new secondary return code NAP_LUNAME_CGID_MISMATCH).
(12) JAGab70700/8606106276
A problem in the ASCII to EBCDIC translation tables used
for German meant that the symbol ')' which we write to
the screen when the host locks the 3270 keyboard was not
being translated correctly to ASCII. The incorrect
translation meant that we output a control code instead
of ')', and this caused the terminal to switch into the
file transfer state!
When using German ISO88591 the problem was not as
serious, but the bug still existed - the '()'
symbol is displayed as '(-' instead.
Resolution:
Correct the EBCDIC to ASCII translation code for German
codepage 273.
(13) JAGab68385/8606102608
Missing flag in the streams definition.
Resolution:
Add flag to streams_info section of all kernel drivers.
(14) JAGab65516/5003455253
The SIGTERM signaling problem is caused by an interaction
with the DCE threaded library which the HLLAPI library is
linked with. We need to provide a single-threaded R6
version of the HLLAPI library for HP-UX 10.20.
Resolution:
Currently we provide the following objects:
HP-UX 10.20
-----------
libhapi.0 R5 compatible, single threaded
libhapi.1 (libhapi.sl link) R6 multi-threaded (DCE threads)
HP-UX 11.0
----------
libhapi.0 R5 compatible, single threaded
libhapi.1 R6 10.20 compatible (DCEthreads)
libhapi.2 (libhapi.sl link) R6 multi-mode
pa20_64/libhapi.2 R6 multi-mode 64-bit.
Fix is to modify the build system etc so we build the
following:
HP-UX 10.20
-----------
libhapi.0 R5 compatible, single threaded
libhapi.1 (libhapi.sl link) R6 single threaded
libhapi_r.1 (libhapi_r.sl link) R6 multi-threaded
(DCE threads)
HP-UX 11.0
----------
libhapi.0 R5 compatible, single threaded
libhapi.1 R6 10.20 compatible, single threaded
libhapi_r.1 (libhapi_r.sl link) R6 10.20 compatible
(DCE threads)
libhapi.2 (libhapi.sl link) R6 multi-mode
pa20_64/libhapi.2 R6 multi-mode 64-bit.
(15) JAGab65399/4701429621
If CH (Conventional Half-session) receives a segmented
logon screen to an off-line LU, for example a dependent
LU6.2 LU, CH may send an invalid response or, more likely,
no response at all. This could potentially cause problems
when trying to use APPC over that LU.
Resolution:
The nch_sscp_receive now stores the RH from the BBIU
(!EBIU) in the SSCP section of local data. When the EBIU
arrives, the RH is copied in and it is turned round as a
negative RSP.
(16) JAGab65397/4701428094
Firmware problem:
It is possible for the state machine to remain in the
CLOSE_PEND state indefinitely because it expects only
timeout events when in this state. The firmware state
machine should declare an error and restart itself
when it remains in the CLOSE_PEND state too long.
Driver Issues:
From looking at the driver code and the traces it produced,
it is apparent the driver is not initiating any action with
the firmware if the link unexpectedly goes down (simulated
by disconnecting the cable between the 9000 and the modem).
The protocol between driver and firmware requires a message
exchange for the link to start up.
Resolution:
The fix is to make sure that the driver and firmware, once
they become 'unsynchronized', have a way to be
re-synchronized. This is done by:
A)
When the firmware hits an error condition which it does
not know how to handle, it will set a system error and
'jump' to the first line of the firmware code (i.e., first
line of main). This is done by timing out on inactivity in
the OPEN_PEND and CLOSE_PEND states of the firmware. The
declaration of system error will allow the firmware code
code to reset.
B)
To make sure that the driver will not be hung, the driver
code will start out with a credit of 2 when it initiates
data transfer with the card. This is to prevent situations
in which both the firmware and the driver are waiting for
each other to send a message. With the new credit
assignment during initialization, the driver will always
able to initiate action on the card. Since the code is
designed assuming only 1 outstanding message to the
firmware at a time, the driver has a credit check to make
sure the credit value is not greater than 2.
(17) JAGab65393/4701424812
Part of the problem here is that ASCII admin is setting a
number of fields to their default values without checking
to see whether it is dealing with an existing record or
creating a new one.
For these particular fields - namely, mu_credit,
link_deact_timer, use_default_tg_chars and address, we
should only put in the default if we are creating a new
record (i.e. if NofDefAct != NOF_DEF_MOD).
The problem with the options on sdlc ports (i.e. the NRZI
stuff) is that the port dialog is using the stub_spec_data
on the link_spec_data to determine whether it should
initialise the dialog with the NRZI or NRZ.
In fact it should be using the stub_spec_data on the
port_spec_data instead.
Resolution:
Modify dsdlc.c so that mu_credit is only defaulted for
new dlcs.
Modify dsls.c so that link_deact_timer and
use_default_tg_chars are only defaulted for new LSs.
Modify dsport.c so that address is only defaulted to zero
for new ports and to check that settings for 'opt1' field
are taken from the port_spec_data rather than
link_spec_data.
(18) JAGab65392/4701424069
The issue stems from the way signals are handled in a
multi-threaded process. The default signals behaviour in a
multi-threaded environment is that any thread of a process
can be interrupted by the OS to run a signal handler when
an externally-generated signal is received for that
process. In particular, this means that any threads created
for internal use by the SNAplus2 libraries (eg in the path
manager to wait for messages to appear at the stream head)
could be interrupted and used by the OS to handle a signal
sent to a multi-threaded user application. This could
potentially happen at a point where the application has
masked off its own threads from being interrupted by
signals. In such a case, the application might fail
because its signal processing routine was being run
when it did not want it to be.
Resolution:
The fix is to modify the multi-threaded path manager and
HLLAPI libraries so that the new threads they create
protect themselves against being interrupted by external
signals. To do this, we just need to make the appropriate
call to sigprocmask() at the start of the new thread of
execution.
(19) JAGab25510/5003467290
SNAplus2 does not negotiate down the maximum BTU size
from the largest frame size received in the RIF on the
TEST_RSP (route discovery frame) received from the
token ring driver.
Resolution:
Correct the code to process the RIF before sending the size
back to SNAP APPN.
PHNE_18000:
(1) JAGab71162
Cause of problem:
Having looked at the dump file it appears that the root
cause of this crash is down to a timing problem:
For switched links we send a REGISTER_STATION message
from SNAP LINK to the HMOD stub. The HMOD receives
this message, sends a response and issues an
svphopen call to the actual HMOD. Upon receiving the
REGISTER_STATION response SNAP LINK sends down a DIAL
PORT message. The HMOD doesn't do anything with
this DIAL PORT - it simply sends a response back
immediately. When the HMOD open routine comes back
(via the HMOD call back) the HMOD stub sends a POLL
PORT message which indicates that we have successfully
raised DSR.
SNAP LINK must get the DIAL PORT response before the POLL
PORT.
However, if the HMOD open returns prior to the DIAL PORT
response being sent, SNAP LINK will decide this is an
error and issue a RESET PORT message.
Normally we don't see this as the svphopen takes long
enough to come back that we have processed the DIAL
PORT. However, turn on tracing and the SNAP LINK/HMOD
processing takes longer and there is a chance that we
will get the svphopen back sooner than the DIAL PORT
response (note that the customer hit this timing
window with just SDLC level 2 trace - to reproduce
it we had to turn on internal trace in the SDLC
driver as well).
The reason the box is actually crashing is even more
convoluted:
The reset port message that we send down (which is
built on top of an alert held on the port control
block in SNAP LINK) doesn't have the correct
port_handle put onto it. Consequently when we get the
reset_port response we assume that the port control
block has been destroyed and we simply free off the
reset port message. Unfortunately this leaves us
in a bad state:
- Firstly the resetting flag on the pcb is wrong - it
is left as resetting == TRUE.
- Secondly, because the reset port is freed off, this
frees the alert stored on the pcb.
The first assert is occurring because SNAP LINK issues
reset port again to reset the port group. At this point
we spot that the resetting flag is wrong.
The second assert happens when we are doing a retry of
the LS - the same timing problem as before happens
meaning we try and send a reset port using the alert
held on the pcb. Unfortunately this has been set to NULL
(hence the assert) because of our failure to reconcile
the reset port response. Moreover, when we try and access
this memory in the sdl_reset_port routine we crash.
Resolution:
There are two fixes here:
Fix one - ensure that the correct handle is sent on the
reset_port message.
Fix two - ensure that for switched outgoing links we
don't open the HMOD until we receive the DIAL PORT message
For fix one we simply add a line to sdl_reset_port to
ensure that in the reset port case (like the close port)
we set the port_handle to be the pcb->pcb_handle.
for fix two we add a test into vhs_register_station so
that vhs_hmod_open is only called if the port_type
not switched outgoing.
We then add a call to vhs_hmod_open to vhs_dial_port.
(2) 5003466367
The problem occurs under load when nch_sscp_receive() has
two NOTIFY messages on its normal request pending queue.
The routine loops round trying to process everything on
the queue. The first NOTIFY is sent out OK. However this
sets the flag 'LOCAL.norm_flow_rqs_blocked = TRUE'. This
means that the second NOTIFY is then placed back on the
queue by nch_df_sscp_send() resulting in a continuous loop
trying to empty the normal request pending queue.
Resolution:
Modify the while() test in nch_sscp_receive() that
processes the normal request pending queue so that it also
checks LOCAL.norm_flow_rqs_blocked and drops out when this
is TRUE. The end result of this change is that the first
NOTIFY is sent as at the moment, and we then drop out of
the queue processing and any further NOTIFYs or other MUs
on the queue are then dealt with when the NOTIFY RSP
returns.
(3) 5003455659
When converting SNAplus config files to SNAplus2
snapmigrate sets react_timer_retry = 0 in Link Station
definitions. This means SNAplus2 does not attempt to
reactivate the LS if it fails. It should in fact be set
to 65535 to match the behaviour of R4 (under which failed
connections are always retried).
Resolution:
Set react_timer_retry to 65535 instead of 0.
(4) 5003424085
This problem has occurred because SNAplus2 is based on
different code from SNAplus and simply didn't include a
way to do this.
Resolution:
In the GUI, when the user clicks OK in the define session
dialog, we will check to see whether the LU being assigned
is a single LU or a pool. If a single LU we will check to
see whether the Session record is new, or is a modification
of an existing record. If a modification we will see
whether the LU name is being changed. The rationale here
is that we don't want to pop up a warning when someone
modifies the description of an LU - we assume they have
already seen a warning.
Assuming a new session or the LU has changed, we will do a
search through all the sessions defined for all users
looking for a session with the same LU. If we find one then
we will throw up a confirm dialog. If we don't find one we
just carry on processing.
The fix is in routine vga_issue_def_3270_session. We also
add the routine vga_define_3270_confirm_reuse_cb.
We add a new message to the GUI message catalog
(vgacat.pre) -- MSG_CONFIRM_REUSE.
(5) 5003392456
Currently no functionality to allow viewing of LU6.2
session usage.
Resolution:
This enhancement adds support for a 'query_conversation'
verb to the command line administration program.
The query_conversation command returns information about
conversations using a particular local LU. This command can
be used to obtain summary or detailed information about a
specific session or a range of sessions, depending on the
options used.
(6) 4701430017
When configuring SDLC, SAM installed sixw driver on all
machines including those with no PCI bus. When the sixw
was removed, SAM considered the SDLC configuration to be
incomplete and removed entries in the sna.ini config
file. The sixw driver should only be installed on
machines with a PCI bus.
Resolution:
Change the installation of SNA drivers for SDLC
configuration in sdlc::getPSI_DriversForSDLC(). On
machines with:
NIO bus: Install sixl and psi0
EISA bus: Install sixl and psi1
PCI bus: Install sixw, psi1, and sixl.
PCI and EISA buses: Install sixw, psi1, and sixl.
On a T600 machine that has both PCI and NIO buses
install sixl, psi0, and sixw.
(7) 4701427021
Latest version of some binaries is required.
Resolution:
Updated binaries provided to permit a multi-version patch
to be created for the R6.11.00 release of SNAplus2, as
documented in SR text.
(8) 4701425561
The Streams/UX subsystem on hp-ux 11.0 , unlike SVR4
streams, does not provide any form of locking when
accessing a streams Q. Thus, on HP-UX it is not safe to
perform a PUTQ to a stream from outside its context
(i.e. from the put or service routine of another queue).
Resolution:
The streams call PUT() does contend for ownership of a
given queue, because HP-UX guarantees that only a single
put or service routine for a queue will be run at one time.
Thus, to ensure the streams queues are protected we modify
the SNA code to:-
- issue put() rather than putq()
- have the put routine for the streams Q issue
the putq() to defer processing to the
service routine.
(9) 4701425355
Problem is that ACC stack is calling QLLC put routine from
interrupt context. QLLC module is not designed to cope with
this: all other drivers/stacks queue their messages in a
simple service routine so they can be sent upstream outside
interrupt context. We have nevertheless agreed that we
will add queuing to our QLLC module so that it works with
the new ACC X.25 driver.
Resolution:
Fundamental fix is to move from the put() routine to the
service() routine all the read-side processing in the QLLC
module. In practice this only affects M_PROTO messages.
Examination of the QLLC module code suggested that code
processing these messages in put() routine could simply be
removed -- provided the messages were then queued to the
service routine -- because the service routine already has
to handle them (via an FSM) in situations of buffer
shortage. Empirical testing bore this out.
So fix actually simplifies code by removing 'special case'
processing for data messages when there are buffers
available and no control messages queued in front of
them.
(10) 4701424069
The issue stems from the way signals are handled in a
multi-threaded process. The default signals behaviour in a
multi-threaded environment is that any thread of a process
can be interrupted by the OS to run a signal handler when
an externally-generated signal is received for that
process. In particular, this means that any threads created
for internal use by the SNAplus2 libraries (eg in the path
manager to wait for messages to appear at the stream head)
could be interrupted and used by the OS to handle a signal
sent to a multi-threaded user application. This could
potentially happen at a point where the application has
masked off its own threads from being interrupted by
signals. In such a case, the application might fail
because its signal processing routine was being run
when it did not want it to be.
Resolution:
The fix is to modify the multi-threaded path manager and
HLLAPI libraries so that the new threads they create
protect themselves against being interrupted by external
signals. To do this, we just need to make the appropriate
call to sigprocmask() at the start of the new thread of
execution.
(11) 4701421081
The root cause of this Java hang is believed to be a
bug in the HP JVM ,however a workaround has been found
by implementing a code change in the SNAPlus2 library.
Note that since the workaround within SNAPlus2 requires
changes to the customer's code, other customers may also
hit the same issue and so we should still pursue a fix
for the underlying JVM problem with the Java team.
Resolution:
This fix is a workaround to the problem. It forces the path
manager not to rely on the stream being non-blocking. In
particular, vpm_sleep() should check the wait condition
after every call to vpm_rcv_from_stream(). This can be done
by using the vpm_io_limit to limit the number of calls to
vpm_rcv_from_stream which we make before checking the wait
condition.
This requires the application to call SNA_SET_IO_LIMIT(1);
before it makes any call to the API libraries.
(12) 4701418707
The outgoing attach was being sent with the password
from the previously rejected incoming attach causing a
validation error.
Resolution:
Add code to copy the password from the START_TP signal into
the tcp_ptr in nrmsttp.c
(13) 1653307231
The problem occurs under load, when a temporary shortage of
internal buffers means that the software must queue an
incoming message (from the host) and deal with it later,
when a buffer has become available. The reason for the
problem is that the calculation of the RU length in the
APPN node counts the same MU twice, once when it arrives
(before it gets queued because there are no BUFFERs
available) and again once it has been dequeued (because a
BUFFER is now available). Specifically,
ntc_buffers_available() calls ntc_process_btu(), passing
the dequeued buffer. This calls ntc_segment_transfer()
which increments the partial_bui_size field. The same code
(from ntc_process_btu onwards) is called when the MU first
arrived so the size of the MU is counted twice!
Resolution:
The fix is to 'undo' the first addition of the incoming
message length to our running total of RU size if it turns
out that we have to queue the message for later processing.
Specifically, the fix decrements tc_cb->partial_biu_size in
ntc_segment_transfer() if posting is requested and this was
a non-BBIU segment. The process will be called again and
the partial_biu_size re-incremented when the buffer that
was posted for returns.
(14) 1653305805
rom the stack we can see that this a deadlock in the
kernel during snap stop processing. We grab a write lock
on vpr_entity_lock in vpr_stream_close() which we hold
across a number of calls, including the one to nba_term().
It is this lock we are trying to acquire in
vpr_route_ips_on_route() near the top of the stack trace.
Resolution:
We don't actually need to hold the vpr_entity_lock round
the call to nba_term() in vpr_stream_close(). So the fix
is just to release it before that call and reacquire it
afterwards.
(15) 1653305284
The problem is caused by the entry 'sixt gio' in the
DRIVER_DEPENDENCY section of the SNAplus2 master file.
The SAM patches PHCO_17631 and PHCO_18217 enhanced SAM's
checking of driver dependency information. Sixt has no
dependency on gio, gio also being an invalid driver for
11.X. So when SAM checked the driver dependency for SNA,
it found this error and could no longer continue.
Resolution
Remove the entry 'sixt gio' from the SNAplus2 master file.
After the master file is replaced, run
/usr/sam/lbin/getkinfo -b
to have the changes take effect.
(16) 1653301382
SNAplus2 is getting an error with errno set to ERANGE on a
STREAMS call in the path manager. It turned out this error
is because the customer's machine is configured to have
the maximum STREAMS message size (STRMSGSZ) of 8K
configured in the kernel. The solution is to remove this
limit (or set it to at least 64K).
Resolution:
The resolution for the customer is to increase the STRMSGSZ
kernel parameter.However the SNAPlus error log was
incorrectly recorded. The correct log message was already
included in the code,but did not reach the sna.err file
because it was being made after the call to
vpm_error_event(). The solution is simply to move it from
the subsequent line to the preceding one.
This ensures that the error log does get written out
correctly as follows.-
SNAPATH Message 16388 - 13, Subcode: 1 - 1
Log category: EXCEPTION Cause Type: External
System: borg
Process ID: 1155 (snapcfgdae)
I/O Error writing message to Stream.
Errno = 34 (Result too large)
Cause: The Stream to the SNA Subsystem has reported a write
error.
Action: Check the operating system error code shown, and
retry the operation that led to this message. If the
problem persists, provide support services with the audit
and error log files, and trace files if available.
(17) 1653301333
This problem arises because the V0T_TRACE_HEADER_MAX_LEN in
v0/v0trc.h was not updated when the enhancement to add time
zone information to trace files was made and two extra
variables were put in the trace file header. This meant the
amount of space allowed in which to build the trace file
header was insufficient for all possible headers. On most
machines we weren't up against the limit -- which was why
this defect didn't show up when the time zone enhancement
was tested. However, with the recent switch to daylight
saving time, the time zone abbreviation string in much of
Europe changed from 'MET' to 'METDST'. On some machines
this made the header too big for the buffer in which it
was being built.
Resolution:
Changed the code to make the buffer much bigger than
required -- rather than just big enough for the theoretical
maximum header. Also enforce limits on the individual
fields in the header (because eg anybody can change the
time zone abbreviation string just by setting the TZ
environment variable) and truncate if they are too long
-- instead of running the risk of being unable to write
the header and therefore being left with unformattable
trace files.
It also turns out that time zone abbreviations are
truncated to 5 chars when printed in snaptrcfmt output.
This is just an incorrect formatting assumption in a
printf statement in snaptrcfmt, which has also been fixed.
(18) 1653299685
The problem stems from the fact that the 'Send node name in
XID3' checkbox (on the 'Advanced' submenu of the 'Modify LS
configuration' dialogue in ASCII Admin) corresponds to two
separate configuration settings in the sna_node.cfg file.
On the equivalent menu in Motif Admin there are two
separate buttons for these two settings, so the problem
does not arise. If the settings are not both on or both
off then ASCII Admin incorrectly changes the first to match
the second.
Resolution:
It has been agreed that we will not make further
enhancements to ASCII Admin, so we do not propose to add
an extra check box to this Menu. However, given the actual
wording of the current check box, we believe it makes sense
to modify its action to stop it changing the adj_cp_type
field (ie whether or not we check the CP name on incoming
XID3 messages). This simply involves a change to
set_nof_independant_ls() [sic!] in vaa/dsls.c.
(19) 1653299073
The problem is caused by the corruption of the lan3 data
structures involved in DMA transactions by the psi0 DMA
transaction processing.
Resolution:
Changed the handling of DMA transactions. The transactions
are still queued but the DMA engine processes only one
transaction at a time. It does not prefetch DMA
transaction because we force it to stop and generate an
interrupt after having processed a transaction. When the
driver gets the interrupt related to the completion of
a DMA transaction, it starts processing the next DMA
transaction in the queue.
(20) 1653293878
The TP is failing to start because the userid under which
it is running has been misconfigured so that it can't
retrieve its own group name. This may be due to local
access to the group file or with running NIS (Network
Information Service) to share user and group IDs across
more than one machine. There are two reasons for the
cryptic error logs recorded by SNAplus2:-
- Failure of the getpwuid() or getgrgid() system calls
was not logged as an error message.
- The VSM_AS_TP_FAILURE internal error code was not getting
put in the right part of the DLOAD_RSP_ERR message sent
from the Service Manager to the APPC Stub. This meant that
the APPC stub was misinterpreting it as an APPC sense code.
Resolution:
The root cause of the problem is to correctly configure the
Unix user/group under which the TP is to be run.
However changes to SNAplus2 have been made to improve the
logging in this area as follows:-
In vpm_build_user_info() in vr/vpmu.c we add error logs for
the cases where getpwuid() or getgrgid() system calls fail.
However, failure of these system calls leads to a path
failure. So to make sure these new error logs actually
reach the sna.err log file, we also modify
vlm_user_write_log() in vdiag/vlmuser.c so that even if we
fail to open a path we still attempt to send the datagram
containing the log (in addition to attempting to write it
locally).
In the error reply arm of vsm_rcv_dload_confirm() in
vr/vsmdload.c we put the error code in the dld_status field
rather than the ld_sense_data field of the DLOAD_RSP_ERR
message -- because this is where the vas_datagrams()
routine in the APPC Stub expects to find it. We also change
the exception logged in vsm_rcv_dload_confirm() from the
generic one, with its rather misleading reference to errno
to a new specific error.
Texts of the new logs are in the vdiag/*.txt files.
(21) 1653245936
This problem occurred because SNAplus2 is based on
different code from SNAplus and simply didn't include
this feature.
Resolution:
The fix we have put in place here is to add some additional
formatting to snaptrcfmt that decodes RECEIVE_FRAME and
TRANSMIT_FRAME from the IPS trace in a more usable format
(decoding out the address and control part of the
message including Poll/Final and Nr/Ns bits).
A README file is going with the enhancement, but in
summary:
SDLC frame trace can be switched on using the following
snapadmin command or from xsnapadmin:
snapadmin set_trace_type, trace_flags=SDLC
Formatting of an SDLC frame trace is done by default with
the snaptrcfmt utility.
PHNE_17229:
(1) 5003450536
This snapmigrate problem is that we assume partner LUs
have the same Network IDs as the CP of the adjacent node
on which they are defined -- which isn't necessarily
true.
In delivering this fix, we have also uncovered a problem
with the psf file. The snapmigrate binary _should_ be
delivered as '/opt/sna/bin/snapmigrate2' and called from
a wrapper script, '/opt/sna/bin/snapmigrate'. However, an
error in the psf file means that, on 11.0, we have
delivered the binary file as both
'/opt/sna/bin/snapmigrate' and '/opt/sna/bin/snapmigrate2'.
Resolution:
Fix to the original problem is to replace the
DEFINE_ADJACENT_LEN_NODE verb(s) issued
in do_len_directory() in vmigrate/miggtfld.c
with DEFINE_DIRECTORY_ENTRY ones. Because the
assumption about the Network IDs is contained within
the logic of the DEFINE_ADJACENT_LEN_NODE verb
(which actually issues DIRECTORY_ENTRY verbs
internally).
Also fixed the duplicate checking in this function.
Originally only checked for duplicate aliases.
Added checking for duplicate fully qualified names.
In doing this have had to change the order of the
duplicate check.
Previously the entry being processed was checked
against the remaining unprocessed ones in the
list of PLU names, and duplicates marked so they
would not be processed.
However, fully qualified name is not easily
available until an entry has been processed. So now,
entry about to be processed is checked against ones
that have already been processed (in which we have
recorded the fully qualified name). This has the side
effect that, if an entry is multiply duplicated
(ie occurs three or more times) then multiple warning
messages will be printed for it instead of just one.
for it instead of just one. The functionality of the
migration program is not however impaired by this.
The other problem found when processing the
customer's config files lies in mig_output_appc_lus(),
again in vmigrate/miggtfld.c.
The problem is very similar to the second part of
the one above. We are already checking for duplicate
aliases. Just need to add check for duplicate
fully qualified PLU names.
Also the /opt/sna/bin/snapmigrate script is supplied
with this fix.
(2) 5003446971
Panic caused by attempting to dereference null pointer
while examining posted_list LQE in nbm_info structure to
see whether it is empty.
Resolution:
Add boolean flag to nbm_info to say whether posted list is
empty or not.
(3) 5003446278
This problem has not been reproduced. However, a system
chown call is used to perform the change of ownership of
the received files, and it is suspected that this call has
failed.
Resolution:
If this OS call really is returning an error in certain
circumstances, we need to alert the user so that the root
cause of the problem can be discovered and remedied. So
code has been added to check return codes from chmod &
chown and log an error if either fail. Also, just to be
doubly sure that the file attributes really have been
modified correctly, we then use the OS stat function to
query them and compare with what they are supposed to be
set to.
(4) 5003442426
The domain config file can be corrupted if client server
environment and file size > 64K.
Resolution:
Prevent configuration corruption by correcting the code
which processes sending domain config file down paths
via the Path Manager - snapcfgdae code changed to
ensure that when processing 3270 user records we always
setup control variables to ensure that if our buffer
fills up we restart processing exactly where we left
off.
(5) 5003441717
If the kernel initialisation fails, it is possible that
the snaperrlog process could hang - waiting for a signal
from the kernel which never arrives.
Resolution:
A code change has been made to ensure that ,if the
kernel initialisation fails, a failure notification is
sent to the snaperrlog process so it can exit
cleanly.
(6) 5003434860
Various xsnapadmin problems.
Resolution:
Code change made to resolve problems with
xsnapadmin configuration/administration when
used from client system.
(7) 5003433862
snapmigrate incorrectly converts and SDLC leased
connection from SNAplus to SNAplus2.
Resolution:
Set up the 'address' field on the SDLC port to
be 0x00, to allow all poll addresses through - the
poll address configured on the LS being used will
then be used to check, rather then the one on the
port.
(8) 5003427518
The real cause of the ASSERTs being generated is
the config daemon detecting duplicate entries in the
sna_tps file.
As the config daemon reads the sna_tps file, it sets a
flag to say that an update datagram needs to be sent
to inform the router of the TP, and records the TP as
being defined.
When the daemon comes across the second identical
entry for the TP it decides that as the TP is already
known about, the update datagram flag should not be
set, so unsets it. However, the config daemon still
thinks the TP has been defined.
So the datagram does not get sent out, and the TP
is never defined to the router, but the config daemon
has recorded it as having been defined.
The ASSERT is issued later when the config daemon
issues a datagram to the router telling it to delete
the TP definition.
The router does not have the TP definition, and
raises the ASSERT to say that the action code in a
datagram for an undefined TP must be 'add a new TP'.
Resolution:
A change has been made to the code so that when a TP
definition is read from the sna_tps file, and this is
found to be a duplicate, the status flag is checked to
make sure it is retained, rather than overwriting it.
(9) 4701413054
Small timing window when there is an empty list of LULU
control blocks when processing SSCP_INIT_SIGNAL_NEG_RSP
ISP.
Resolution:
Code changed to check whether LULU list is empty before
trying to obtain first element of it.
(10) 4701407528
Possible memory leaks may occur.
Resolution:
Code change made in PCI SDLC Glue module to
prevent possible memory leak. This change was to use the
streams function copymsg rather than dupmsg when passing
SDLC frames to the WAN
module.
(11) 4701400929
snapmigrate cannot correctly convert qllc configuration
from SNAplus as the port information is missing.
Resolution:
snapmigrate code changed to print a warning when
a QLLC port or LS is migrated telling the user to
manually check configuration.
(12) 4701399527
Assert errors occur because a large SSCP message is
received from the host for an LU6.2 session. This
errors as these messages are not expected for an LU
type 6.2.
Resolution:
A Code change has been made to prevent Assert
errors occurring when a large USSMSG10 is received
for an LU6.2 session.
The maximum amount of data permissible on the SSCP
screen has been increased to 2048 bytes, to ensure
segmented data on SSCP screen handled correctly.
(13) 4701399279
The PSI f/w header string was not changed with the
release of SNAplus2 as the f/w is common to both
SNAplus & SNAplus2.
Resolution:
- a new what string for the NIO firmware
- a new what string and a new compilation format for
the EISA firmware
The ']' character has been added at the beginning of
each PSI firmware library header line so that the
header can be recognized by the snapwhat command.
(14) 4701392670
Various assert errors logged due to the LAN logger
component (which handles central logging) incorrectly
registered itself with the service manager as a server.
This means that a server could end up twice in
the service table (for example, once as a backup, then
again as a master server). This leads to extremely
unpredictable and unreliable client/server operation.
Resolution:
Code change made to prevent incorrect registering of the
LAN logger component as a
server.
(15) 1653289686
TN3270 client was locking up when the clear key was
entered because TN Server was passing the clear
command to the Host instead of processing it locally
(as is done in the Motif 3270 emulator for example).
Resolution:
Code changed to add check and special handling for the
clear key at the beginning of the TN Server SSCP inbound
MU processing.
(16) 1653289603
TN3270 client was receiving SSCP datas when the clear
key was entered because TN Server was passing the clear
command to the Host instead of processing it locally (as
is done in the Motif 3270 emulator for example).
Resolution:
Code changed to add check and special handling for the
clear key at the beginning of the TN Server SSCP inbound
MU processing.
(17) 1653285478
Here is the sequence of events that leads to the error :
1. Error case which sets the HDWR_DEAD bit is set in
the PDA flags field (pda->flags)
2. Stop SNA
3. Start SNA
4. first ioctl in libsnainit.1 fails because the
HDWR_DEAD bit is set ...
Resolution:
The HDWR_DEAD flag has to be reset each time the
firmware is downloaded on the card. Thus, the fix
consists in calling the CLEAR_PDA_FLAGS ioctl just
before downloading the firmware on the card.
It is implemented in the libsnainit.1 library.
(18) 1653280123
snap2adm encountered configuration problems with
ports and LS's due to incorrect OBAM ui description
files.
Resolution:
OBAM ui description files corrected.
(19) 1653279703
Assert errors logged when starting an RJE workstation
and incorrectly listening for RTM requests.
Resolution:
Code change made to fix a problem with Assert errors
being logged when an RJE workstation is started.
The ASSERT should only be produced if an application
has opened the SSCP session and is listening for RTM
requests (RJE does not do this so it should not be
logged as an error).
(20) 1653276543
3270 session hang, due to NOTIFY not being sent
because the CH queue is not being flushed.
Resolution:
Code change to prevent 3270 session hang, due to NOTIFY
The fix applied is to ensure that any pending NOTIFY
requests are flushed from the CH queue in the APPN
node when a CLOSE_SSCP message is received (indicating
that the emulator has been stopped).
(21) 1653273979
Snaplus2 has a restriction of a single PU on an SDLC
leased line.
Resolution:
Enhancement made to allow multiple PUs to be used on
a secondary leased link. This means that if an SDLC
port is connected to a leased line, you can have
multiple LS's active over the port at the same
time.
(22) 1653267179
If you specify AP_SAME on the ALLOCATE verb but did not
configure user validation, then we will send a user ID
consisting of 10 NULLs.
Resolution:
A code change has been made, and the following
behavior applies when AP_SAME is used:
case 1: A TP on Unix invokes a remote TP
The outgoing Allocate will contain a
userID subfield, set to the Unix user ID
the TP is running under.
case 2: A TP on Unix invokes several remote TPs
See case 1.
case 3: Multiple conversations, where an INVOKED TP
issues an ALLOCATE
In this case, the outgoing Allocate will include
the same level of validation which was on the ATTACH
that invoked that TP.
(23) 1653264531
The API failure after fork is because the Path Manager
is initialised by the first API call.However, when you
fork(), the path manager needs to do some post fork
handling - this includes notifying all APIs that have
open paths.
So, the bug was that the path manager notifies the APIs
which have been initialised prior to the fork - but not
any that are initialised after the fork is issued.
Resolution:
A code change has been made to ensure that the path
manager correctly notifies the API's initialised via
a fork call.
Enhancement:
No (superseded patches contained enhancements)
PHNE_18000:
This has enhancements to add support for a
"query_conversation" verb to the command line
administration program and to enable tracing and
formatting of SDLC frames. The enhancements are
documented in SR 5003392456 and 1653245936 and in
the /opt/sna/samples/Common_patch.notes file on the
patched system.
SR:
8606300552 8606291438 8606289627 8606284767 8606268352
8606267173 8606266999 8606242140 8606227887 8606221968
8606218168 8606208520 8606207227 8606202416 8606202100
8606199884 8606195995 8606184850 8606179959 8606178646
8606177656 8606173811 8606173243 8606172744 8606171744
8606171082 8606166322 8606165404 8606164859 8606155977
8606151854 8606144657 8606143222 8606134781 8606132121
8606130719 8606128605 8606128040 8606127677 8606112613
8606111815 8606111714 8606110894 8606108556 8606108322
8606106834 8606106417 8606106415 8606106333 8606106291
8606106276 8606106196 8606105945 8606105839 8606105222
8606105164 8606104367 8606104163 8606102608 8606101599
5003467290 5003466367 5003455659 5003455253 5003450536
5003446971 5003446278 5003442426 5003441717 5003434860
5003433862 5003427518 5003424085 5003392456 4701430017
4701429621 4701428094 4701427021 4701425561 4701425355
4701424812 4701424069 4701421081 4701418707 4701413054
4701407528 4701400929 4701399527 4701399279 4701392670
1653307843 1653307231 1653305805 1653305284 1653301382
1653301333 1653299685 1653299073 1653293878 1653289686
1653289603 1653285478 1653280123 1653279703 1653276543
1653273979 1653267179 1653264531 1653245936
Patch Files:
SNAplus2-Link.SNAP2-LINK,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32/64,v=HP:
SNAplus2-Link.SNAP2-LINK,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32/64,v=HP:
/opt/sna/sdlc.dlf
/opt/sna/sdlc.pbs
SNAplus2-Link.LINK-64ALIB,fr=R6.11.00.200,
fa=HP-UX_B.11.00_64,v=HP:
SNAplus2-Link.LINK-64ALIB,fr=R6.11.00.010,
fa=HP-UX_B.11.00_64,v=HP:
/opt/sna/conf/lib/libpsi0.a
/opt/sna/conf/lib/libpsi1.a
/opt/sna/conf/lib/libsixd.a
/opt/sna/conf/lib/libsixl.a
/opt/sna/conf/lib/libsixm.a
/opt/sna/conf/lib/libsixp.a
/opt/sna/conf/lib/libsixq.a
/opt/sna/conf/lib/libsixs.a
/opt/sna/conf/lib/libsixw.a
/opt/sna/bin/snaptnsrvr
SNAplus2-Link.LINK-32ALIB,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32,v=HP:
SNAplus2-Link.LINK-32ALIB,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32,v=HP:
/opt/sna/conf/lib/libpsi0.a
/opt/sna/conf/lib/libpsi1.a
/opt/sna/conf/lib/libsixd.a
/opt/sna/conf/lib/libsixl.a
/opt/sna/conf/lib/libsixm.a
/opt/sna/conf/lib/libsixp.a
/opt/sna/conf/lib/libsixq.a
/opt/sna/conf/lib/libsixs.a
/opt/sna/conf/lib/libsixw.a
/opt/sna/bin/snaptnsrvr
SNAplus2-Common.SNAP2-SAM,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32/64,v=HP:
SNAplus2-Common.SNAP2-SAM,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32/64,v=HP:
/opt/sna/sam/SNA_SAM.sl
/opt/sna/sam/C/mainMenu.ui
/opt/sna/sam/sna.da
/opt/sna/sam/sna.tm
SNAplus2-Common.SNAP2-NOTES,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32/64,v=HP:
SNAplus2-Common.SNAP2-NOTES,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32/64,v=HP:
/opt/sna/samples/Common_patch.notes
SNAplus2-Common.SNAP2-CORE,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32/64,v=HP:
SNAplus2-Common.SNAP2-CORE,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32/64,v=HP:
/opt/sna/lib/libpe.1
/opt/sna/lib/libsnainit.1
/opt/sna/lib/nls/msg/sna.cat
/opt/sna/conf/master.d/sna2
/opt/sna/lib/nls/msg/sna_log.msg
/opt/sna/bin/snaperrlog
/opt/sna/bin/snaptpsrvd
/opt/sna/init.d/snaplus2
/opt/sna/bin/snapmsgf
SNAplus2-Common.SNAP2-ASCIIADM,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32/64,v=HP:
SNAplus2-Common.SNAP2-ASCIIADM,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32/64,v=HP:
/opt/sna/bin/snap2adm
/opt/sna/bin/snap2adm.ui
/opt/sna/bin/snap2adm_v.ui
SNAplus2-Common.SNAP2-ADMIN,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32/64,v=HP:
SNAplus2-Common.SNAP2-ADMIN,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32/64,v=HP:
/usr/newconfig/etc/opt/sna/app-defaults/XSnapadmin
/opt/sna/bin/snapadmin
/opt/sna/bin/snapmigrate
/opt/sna/bin/snapmigrate2
/opt/sna/lib/X11/uid/XSnapadmin/xsnadlg.uid
/opt/sna/bin/X11/xsnapadmin
SNAplus2-Common.CORE-64SLIB,fr=R6.11.00.200,
fa=HP-UX_B.11.00_64,v=HP:
SNAplus2-Common.CORE-64SLIB,fr=R6.11.00.010,
fa=HP-UX_B.11.00_64,v=HP:
/opt/sna/lib/libmgrdce.1
/opt/sna/lib/pa20_64/libpe.1
/opt/sna/conf/lib/libsixc.a
/opt/sna/bin/snapcfgdae
/opt/sna/bin/snapslim
/opt/sna/bin/snaptrcfmt
/opt/sna/bin/snapfilter
SNAplus2-Common.CORE-32SLIB,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32,v=HP:
SNAplus2-Common.CORE-32SLIB,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32,v=HP:
/opt/sna/lib/libmgrdce.1
/opt/sna/conf/lib/libsixc.a
/opt/sna/bin/snapcfgdae
/opt/sna/bin/snapslim
/opt/sna/bin/snaptrcfmt
3270-Common.MANAGERS,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,
v=HP:
3270-Common.MANAGERS,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,
v=HP:
/opt/sna/include/values_c.h
/opt/sna/lib/pa20_64/libsna.2
/opt/sna/lib/pa20_64/libcsv.1
3270-Common.MANAGER-64SLIB,fr=R6.11.00.200,
fa=HP-UX_B.11.00_64,v=HP:
3270-Common.MANAGER-64SLIB,fr=R6.11.00.010,
fa=HP-UX_B.11.00_64,v=HP:
/opt/sna/lib/libmgr.1
/opt/sna/lib/libcsv.1
/opt/sna/lib/libsna.2
/opt/sna/lib/libsna.1
3270-Common.MANAGER-32SLIB,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32,v=HP:
3270-Common.MANAGER-32SLIB,fr=R6.11.00.010,
fa=HP-UX_B.11.00_32,v=HP:
/opt/sna/lib/libmgr.1
/opt/sna/lib/libcsv.1
/opt/sna/lib/libsna.2
/opt/sna/lib/libsna.1
3270-Common.3270API,fr=R6.11.00.200,fa=HP-UX_B.11.00_32/64,
v=HP:
3270-Common.3270API,fr=R6.11.00.010,fa=HP-UX_B.11.00_32/64,
v=HP:
/opt/sna/lib/libhapi.0
/opt/sna/lib/libhapi.1
/opt/sna/lib/libhapi.2
/opt/sna/lib/libhapi_r.1
/opt/sna/bin/snapprtsty
/opt/sna/lib/pa20_64/libhapi.2
/opt/sna/lib/HLLAPI_patch.notes
what(1) Output:
SNAplus2-Link.SNAP2-LINK,fr=R6.11.00.200,
fa=HP-UX_B.11.00_32/64,v=HP:
/opt/sna/sdlc.dlf:
] SNAplus2 EISA FW v2.9
] (00/12/15 02:30:07)
]
/opt/sna/sdlc.pbs:
]SNAplus2 NIO FW v2.1
](98/11/13 11:58:22)
SNAplus2-Link.LINK-64ALIB,fr=R6.11.00.200,
fa=HP-UX_B.11.00_64,v=HP:
/opt/sna/conf/lib/libpsi0.a:
]R6.11.00.204 SNAplus2 R6 NIO PSI driver (64 bit)
] (PHNE_19527: 99/11/23 13:15:53)
]
/opt/sna/conf/lib/libpsi1.a:
]R6.11.00.200 SNAplus2 R6 EISA PSI driver
] (11.00.R6 DART45: 99/04/01 16:18:30)
]
/opt/sna/conf/lib/libsixd.a:
]R6.11.00.205 SNAplus2 R6 NDLC to DLPI Mapping (64 b
it)
] (PHNE_24341 : 01/07/25 12:39:02)
]
/opt/sna/conf/lib/libsixl.a:
&