IBM Tivoli Network Manager IP 3.9.0 Fix Pack 4, 3.9.0-TIV-ITNMIP-FP0004

Download Description

***************************************************************************
A change introduced by 3.9.0-TIV-ITNMIP-FP0004 might negatively
affect existing product function.

Please refer to APAR(s) IV60422, IV60742, IV61404, IV61701 and IV62920 for a
description of the problem(s) and corrective action(s).

APAR:IV61404 APAR:IV60742 APAR:IV60422 APAR:IV61701 APAR:IV62920

Evaluate these APARs for the potential impact in your environment.
****************************************************************************
Summary of the Enhancements included in this Fix Pack.

Poller Improvements - Numerous improvements in the Poller related to start up performance and for monitoring Poller load. Poller metrics are introduced that can be output as bar charts on the command-line interface.
OSDB Support

• - SUSE 11 SP2 / SP3
• - DB2 HADR (supported on DB2 9.7, 10.1)
• - SUSE KVM

TIP, Browser & JRE support

• - TIP version 2.2.0.13
• - FireFox24 ESR, Internet Explorer 10
• - Client JRE : Oracle JRE 1.7

GUI performance improvements - Incorporates GUI Performance improvements added in ITNM 3.9 FixPack3 Interim Fix 0001.
Enhanced Domain-Agnostic Entity Structure Search - Use entity smart search to search for an entity across all Network Manager domains or across a custom NCIM database table in an extended NCIM database.
Multi-threaded XML-RPC Helper- Pre FP4: XML-RPC Helper limited to a single request at a time. FP4: We have massively reduced the risk of this bottle neck by making the XML-RPC helper support multiple concurrent requests. Improvement:

• - discover increasing numbers of Collectors (EMS)
• - discover increasing volumes of data from those Collectors (EMS)
• - complete discovery in a reasonable amount of time

Several enhancement to ITNM Agents, including :

• - RFE 20180 Discovery support for ZTE M6000, T1200 and T600 Physical Inventory
• - RFE 28742 VLAN Discovery support for ExtremeSwitch version 12
• - RFE 16432 HTTPS support for Perl Collector Framework and Alcatel5620SamSoap

Cross Domain for Visualization.

Full detail for above Enhancements, download attached 3.9.0_INSTALL_FP0004 file.
============================================================================
Known problems with V3.9.0 Fix Pack 4
============================================================================
Known problems are documented in the form of individual Technotes in the Software Support knowledge base.

List of known problem in 3.9 Fix Pack 4.
https://www.ibm.com/support/knowledgecenter/SSSHRK_3.9.0/com.ibm.networkmanagerip.doc_3.9/itnm/ip/wip/relnotes/reference/nmip_rn.html

Go to the following link to perform an advanced search of the knowledge base:
http://www-01.ibm.com/software/sysmgmt/products/support/IBMTivoliNetworkManagerIPEdition.html

Go to the following link to retrieve all technotes for version 3.9 of Network Manager:
http://www-01.ibm.com/support/search.wss?tc=SSSHRK&q=technote+3.9

===========================================================================
Summary of fixes in V3.9.0 Fix Pack 4
============================================================================

APAR : IV16886
-----------------------
Abstract: MAINTENANCESTARTTIME AND MAINTENANCEENDTIME IS NULL IN MANAGED
STATUS TABLE.
Description: When we keeping any device in maintenance mode then no data is
saved in the managedstatus table like MaintenanceStartTime and
MaintenanceEndTime, its always NULL but the changetime column is
being filled with correct date-time.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV16886

APAR : IV17929
-----------------------
Abstract: DOMAIN SPECIFIC EVENTGATEWAYSCHEMA.<DOMAIN>.CFG FILE NOT RE-READ
BY KILL -HUP OF NCP_G_EVENT PROCESS
Description: When a schema file is read, check for domain-specific file. It looks for a
domain-specific version, and uses it if found. If not found, it defaults to the generic
version. The value changed is not picked up by the running
ncp_g_event. The action of kill -HUP of the ncp_g_event pid does not
cause it to re-read the domain specific EventGatewaySchema.domain.cfg file,
http://www-01.ibm.com/support/docview.wss?uid=swg1IV17929

APAR : IV20912
-----------------------
Abstract: ITNM AEL TOOLS DO NOT TRANSLATE TO FRENCH CHARACTER SET AS NCW T
OOLS DO ON THE SAME DROP DOWN MENU
Description: Active Event List tools for ITNM do not translate as they should.
Browser setting on French, TIP, Webtop and most parts of ITNM
automatically translate all of the text as expected, but tools from the 'alerts'
menus do not.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV20912

APAR : IV22660
-----------------------
Abstract: NCO_P_NCPMONITOR HUP
Description: When sending the HUP signal to the nco_p_ncpmonitor probe with
ITNM, the probe responds to this signal by reconnecting to the objectserver.
The objectserver reports this by two ConnectionWatch events. However,
the ConnectionWatch is connection event (problem event) is often send one second
later than the ConnectionWatch connection event (resolution
event) when the ITNM probe responds to the HUP signal. Because
LastOccurrence of the resolution ConnectionWatch event is
earlier than the problem ConnectionWatch, the generic clear
trigger doesn't clear the problem ConnectionWatch event.
This behavior is noticed so far only with the ITNM probe when
it responds to the HUP signal. For other Omnibus probe that we
operate, the HUP respond of the probe doesn't have this effect.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV22660

APAR : IV22788

-----------------------
Abstract: UNMANAGE/MANAGE ACTION CAUSES BACKUP POLLER TO POLL DEVICE.
Description: Unmanage/manage action causes backup poller to poll device.
In a failover environment with a Primary and Backup Poller
running, with the Primary domain healthy and acting primary, if
the user unmanages, then manages a device, the backup poller
will begin polling that device, and raising duplicate events.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV22788

APAR : IV23287
-----------------------
Abstract: MYSQL DB USING TOO MUCH CPU CREATING REPORTS
Description: DB push CPU usage to 100% because it have issues with
inner joins, group by and order by on SQL statements that
retrieve large amounts of data.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV23287

APAR : IV25953
-----------------------
Abstract: FIND IN MAP FUNCTION FINDS ELEMENTS THAT ARE NOT IN MAP
Description: Attempting to search for a device from a network view finds the
device even if it is not in that view.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV25953

APAR : IV26760
-----------------------
Abstract: DEFAULT INTERFACE PING POLL DEFINITION HAS 'IFADMINSTATUS'
FILTER CASUING POLLER NOT TO POLL VALID IPADDRESSES.
Description: The 'Default Interface Ping' Poll definition has following
filters under Interface filter tab against 'ncim.interfaces'
table i.e. ipAddress IS NOT NULL ifAdminStatus = 'up'.
There could be many cases where we have valid IPAddress for an
interface where ifAdminStatus is not populated. All such
scenarios won't be polled by 'Interface Ping' policy.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV26760

APAR : IV27702
-----------------------
Abstract: NETWORK VIEWS ARE TOO SLOW TO OPEN DUE TO HUGE TOPOLOGY
Description: Customers with huge topology struggle with performance due to the
long time it takes for the data to be retrieved from the database.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV27702

APAR : IV28121
-----------------------
Abstract: NCP_PERL FAILS WITH ERRORS IN WINDOWS APPLICATION ERROR
Description: ncp_perl.exe failed and the following error created in windows
application log. Faulting application name: ncp_perl.exe xxxxxxx
time stamp: 0x4d52b578 Faulting module name: libxxxxxx.dll..
http://www-01.ibm.com/support/docview.wss?uid=swg1IV28121

APAR : IV29828
-----------------------
Abstract: WEBTOOL MANUAL LOGON FAILS WHEN M_PASSWORD HAS SPECIAL CHARACTERS
Description: Login or authorization failure using Webtools when attempting a
manual connection. Webtools manual login fail when password contains special
characters.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV29828

APAR : IV30425
-----------------------
Abstract:ON FAILOVER/FAILBACK, ACTIVE POLLER SHOULD CONTINUE POLLING
UNTIL NEW POLLER IS READY TO TAKE OVER
Description: When NCIM database is remote, Poller process takes very long
time to sync with NCIM and to evaluate the scope during
start-up. The start-up time could vary from 30 mins to 3 hours
depending upon domain size, number of active policies, number of
entities to monitor and Interface/Device filters etc.
Same delay is applicable in a failover scenario, when Primary
goes down, Backup domain takes over in the given configured
period - default is 5 minutes. But Poller isn't doing any
monitoring as the poller synchronization with NCIM and scope
evaluation of Poll policies/definitions could take few hours.
Same is true for failback scenario. On failback, the active Poller was sending it's heartbeat
once the monitors were activated, but before they started to
poll resulting in a gap in monitoring.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV30425

APAR : IV31296
-----------------------
Abstract: ASSOCADDRESS AGENT TO USE IPADENTADDR VALUE INSTEAD OF THE INDEX
FOR THE IPADDRTABLE ENTRY.
Description: Unable to discovery Redware Alteon Application Delivery Controller (ADC) and
VADC (Virtual ADC) properly. Introduce a new option called UseSnmpIpValue for
AssocAddres agent to use the actual ipAdEntAddr value instead of using the index of the
ipAddrTable entry, if the device's matches the OID filter defined in this flag. By default, the
agent will use the IP value from ipAdEntAddr column for Alteon devices.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV31296

APAR : IV31587
-----------------------
Abstract: SNMP TIMEOUT
Description: For every first SNMP query to determinate if the device responds
to SNMP, the timeout configured is not used. The process does not use wait 3 seconds to
start the first retry but it reties after 1 sec of timeout. During the initial SNMP credential
discovery the configured timeout and retries for a given IP address were not always
used. Meaning that sometimes ITNM failed to gain SNMP access
to devices that were responding slowly.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV31587

APAR : IV31984
-----------------------
Abstract: IN A DISTRIBUTED ENVIRONMENT REMOTE PROCESSES UNABLE TO
RE-ESTABLISH CONNECTION WHEN THE BROKER DAEMON RESTARTED
Description: In a distributed environment - for e.g.
ServerA : ITNM Core except ncp_model
ServerB : ncp_model & NCIM database
When the broker (ncp_brokerd) on ServerA is restarted for
whatever reason - the processes on ServerB notices a
disconnection, but unable to re-establish connection with new
broker instance upon restart.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV31984

APAR : IV32386
-----------------------
Abstract: ITNM PROCESSES FAIL TO STOP AFTER DB2 DATABASE IS RECYCLED.
Description: DB2 database is recycled then some of the ncp processes are not
terminated with itnm_stop script. The proccess like ncp_model,
ncp_poller and ncp_g_event are not terminated with itnm_stop
means process are still running after itnm_stop script is completed.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV32386

APAR : IV32747
-----------------------
Abstract: NCP_POLLER CORE AT SHUT DOWN OF ITNM
Description: When shutting down the ncp_ processes, typically via
itnm_stop ncp, a core file is generated for the ncp_poller
process. When shutting down the ncp_ processes, typically via
itnm_stop ncp, a core file is generated for the ncp_poller
process when network contains devices managed via SNMP v3
and at least one SNMP poll policy is active.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV32747

APAR : IV33079
-----------------------
Abstract: NCP_DISCO CORE DUMPS WHILE EXITING IF NAT TIMER STITCHER (RUNS
EVERY HOUR OUT-OF-THE-BOX) TRIGGERED DURING SAMETIME.
Description: If the disco process is getting recycled (stopped and restarted)
on top of the hour say 10am, and the NATTimer stitcher which
runs every hour out-of-the-box is triggered at same time will
cause Disco process to core while exiting.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33079

APAR : IV33095
-----------------------
Abstract: TOPOVIZ.TRACE.LEVEL VARIABLE MISSING IN TOPOVIZ.PROPERTIES
CONFIG FILE
Description: topoviz.properties config file does not contain variable
topoviz.trace.level. Making it hard for users to adjust the
trace message level. missing parameter from properties file
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33095

APAR : IV33114
-----------------------
Abstract: MODEL MANAGED STATUS WORK
Description: Managed status state transitions and propagation from chassis
to sub components must follow a number of rules (as documented in
the ITNM best practices guide.)
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33114

APAR : IV33118
-----------------------
Abstract: REMOTE POLLERS ARE NOT STOPPING REMOTELY
Description: During the stop of ncp_ctrl all the local children are stopped and the remote
processes are stopped via the ncp_ctrl slave.

APAR : IV33128
-----------------------
Abstract: NETWORK VIEWS FAIL TO HANDLE INVALID FILTERS GRACEFULLY
Description: If a filter is invalid it throws exception causing the Wetop
Entity registration to terminate, hence some views have no
status. Status not visible on some or all nodes in a network view.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33128

APAR : IV33250
------------------------
Abstract: INVALID OID REQUESTED WHEN INTERFACE FILTERING IS ENABLED AND
ALL IN-SCOPE DEVICES ARE UNMANAGED AND REMANAGED
Description: Enable SNMP polling and apply an interface filter. The
interface filter prevents the poller from simply walking the
interface index and ensures that it has to construct the
interface index as part of the poll.
Select all the hosts that comprise the entire scope of the poll
policy. Then toggle the managed status of the devices, i.e.
unmanage them all, and then remanage them all.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33250

APAR : IV33478
-----------------------
Abstract: RECREATEANDSENDTOPOLOGY.STCH IS MISSING THE CALL TO SONMPLAYER
Description: During a rediscovery the SONMPLayer database is not
included in the list of arguments to the Merge Layers stitcher
in RecreateAndSendTopology.stch leading to incorrect
connectivity for devices advertising data via SONMP such as
Nortel switches.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33478

APAR : IV33533
-----------------------
Abstract: STRUCTURE BROWSER SEARCH FAILS TO FIND MATCH VLANS
Description: VLANs do not seem to be included in the search pattern of the
Structure Browser. The search request statement needs to be
altered to query the VLANs, VPNs, etc..
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33533

APAR : IV33578
-----------------------
Abstract: THE ITNM_DISCO.PL CORE DUMP WHEN IT IS STOPPED BY CTRL+C
Description: The itnm_disco.pl core dump when interrupted by Ctrl+C.
The issue is caused by the way that AIX and perl handle signals. On Linux and Solaris the
signals end up being called within the main thread but on AIX they can get called on the
IO thread. In addition the use of multiple signal handlers can result in inconsistent data ( i.e. cores ) according to the perl manual. The ncp_perl script has a
signal handler and as the perl script also had a signal handler it caused a core.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33578

APAR : IV33996
-----------------------
Abstract: ITNM REJECTS LINKS DISCOVERED BY SONMP STITCHER
Description: Incorrect/Missing layer 2 connectivity between Nortel Avaya switches. In
SONMPLayer.stch, ITNM rejected those links that coming from devices with
unknown/unsupported chassis type. The code was too restrictive to devices which has certain known chassis type number.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV33996

APAR : IV34108
-----------------------
Abstract: DOCUMENT THE "MONITORING POLICY DETAILS" REPORT
Description: The "Monitoring Policy Details" report does not show a list of
nodes to be queried as documented. Instead it shows a list of
nodes for a policy that has the store option set, and that is a
prerequisite for getting any report.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV34108

APAR : IV34113
-----------------------
Abstract: ITNM 3.9 NETWORK VIEW .SVG AND .PDF CANNOT BE SAVED
Description: When loading a Network View and clicking the save icon. If .svg
or .pdf format are selected, an error message "cannot be saved"
is displayed and a 0 length file created.
Unable to save network views as pdf or svg format
http://www-01.ibm.com/support/docview.wss?uid=swg1IV34113

APAR : IV35501
------------------------
Abstract: DUPLICATES IN 'SHOW CONNECTIVITY INFORMATION' GUI FOR DEVICE
Description: For VLAN (or other sub-connections), its containing entities
will be displayed in the 'show connectivity' GUI. The filter
method iterate through the connections to make sure only one
instance of such is displayed but sometimes it doesn't filter
correctly.The problem will appear as an identical connections.
Investigation into model and ncim show that there are not
duplicate interfaces to be found.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV35501

APAR : IV35661
-----------------------
Abstract: GET_POLICIES.PL DOES NOT IMPORT POLICY INTERFACE FILTER WHEN
MIGRATING FROM 3.8 TO 3.9
Description: When trying to migrate policies from 3.8 to 3.9 ,
policy interface filter is lost. In ITNM 3.8 interface filter is stored on policy.ifScopeId
column. In 3.9 interface filter is defined at template level (template.ifScopeId).
The migration process:3.8 - ncp_perl get_policies.pl -from domain=NCOMS -to
file=get_policies_NCOMS.dat. 3.9 - ncp_perl get_policies.pl -from
file=get_policies_NCOMS.dat -to domain=NCOMS
fails to transfer the interface filter (policy.ifScopeId -> template.ifScopeId).
http://www-01.ibm.com/support/docview.wss?uid=swg1IV35661

APAR : IV35842
-----------------------
Abstract: POLLER THROWS UNSUPPORTED DATABASE TYPE: ORACLE11 WHEN POLLING
IS FILTERED ON A VIEW.
Description:
Create a snmp policy. Filter the policy on a view created in the
Network Views. Save the policy.
In the ncp_poller.DOMAINNAME.trace we get the following error
message: Exception in thread "Thread-21" java.lang.RuntimeException:
Unsupported database type: oracle11 at xxxxxxx.java:66.
The poll might succeed, but in some cases they will not.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV35842

APAR : IV36091
-----------------------
Abstract: SNMP COMMUNITY STRING INDEXING FAILED ON NORTEL DEVICES AND BAD
SNMP COMMUNITY STRING REPORTED
Description: NortelSwitch agent uses community string indexing to query FDB data,
causing SNMP authentication failure alarm generated. It has been found out that
Nortel/Avaya switches that supports standard FDB MIBs do not support community string
Indexing. The NortelSwitch agent code has been updated to avoid using community string
indexing if the device do not support private FDB MIBs and only supports standard FDB
MIBs. NortelSwtich agent would use community string indexing only if the device supports
private FDB MIBs.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV36091

APAR : IV36109
-----------------------
Abstract: DISPLAYING THE DEVICE ICON IN LEGACY MODE IS NOT WORKING.
Description: Displaying the Device icon in Legacy mode is not working as per
the below doc to change the device icon as like ITNM 3.8 version
https://www.ibm.com/support/knowledgecenter/SSSHRK_3.9.0/com.ibm.networkmanagerip.doc_3.9/itnm/ip/wip/admin/task/nmip_adswitchingtov38vismode.html
Label is coming on top the device icon,so can not select the
label to move or changes the position. When network view map was displayed in legacy
mode, labels appeared on top of the icons in the center.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV36109

APAR : IV36368
-----------------------
Abstract: RUNTIME EXCEPTION ERROR DISPLAYED WHEN SELECTING "DEVICE
MEMBERSHIP" LINK FOR A POLL POLICY.
Description: When selecting the "device membership" link for poll policy with
a device filter that uses either the displayLabel or description
field from the mainNodeDetails table in ncim, the following
error is displayed:
CTGTD1000E An error occurred while processing the request to the server.
Detail: RuntimeException Error executing query: Getting device membership for policy:
Ambiguous column(displayLabel).
http://www-01.ibm.com/support/docview.wss?uid=swg1IV36368

APAR : IV36581
-----------------------
Abstract: PREVIEW OF COLLECTION VIEWS IS NOT ORDERED AND IS HARD TO
SCAN/SEARCH
Description: Preview of collection views is not ordered and is hard
to scan/search When we have to create a Network View and I select
as type "VPN" the order of available collection is not
alphabetical or numerical, so it is very difficult to identify and then select a
certain Virtual Circuit (VC).
http://www-01.ibm.com/support/docview.wss?uid=swg1IV36581

APAR : IV36679
-----------------------
Abstract: NULLPOINTER EXCEPTIONS REGISTERED ON GUI WHEN SEARCHING AN
EMPTY DATASTORE
Description: Open a NetworkView viewType which contains an empty view tree,
search for a view name. NullPointer Exceptions registered on GUI when searching an
empty data store.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV36679

APAR : IV36728
-----------------------
Abstract: AEL LIST NOT UPDATED WHEN "SHOW NETWORK VIEWS TREE?" DISABLED
Description: Replication steps:
1) open a "Network Health View" window. Select device or group
and AEL is updated.
2) Click the top right down arrow and select personalize.
Disable the "Show Network Views Tree?""and set "Map ID" to 1.
3) Save and again select devices, AEL will not update in Health View, personalize the
network views page to display map only, not showing the network views tree. Then
click on devices on the map, AEL won't filter by device selected.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV36728

APAR : IV36856
-----------------------
Abstract: MEMORY USAGE REPORT USES INCORRECT POLL POLICY
Description: The "Memory usage report" uses the incorrect policy.
The code, and consequently the documentation are incorrect as the
"Memory usage report" does not depend on stored data from the
CiscoMemoryPool policy.
https://www.ibm.com/support/knowledgecenter/SSSHRK_3.9.0/com.ibm.networkmanagerip.doc_3.9/itnm/ip/wip/ref/reference/nmip_ref_report_memoryusage.html
The "Memory usage report" uses the memoryPoll policy.
The report uses a policy called ciscoMemoryPoll that don't
store data on database and results in an empty report
http://www-01.ibm.com/support/docview.wss?uid=swg1IV36856

APAR : IV37074
-----------------------
Abstract: POLLER PRODUCES A LARGE AMOUNT OF EVENTS AFTER STARTUP
Description: The poller generates hundreds of Link State change events on
startup. Most of these events may be for empty ports or access
ports if the Link State polls did not filter out these
interfaces, and of little value. With the introduction of FP02 the initial state of all polls are
determined by existing events. If there are no existing events then the initial state is
determined to be "clear". In networks where ports are not yet used, but have not be
placed into AdminStatus down, having the initial state set to clear can result in
unnecessary Snmp Link State down events. Previously the first poll would have been used
to establish the initial state for the Snmp Link State polls, resulting in no link down events.
for customers who do not set their port's AdminStatus this was the desired behavior.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV37074

APAR : IV37322
-----------------------
Abstract: INCORRECT VLAN MODELING WITH HUAWEI SWITCHES
Description: Incorrect VLAN modelling with Huawei switches via StandardSwitch agent. The
Huawei device uses dot1dBasePortIfIndex to map the port number (PortList) into interface
index. StandardSwitch agent code does not perform this mapping while discovering VLAN
information.
The StandardSwitch agent has been updated to add the port to ifIndex mapping into the
StandardSwitch agent while parsing dot1qVlanCurrentEgressPorts data.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV37322

APAR : IV37587
-----------------------
Abstract: MIB GRAPH DOES NOT RENDER
Description: In a failover setup using a non replicating NCIM the backup
poller can interfere with the proper scope evaluation of the primary poller
Error message:
"No data was received from server after multiple polling
intervals. Please check the graph configuration, or see the
log for more information.
The backup Poller would evaluate it's scope while in stand by
mode and update the scopedEntity table. This table would
then be used by the primary Poller when re-evaluating scope,
resulting in entities to be missed. This issue impacts Mib
Grapher and regular policies.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV37587

APAR : IV37779
-----------------------
Abstract: MINOR CODE FIXES FOR ITNM_PATHTOOL.PL
Description: Added function
prototype. Added checks for null values. Function called too early to
check prototype. Use of uninitialized value in pattern match (m//).
Use of uninitialized value in concatenation (.) or string.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV37779

APAR : IV38084
-----------------------
Abstract: WARNING IN NCP_POLLER LOGS "CANNOT CAST ATOMIC VALUE"
Description: Following message seen in the poller logs:
Warning: W-RIV-002-043: [3982490528t] xxxxxx.cc(507)
Cannot cast atomic value from parsed to stored ! Clause & p2
evaluated to 'inbandwidth' of type String which could not be.
This is a cosmetic warning message and should be suppressed poller prints warning
message when poll incomplete.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38084

APAR : IV38239
-----------------------
Abstract: MISSING LINK DUE TO INVALID ARP ENTRY FROM NETSCREEN DEVICE
Description: Missing link from NetScreen device due to Invalid ifIndex found
in ipNetToMediaPhysAddressArp entry was rejected by ArpCache
agent because the agent failed to find the interface data with
ifIndex=0, which is the first numeric number from the index of
ipNetToMediaTable. Missing connectivity due to invalid ifIndex found in
ipNetToMediaPhysAddressArp entry in NetScreen device was
rejected by NetScreenArpCache agent because the agent failed
to find the interface data with ifIndex=0, which is the
first numeric number from the index of ipNetToMediaTable.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38239

APAR : IV38395
-----------------------
Abstract: FIND IN NETWORK VIEW TOOL FAILS FROM DEVICE ICON IN HOP VIEW
Description: On some occasions the attempt to right click a device icon from
the Network Hop View and select "Find in Network View" can fail.
Other occasions the attempt succeeds but can take as long as 10
minutes with large view sizes and containment.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38395

APAR : IV38462
-----------------------
Abstract: SCOPE.SPECIAL EXTENDED TO CHASSIS AND INTERFACES.
Description: "scope special" is an existing feature we extend to chassis and
interfaces. Please see following RFE..
http://www.ibm.com/developerworks/rfe/execute?use_case=viewRfe&CR_ID=31587
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38462

APAR : IV38610
-----------------------
Abstract: ATTEMPTING TO CREATE A COPY OF MIB GRAPHER FAILS
Description: In MibGraph, click on the copy button, new mibgraph appeared
with same configuration. Click on copy button from this new
mibgraph, the next copy lost all the configuration information.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38610

APAR : IV38617
-----------------------
Abstract: MODIFYING POLL POLICIES POLLER STATUS ICON TO CHURNS
Description: Poll GUI has been showing the spinning "Status Pending". This
happens after modifying the poll policy.
The status never changes even after restarting browser and
restarting the TIP server. After modifying poll policy, Poll GUI showed the spinning "Status
Pending" and never changed.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38617

APAR : IV38618
-----------------------
Abstract: MIB GRAPHER UNNECESSARILY POLLS WHEN IN HISTORICAL MODE
Description: When it's set to draw historical graph, mibgrapher also sent
realtime poll requests. Although data is discarded, the poller
should not be polling the network.
Realtime polling also runs when mibgraph is running in historical mode
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38618

APAR : IV38744
-----------------------
Abstract: AFTER APPLYING ITNM V3.9 FIX PACK 3, NCP_MIB CORE DUMP FILE. REFERENCE
APAR IV27953.
FAULTY MEMORY HANDLING RESULTS IN NCP_MIB CORE DUMP FILE.
Description: This APAR is created as AE/MDV APAR with "MDVRGRS" class.
APAR IV27953 APAR: MDVPARTL 3.9.0-TIV-ITNMIP-FP0003
After applying ITNM v3.9 Fix Pack 3, ncp_mib cores. Reference
APAR IV27953. Faulty memory handling is leading to
undefined behavior which results in
occasional cores during mib processing on some platforms.
This problem originated in FP03.ncp_mib core dumps due to memory corruption when processing a MIB variable defined
as <mibVariable> = { <definedMibVariable> <value1> <value2>
... } as opposed to simply { <definedMibVariable> <value1>}.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38744

APAR : IV38974
-----------------------
Abstract: STATUS MISMATCH BETWEEN THE NETWORK VIEWS AND NETWORK VIEW TREE
Description: Network View tree highest severity status is less than or not
as severe as some of the contents/entities of the network view.
The Network View Tree and the Network View construct the status
filter based on different criteria (event field):
Network View Tree -> NmosEntityId IN (722382, ...)
Network View -> NmosObjInst IN (722382, ...)
When the event has the NmosEntityId and NmosObjInst set to
different values: NmosEntityId 86958 Vlan interface [Vl686] ifIndex 20
NmosObjInst 722382 main node , the Network View Tree and the
Network View will show different severity status.
Both (view and tree), should build the filter using the same
event field.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV38974

APAR : IV39072
-----------------------
Abstract: ATOMIC CAST FAILURES FOR HIGH DISCARDRATE POLICY USING INTEGER
Description: The High Discard Rate policy has incorrect expression.
It was casting results to int when it should have been long64.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV39072

APAR : IV39195
-----------------------
Abstract: MONITORED DEVICE DETAILS REPORT PDF FORMAT IS NOT CORRECT.
Description: Monitored Device Details report format is not correct when
exporting as a PDF format.
To replicate the issue:
Common Reporting--> Public Folders > Network Manager >
Monitoring Reports --> Run Monitored Device Details as PDF
format. Customer views monitored device details report in PDF format
and report is broken up into many pages with too much
white space between charts.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV39195

APAR : IV39203
-----------------------
Abstract: DEVICE AVAILABILITY SUMMARY REPORT HAS WRONG UPTIME AND DATE IS
MISSING.
Description: Device Availability Summary report showing wrong SysUpTime and
Uptime, also there is no start Date-time and End date-time in
the report.
Replication Steps:Common Reporting --> Public Folders > Network
Manager > Summary Reports --> Run the Device Availability
Summary report. Report does not include the Start and End time of when the
user ran the report. Also the sysUpTime displayed in the
header is incorrect.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV39203

APAR : IV39428
-----------------------
Abstract: ITNM GUI CONSTRUCTS WRONG FILTER FOR ALERT VIEWS
Description: ITNM GUI constructs wrong filter for Alert views. When opening
the AEL from the Network View the creation filter is incorrect. In some cases there is a
mismatch between the status in the
Network View Tree and the devices in the Network View
http://www-01.ibm.com/support/docview.wss?uid=swg1IV39428

APAR : IV39551
-----------------------
Abstract: CORE DUMP OF NCP_G_EVENT ON
SOLARIS IN ANY CURRENT RELEASE OF 3.9.
Description: Core dump of ncp_g_event on Solaris whilst performing isolated
suppression.
Memory corruption is occurring as a result of a bug in SunStudio
11. There appears to be a problem with the std::deque STL
container in the version of SunStudio used to build 3.9. This is
Oracle issue CR 6363210: std::deque memory corruption. This is
fixed in subsequent patched versions of SunStudio, to be
integrated into the build system for future releases of ITNM.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV39551

APAR : IV39822
-----------------------
Abstract: POLLER FALLS BEHIND ON SNMP BANDWIDTH POLLS WHEN NOT USING AN
INTERFACE FILTER
Description: Poller falls behind on Snmp Bandwidth polls when not using an
interface filter in two scenarios :
a) During start-up
b) Upon completion of discovery while processing topology updates.
And it takes minutes/hours to recover depending on the number
of monitors in the polled scope.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV39822

APAR : IV39998
-----------------------
Abstract: NETWORK MANAGER IP VIEWS OUT OF MEMORY
Description: ITNM 3.9 FP3 - This APAR is created as AE/MDV APAR with
"MDVPARTL" class.
MDVPARTL 3.9.0-TIV-ITNMIP-FP0003
Currently the ITNM WebGUI is not translating mainNode only
views into aggregated NmosObjInst filters, this is causing a
functional issue whereby device specific views are not
generating alerts on each devices containment, rather just on
the device itself. There is also a knock-on performance problem
whereby, when endNodes="true" is set the generated filter is
based on the full set of device containment as NmosEnityId
rather than the more efficient NmosObjInst. "Hung Thread" and
"Out-of-Memory" errors in logs; followed by system slowdown.
In extreme cases the UI/Tip server java process crashes.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV39998

APAR : IV40185
-----------------------
Abstract: DEVICES MISSING COLLECTOR DISCOVERY COLLECTOR ADDRESSTRANSLATION
Description: Customer has collector only discovery.
Missing device is in CollectorDetails.returns
seen in CollectorInventory.returns, but is missing an entry for it's own IP address in the
'm_ExtraInfo' => {'m_AssocAddress' => [section.so ip is not found here
select * from translations.collectorUniqueToWorkAddress where
m_WorkAddress = or
select * from translations.collectorUniqueToWorkAddress where
m_UniqueAddress=
and missing from scratchTopology.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV40185

APAR : IV40211
-----------------------
Abstract: DBI_FACTORY DOES NOT INITIALIZE CLIENT_LOCALE REQUIRED FOR
INFORMIX ON WINDOWS.
Description: Windows Operating System with Informix database
containing multibyte characters.
If you have a Windows installation of ITNM using the default
Informix (IDS) database and you have data in the IDS database
that includes multibyte characters, then using the perl scripts
to interact with the data will fail.
For example, if you name your Network Views using Japanese
characters, then use get_network_views.pl or
ITNMExportNetworkViews.pl to export the Network Views, the
script will fail to export those views with Japanese characters.
You may receive an error from the script such as:
DBD::ODBC::st fetchrow_hashref failed: [Informix][Informix ODBC
Driver]Inexact character conversion during translation.
(SQL-HY000).
http://www-01.ibm.com/support/docview.wss?uid=swg1IV40211

APAR : IV40479
-----------------------
Abstract: DEADLOCK OCCURS WHEN MULTIPLE THREADS TRY TO LOCK BROKER
CONFIGURATION FILE LEADING TO HANGING.
Description: This behavior occurs specially on Windows systems, where
both ITNM and ITM agent for ITNM are installed, but might
happen on other operating systems.
When deadlock occurred , error messages like:
Error: An unrecoverable error occurred. Please contact IBM
Support.
Fatal Error: Unable to lock mutex. found in file xxxxxxxx.cc at
line 112 - Invalid argument in mutex created from xxxxxxxxx.cc.
Application terminating due to error
are generated in logs associated with processes that are trying
to lock the config file.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV40479

APAR : IV40995
-----------------------
Abstract: INCOMPLETE LOGIC UNDER RESOLVE VRRPESSES.STCH CAUSES
VIRTUAL VRRP IPADDRESS TO BE MAIN CHASSIS IP OF THE DEVICE
Description: VirtualIP of VRRP assigned as main chassis IP when it should
not.During stitching the stitcher 'ResolveVRRPAssocAddresses.stch'
that processes all records from Nokia VRRP & other agents is
supposed to find MasterIp and make that as BaseAddress if we
have found the device in the discovery Phase1 using Virtual IP
Address, but it does not do that extensive check, resulting
mainnode base address with Virtual IPAddress.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV40995

APAR : IV41267
-----------------------
Abstract: ON RESTART MODEL SHOULD CHECK FOR (AND FIX) MISSING
DOMAINMEMBERS ENTRIES IN NCIM
Description: If rows are deleted from the domain Members table this can result
in missing entities and links in the network views. Upon a
re start the model process does not detect this missing data and
will therefore not restore the missing entities and links to the views.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV41267

APAR : IV42617
-----------------------
Abstract: DOMAIN-SPECIFIC DBLOGINS.CFG MANDATORY FOR POLLER WITH
MULTIPLE DOMAINS
Description: Polling failover failed if domain specific version of
DbLogins.cfg didn't exist.
The get_policies.pl script is used to write changes in Poll
Policies to the domain's version of MonitorPolicies.xml.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV42617

APAR : IV42667
-----------------------
Abstract: SAE AUTOMATION THAT SHIPS WITH ITNM'S SERVICE AFFECTING EVENTS CAPABILITY DOC UPDATE
Description: Document update on ITNM's Service Affecting Events capability can be
deployed correctly in a multi-tier OMNIbus environment.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV42667

APAR : IV42743
-----------------------
Abstract: ENTITYCOUNT FIELD FROM NCIM.DOMAIN SUMMARY IS NOT RELIABLE
Description: The entityCount field from ncim.domainSummary table should store
the number of entities in the domain.
However, count(*) from  ncim.entity table is not equal
with entityCount value
This value suppose to be updated at the end of each discovery.
The problem occurs only on ITNM 3.9 version.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV42743

APAR : IV42747
-----------------------
Abstract: SERIAL NUMBER ATTRIBUTE IS MISSING FOR JUNIPER ENTITIES
Description: Serial Number info is missing for Juniper Entities. ITNM Juniper
agent does retrieve the card serial number info but gets
removed at the stitcher level. Card serial number info retrieved from Juniper agent was
removed at the stitcher level.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV42747

APAR : IV43090
-----------------------
Abstract: FAILOVER NCP_G_EVENT SERVER ATTEMPTING TO UPDATE OMNIBUS PRECISION FIELDS
Description: The SAE plugin in ncp_g_event on the backup server attempts to
update OMNIbus after discovery completion. These transactions
should not take place from the backup and are causing Warning
messages: ncp_g_event.NCOMS_B.trace:Wed May 15 16:21:25 2013 Warning: A
generic error has occurred found in file xxxxxxx.cc at line xxxx
General failure: Failed to retrieve results for query..At the end of a discovery, the gateway
on the primary updates the SAE information in Omnibus from the primary gateway.
ITNM failover GW also try to update this information. It should not do this.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV43090

APAR : IV43140
-----------------------
Abstract: CREATEMPLSPE.STCH BAD EVAL LINE 534 CAUSING DISCO SYNTAX ERROR
Description: CreateMPLSPE.stch stitcher, line 534 has a syntax eval error.
Error causes syntax error in line 29 in disco log file.
where
(
m_BaseName = eval(text,'$entName')
AND
m_IpAddress = (eval,text,'$bgpAddress')
);
http://www-01.ibm.com/support/docview.wss?uid=swg1IV43140

APAR : IV43184
-----------------------
Abstract: NAT ADDRESS SPACE IS NOT POPULATED CORRECTLY.
Description: NAT address space is not populated correctly and m_AddressSpace is missing for some of the entities.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV43184

APAR : IV43870
-----------------------
Abstract: NCP_POLLER STOPS POLLING AND LOGS OUT-OF-MEMORY WHEN
COLLECTION RATE EXCEEDS STORAGE RATE
Description: ncp_poller stops polling and logs Out-of-memory when collection
rate exceeds storage rate. The poller buffers data in memory when it is unable to store
due to database connection or collection rates exceed storage
rates. The buffer increases until memory is exhausted during prolong database outages, or
excessive polling load, the Poller would continue to collect poll data and store in
a memory queue. This queue would continue to grow until
poller ran out of memory.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV43870

APAR : IV43944
-----------------------
Abstract: DISCOVERY TIME INCREASE AFTER EACH FULL DISCO PROCESS.
Description: The problem is that discovery keeps on taking longer and longer the more
often it runs.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV43944

APAR : IV44090
-----------------------
Abstract: HOP VIEW, SUBNET CONNECTIVITY IS ONLY SHOWING THE SEED DEVICE.
Description: In the hop view, if I choose subnet view, only the seed device
is displayed. However if I go to the network view and choose
subnet view, I see the seed device and subnet icons. In the hop view, if I choose subnet
view, only the seed device is displayed. However if I go to the network view and
choose subnet view, I see the seed device and subnet icons.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV44090

APAR : IV44161
-----------------------
Abstract: CISCOSWITCHSNMP AGENT CAN FAIL TO DISCOVER CONNECTIONS WHEN
MULTIPLE VLANS ARE CONFIGURED
Description: In some cases the CiscoSwitchSnmp agent can fail to retrieve
connections when the device is configured with multiple VLAN. Missing link between devices which don't have CDP enabled.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV44161

APAR : IV44176
-----------------------
Abstract: MISSING EXTRAINFO DATA FROM DETAILS COLLECTOR
Description: All collector data which added by customer via extraInfo.cfg is missing in the TIP.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV44176

APAR : IV44266
-----------------------
Abstract: NCP_TRAPMUX NOT FRAGMENTING SNMP PACKETS FOR LARGE MESSAGES
Description: SNMP packet is too
large (i.e. > 1500bytes).
This seems related to IV00342.
When ncp_trapmux receives a trap exceeding 1500 bytes, it
can result in the error message 'System Error sending packet.
Message too long' when it tries to forward the trap to a remote sink Host.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV44266

APAR : IV44471
-----------------------
Abstract: SAM COLLECTOR FAILED TO DISCOVER VPN INTERFACES
Description: Missing interfaces in discovery, due to absence of
<displayedName> tag in SOAP response for interfaces data
This results the collector to fail.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV44471

APAR : IV45047
-----------------------
Abstract: AMS EXTRAINFO->M_IFOPERSTATUS UNKNOWN
Description: Device's m_IfOperStatus always show 4(Unknown) after downloaded by IDM
5529 collector.Fix collector to download data correctly.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV45047

APAR : IV45185
-----------------------
Abstract: PERFORMANCE REPORTS DO NOT HONOR THE HOSTNAME OR IP
ADDRESS FILTER
Description: When attempting to filter results for Performance reports (for
example "Bandwidth Utilization" or "Device Availability
Summarization", the report does not honor the filter but
instead provides output for every device regardless of the
filter content.
Filtering in performance reports on ip address fails to
display reports correctly. Instead it returns ALL ip addresses.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV45185

APAR : IV45258
-----------------------
Abstract: AFTER APPLYING ITNM V3.9 FIX PACK 3: TAGMANAEDENTITIES.STCH DOES
NOT HONOR ENTITYTYPE=2 FILTER RESULTING IN UNMANAGED ENTITIES
Description: This APAR is created as AE/MDV APAR with "MDVRGRS" class.
MDVPARTL 3.9.0-TIV-ITNMIP-FP0003
TagManagedEntities.stch does not honor EntityType=2 filter when
the condition is based on EntityOID. This was working correctly
in earlier builds and is broken on FP3.
After the enhancement of OQL optimization made in FP3, an
OQL query involving a mix of indexed and unindexed queries
on the same table AND'd together could in some cases result
in the wrong data being returned.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV45258

APAR : IV45467
-----------------------
Abstract: TOPOVIZ CREATES ILLEGAL SQL QUERY WHEN USING ADVANCED SQL FILTER
Description: Topoviz creates illegal sql query when using advanced sql filter.
When creating a network view with an advanced sql filter in topoviz the
backend creates an illegal sql query which doesn't work in mysql.
Caused by: java.lang.RuntimeException: Error executing query:xxxx. The query used in
network filtered view was prefixed by
"t.", to avoid column name conflicts, but didn't work for all cases.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV45467

APAR : IV45561
-----------------------
Abstract: DOCUMENTATION LACKING FOR COLLECTOR BASED DISCOVERIES
Description: When attempting to configure collector based discoveries there
is not enough information provided in the info center nor in
the configuration files to determine the what changes to make
to achieve desired results. This APAR was created in order to
provide a more detailed level of documentation for use when
using all collector based discoveries.
Added Timeout property. Added more detailed information on
the data retrieved by GetEntitites and GetONT.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV45561

APAR : IV45637
-----------------------
Abstract: STRUCTURE BROWSER DOES NOT SEARCH THROUGH ALL ELEMENTS OF THE
DEVICE STRUCTURE TREE
Description: The user is unable to find a node in the Device Structure Tree,
which in fact does have a match to the specified search phrase.
The reason is that the Device Structure Tree Search facility
uses a static SQL query to search for predefined entity
attributes of a specified entity Id, only for the following
entity types: [interface, ip endpoint]. With the addition of new
entity types (VLAN for example), this SQL query became obsolete
and therefore not able to find everything which could be in the
Device Structure Tree.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV45637

APAR : IV45870
-----------------------
Abstract: AVAYA PHYSICAL INVENTORY AGENT CORE DUMP.
Description: The agent fails to complete during discovery, exits
unexpectedly and a core file is generated.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV45870

APAR : IV45912
-----------------------
Abstract : ADDJUNIPERENTITYCONTAINMENT STITCHER ISSUE
Description: Some interfaces of Juniper were not in the correct containment after enabling JuniperBoxAnatomy agent.
In AddJuniperEntityContainment stitcher, some interfaces would missed out while
resolving to the correct containment. The oql statement should use m_BaseName instead
of m_UniqueAddress while retrieving all the interfaces entity from
workingEntities.finalEntity table. The stitcher code includes changes to include the
jnxContentsPartNo and jnxContentsRevision attributes into the entity.
https://www.ibm.com/support/docview.wss?uid=swg1IV45912

APAR : IV46243
-----------------------
Abstract: When Interface alarms incorrectly associated to Chassis, they
suppress other unrelated alarms on chassis
Description: Alarms for interfaces that could not be found in the topology
(either because the interface was removed from the topology to
improve performance or because of other issue with syslog
interface alarms -IV47678 ) were suppressing other unrelated
alarms on the chassis and down stream network. This would
happen because the alarms were being updated to set the
NmosEntityId to the EntityId of the chassis. The
LookupIfEntry.stch prevents the alarm being treated as a
chassis alarm on it's first occurrence but subsequent
occurrences are treated by RCA as at the level of the chassis
because of the value of the NmosEntityId.
An event raised against an interface which is not found will become root cause,
suppressing other events on the containing chassis, if it raised more than once.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV46243

APAR : IV46301
-----------------------
Abstract: CHANGE/IMPROVEMENT OF ITNM START/STOP/STATUS SCRIPTS
Description: Currently the behavior of the itnm stop/start/status is
unsatisfactory. The itnm_status script tries to show the status of all
processes running under nco_pad.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV46301

APAR : IV46436
-----------------------
Abstract: ITNM TOOLS SKIP PROCESSING OF THE ENTITYID LIST LIST WHEN A
COOKIE IS USED
Description: The 'target' parameter contains the parameter name instead of
the entity Id. Using a cookie in the URL Tool definition was breaking the
value for parameter 'target' in the resulted URL.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV46436

APAR : IV46560
-----------------------
Abstract: ITNM GATEWAY (NCP_G_EVENT) CANNOT HONOUR COMMON GATEWAY
PROPERTIES.
Description: There is no capability in the Gateway for the user to
configure common Gateway properties such as IPC.TIMEOUT
ITNM gateway needed to support Ipc.Timeout property (common
property) in order to drop the connection to the Primary ObjectServer
quickly if no activity was detected. Otherwise it would wait 60 seconds
before timing out (default value) and then would retry 2 more times before
failing over.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV46560

APAR : IV46675
-----------------------
Abstract: DLA BOOK HAS A ROUTER CLASS WITH A HARD CODED NAME OF "ROUTER".
Description: DLA book has a router class with a hard coded name of "router".
The Classes Router and Bridge have the instance name hard
coded as Router and Bridge
http://www-01.ibm.com/support/docview.wss?uid=swg1IV46675

APAR : IV46734
-----------------------
Abstract: DYNAMIC DISTINCT VIEWS CHANGE THEIR PARENT-CHILD HIERARCHY IN THE NETWORK VIEW TREE
Description: When the filter for Dynamic Distinct view is created by the user, he may or
may not list the table fields in alphabetic order.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV46734

APAR : IV47106
-----------------------
Abstract: THE ITNM DLA DO NOT EXPORT ENTITY DETAILS THAT HAVE CUSTOMER
MEANINGFUL INFORMATION
Description: The ITNM DLA do not export entity details that have customer
meaningful information. ITNM DLA was updated to export the
entity detail as an extended attribute of the class when the new
property is defined as ncp.dla.entity.details=true..
http://www-01.ibm.com/support/docview.wss?uid=swg1IV47106

APAR : IV47312
-----------------------
Abstract: NCP_DH_ARP CORE BAD PACKET RECEIVED WITH MISSING FIELD.
Description: Arp helper restarts, or essentially cores.
As we got the bad packet to ARP helper, there isn't any field
of RIV_REQUEST -"RivAgentDataRequest".
This caused the Fatal error in DAPRCheckHelperName.
Multiple occurrences of the same client Id detected.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV47312

APAR : IV47318
-----------------------
Abstract : ASSOCADDRESS AGENT ISSUE
Description : ITNM assigned wrong IP address into interface, where the wrong IP address is
a broadcast address ending with 255. In the new ipAddressTable, check the value of
ipAddressType. If ipAddressType value is broadcast(3), discard the entry and do not assign
the broadcast address into the interface record. This fix affects the following agents:
AssocAddress, Interface and InetRouting.
https://www.ibm.com/support/docview.wss?uid=swg1IV47318

APAR : IV47598
-----------------------
Abstract : INTERFACE AGENT FAILS TO CHECK VALIDITY OF IPADDRESSPREFIX
Description : Subnet IP address for certain devices are not presented correctly.
Modified Interface agent to check the validity of ipAddressPrefix response. If the response
is not valid, agent should gather subnet information from old IPTable.
https://www.ibm.com/support/docview.wss?uid=swg1IV47598


APAR : IV47678
-----------------------
Abstract: GATEWAY INTERFACE LOOKUPS SHOULD ACCEPT PLAIN IFNAME IN
LOCAL PRI OBJ IF NO MATCH IN THE FORMAT "IFENTRY.<IFNAME>"
Description: Syslog interface alarms were not being matched to interfaces in
the topology. ExtractIfString.stch assumes that in the ifDescr
value in the LocalPriObj or LocalRootObj were prefixed with a
"ifEntry." string as is the convention for the ifIndex in
interface traps. The problem is that these are not currently matched to the
interface by the gateway.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV47678

APAR : IV47759
-----------------------
Abstract: CHUNK3 AND CHUNK4 ERRORS CREATED BY CONTAIN RESOLVABLE ENTITIES
STITCHER
Description: The variables chunk3/chunk4 are deleted within a loop causing
two errors each time through the loop:
2013-07-20T08:25:17: Warning: W-RIV-002-139: [1543t]xxxxxxxxx.cc(212)
The following stitcher rule failed to complete correctly.
This causes thousands of error messages each time the topology
is stitched.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV47759

APAR : IV47816
-----------------------
Abstract: RCAPATHTOOL RESULTS DIFFER FROM ACTUAL CONNECTIVITY(LAYER2) EVEN
THOUGH RCA END RESULTS ARE ACCCURATE
Description: RCA path tool(ncp_pathtool) supposed to use the shortest path,
instead it uses some random path which is inaccurate. ncp_g_event pathtool
shows too many nodes that are unrelated to the A-Z path.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV47816

APAR : IV47867
-----------------------
Abstract: NETNUMBER AND IPNUMBER IN NCIM.IPENDPOINT INCORRECTLY CALCULATED
Description: The NETNUMBER and IPNUMBER are incorrectly calculated
for IPv6 addresses. Some translations of IPv6 addresses to number form failed on
little endian systems because ntohl() was used with 64-bit
values (it only supports 32-bit).
User running on, say, linux on intel x86 (little endian)
encountered the issue while big endian systems (e.g Solaris
on SPARC, or AIX on PowerPC) wouldn't hit the issue.
The end result was that the IPNUMBER and NETNUMBER values in
NCIM's ipEndPoint table could be wrong.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV47867

APAR : IV47900
-----------------------
Abstract: REPORTS SHOW GARBLED POLICY NAME FOR OUTPUT IN RUSSIAN ENV
Description: (Windows Only) The report output for Bandwidth Utilization Drill down
and Generic Trend Analysis Drill down Report (possibly others)
will show garbled characters for the policy name. The Performance Drill down Cognos
Reports shows the policy name garbel.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV47900

APAR : IV48135
-----------------------
Abstract: NETWORK VIEW VISUALISATION ISSUES
Description: MPLS and subnet views not displayed; IP Filtered Views membership
inconsistent for end nodes included/excluded; hop views and network views do not agree
on links displayed for layer 3 connectivity
http://www-01.ibm.com/support/docview.wss?uid=swg1IV48135

APAR : IV48866
-----------------------
Abstract: THE ALCATEL5620SAMSOAP COLLECTOR PRODUCES NCP_DISCO ERRORS WHEN
PROCESSING THE ALU SGW IPV6 ADDRESSES.
Description: The issue is that we are discovering IPv6 addresses from the
SAM but the Collector framework assumes IPv4 and sets the
protocol to IPv4. Collector framework doesn't support IPv6 (it assumed IPv4)
and the EMS is providing IPv6 addresses.
As a result the IPv6 addresses end up in the Disco tables
with the incorrect protocol set. This leads to warning such as;
Warning: A generic error has occurred found in file
CDiscoStitcherIsInScope.cc at line 155 - IP address: xxx is
not valid
http://www-01.ibm.com/support/docview.wss?uid=swg1IV48866

APAR : IV49264
-----------------------
Abstract: CISCO WEBTOOL SPECIFIED OVER SSH V2 EXCEED THE INITIAL WINDOW SIZE AND FAILED
Description: When creating a cisco webtool (or modify an existing one) that
should execute many commands, exceeding a buffer limit ,
it exceeds the initial window size and failed with
message channel window size adjustment.
This happens only if connect by SSH V2. When connecting by
telnet or SSH V1 it works fine.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV49264

APAR : IV49805
-----------------------
Abstract: DISCO.FILTER CUSTOM TAGS FAILS TO TAG IP ADDRESSES.
Description: Custom tags fails with only garbage in the log file.
disco.filter CustomTags failed to tag the specified
m_CustomTags for the IP address due to the invalid syntax in
AddCustomTags.stch
http://www-01.ibm.com/support/docview.wss?uid=swg1IV49805

APAR : IV49860
-----------------------
Abstract: THE ITNM DATABASE SCRIPT (ORACLE) GRANTS WIDER PRIVILEGES THAN IT NEEDS.
Description: The ITNM database script (Oracle) grants wider privileges than
it needs.The proposed change is to minimise those privileges in database creation scripts.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV49860

APAR : IV49968
-----------------------
Abstract: CIRCULAR RCA
Description: At times, Circular RCA observed without any Root Cause based on
the sequence the events come in.
Circular suppression involving either 3 or 4 entities. e.g.
A suppressing B suppressing C suppressing D suppressing A.
One of the suppression types is always connected
suppression.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV49968

APAR : IV50276
-----------------------
Abstract: TIP RUNS OUT OF MEMORY WHEN USER OPENS TOPOVIZ WITH A LARGE SAVE
VIEW
Description: TIP runs out of memory because we allocate memory for the entire
uncompressed map before sending it to the client. During loading the Network View graph consumes too much memory causing OOM.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV50276

APAR : IV50278
-----------------------
Abstract: PATH TOOL FAILS WHEN USING A NON-DEFAULT TIP PORT
Description: If trying to add a path to trace via the gui, on a TIP using a
non-default port the following error message appears in the GUI:
"Error Tracing Path An error occurred while running the PathTrace Tool.
Check ncp_topoviz log file for a detailed error message or
press View Path Trace Details to view the PATH Tool output. ..."
PATH creation via the GUI failed, if TIP is configured to use non-default port.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV50278

APAR : IV50295
-----------------------
Abstract: VLAN DATA WHERE VLANID GREATER THAN 999 FAILS TO BE DOWNLOADED
Description: Updated the agent to be able to receive and process vlan id greater than 999
http://www-01.ibm.com/support/docview.wss?uid=swg1IV50295

APAR : IV50470
-----------------------
Abstract : NETSCREENINTERFACE AGENT CAUSES DISCOVERY TO HANG WHEN A NSIFNAME CONTAINS COLON(:) - E.G. ETHERNET1/2.1:1
Description : Due to the present of both logical unit identifier and channel in the interface
name. e.g. ethernet0/1.1:1 when deciding the upward connection.
https://www.ibm.com/support/docview.wss?uid=swg1IV50470

APAR : IV50718
-----------------------
Abstract: PERFORMANCE ISSUES WITH POLLER WHEN POLLING HIGH NUMBER OF
ENTITIES
Description: When poller is configured to do ICMP polling of around 10k
entities, poller falls behind with multiple messages in
ncp_poller.SnmpPoller.<DOMAIN>.trace: <snip>
2013-10-02 11:12:43 SEVERE xxxxxx.cc(417): About to submit
batch, but there aren't any threads available..
At this point - poller process memory size starts to
increase and eventually it fails with Out-of-memory error.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV50718

APAR : IV50721
-----------------------
Abstract: IN RCA, MAXAGEDIFFERENCE BETWEEN TWO EVENTS SHOULD BE CALCULATED
USING FIRSTOCCURRENCE ONLY, NOT LASTOCCURRENCE TIMESTAMP.
Description: In RCA, the MaxAgeDifference (this attribute is listed in
RCASchema.cfg) between two events should be calculated using
FirstOccurrence timestamps only, not LastOccurrence timestamps.
Events that first occur an hour apart say but keep on failing still get correlated even
though their FirsOccurrence timestamps are greater than 15 minutes apart.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV50701

APAR : IV50786
-----------------------
Abstract: NCP_POLLER STOPS WORKING AFTER NCIM RESTARTS
Description: ncp_poller memory consumption is increasing indefinitely until
it stops working. This can be seen in the ncp_poller debug logs from the memory
usage messages: Debug: D-MOM-001-051: [1t] Generating heartbeat: memory usage:
110731264 bytes...When database was disconnected the Poller would not always
reconnect
http://www-01.ibm.com/support/docview.wss?uid=swg1IV50786

APAR : IV51240
-----------------------
Abstract: ENTITYFILTER IN MODELNCIMDB.CFG ALLOW FOR DUPLICATE MATCHES.
Description: During Topology updates the following errors are
observed in the ncp_model.<domain>.log
Warning: W-RIV-002-206: [4117887888t] xxxxxxx.cc(xxx) A
database 'execute' operation has failed : SQLRETURN = -1
xxxxxxx.xx.cc line xxx : [Informix][Informix ODBC
Driver][Informix]Unique constraint (xxxx.xxxxxxxxx_pk)violated....
Warning: W-RIV-002-128: [4117887888t] xxxxxxxx.cc(xxxx)
xxxxxxxxxxx::RecordToExternalDb: Failed to update
Cause: The default dbModel.entityMap.EntityFilter values in
ModelNcimDb.cfg file allow for duplicate matching. Filters in
this file should not match more than one record.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV51240

APAR : IV51576
-----------------------
Abstract : IFOPERSTATUS OF INTERFACES IS NOT MERGED INTO FINAL TOPOLOGY FOR NETSCREEN DEVICES
Description : ifOperStatus is not updated onto Interface for NetScreen device.
However the data is available at Chassis level for all Interfaces. NetScreenInterface agent is
modified to retrieve ifOperStatus from IF MIB, and update it to respective Interface.
https://www.ibm.com/support/docview.wss?uid=swg1IV51576

APAR : IV52022
-----------------------
Abstract: MODEL TO NCIM BATCH TRANSFER TIME SPIKES OCASIONALLY
Description: Model to NCIM batch transfer time spikes ocasionally to very high values
https://www.ibm.com/support/docview.wss?uid=swg1IV52022

APAR : IV52877
-----------------------
Abstract: DEADLOCK DETECTION OCCURS WHILE CACHING NETWORK VIEWS(MYSQL
ONLY)
Description: Deadlock detections noticed in following scenario:
While loading the networkView Tree for the very first time, the
Views in the tree are being registered for events. Registering
views for the events causes to cache the Views. During caching
process,deadlock detections are noticed.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV52877

APAR : IV53347
-----------------------
Abstract: INCOMPLETE ENTITY AGENT DATA CAN CAUSE A CORE DUMP IN THE AGENT AND IMPACT ON THE COMPLETION OF DISCOVERY.
Description : Core dump happened when enable Entity Agent in discovery. It's triggered by
incomplete data downloaded being downloaded which later causes core dump when
Entity Agent try to process it. Core message will be similar to below:
0x00227bec in xxxxxxx::ResolveInterfaces (this=0x95d54a0, neRecord=0xf71004e0,
neMedData=0xf6e00488, entityDataByPhysIndex=@0x93de0b4) at xxxxxxxx.cc:1279
https://www.ibm.com/support/docview.wss?uid=swg1IV53347

APAR : IV53975
-----------------------
Title : NETWORK MANAGER IP EDITION 2013 INTEGRATIONS R1 (3.9)
Problem : Support added for the following:
(1) IOT Certified Collector discovery support for Alcatel-Lucent SAM 5620 Release 11 using FindToFile method.
(2) IOT Certified Collector discovery support for Alcatel-Lucent IDM 5529 Release 9.2.30
Access Domain using getInventoryIterator method.
(3) Juniper QFabric discovery agent for Juniper physical inventory discovery.
(4) Discovery support for ZTE M6000, T1200 and T600 Physical Inventory.
(5) VLAN discovery support ExtremeSwitch version 12.
(6) HTTPS support for perl Collector Framework and Alcatel5620SamSoap,
Alcatel5620SamSoapFileToFile and Alcatel5529IdmSoap collector.
https://www.ibm.com/support/docview.wss?uid=swg1IV53975

APAR : IV54305
-------------
Abstract: COLLECTORINVENTORY AGENT CRASHED
Description: The CollectorInventory Agent could crash due to complications caused by an
entity existing twice (it tries to delete the data twice).
Two Agents are potentially vulnerable to this bug:CollectorInventory and CollectorVpn.
Prior to the crash it would output the following warning;
http://www-01.ibm.com/support/docview.wss?uid=swg1IV54305

APAR : IV54522
-----------------------
Abstract: INAPPROPRIATE IOCTL CALL CAUSING KERNEL LOG ERRORS (APAR=IV54522)
Description: Thousands of following errors were logged to /var/log/messages
during discovery by ncp_d_helpserv, ncp_disco, ncp_df_*
ncp_agent processes and also by ncp_virtualdomain irrespective of the discovery status.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV54522

APAR : IV54682
-----------------------
Abstract: MISSING STACKED ENTITY INFORMATION ON CISCO DEVICE
Description:Chassis information is missing from containment when Cisco's
device is configured as Stack Type
http://www-01.ibm.com/support/docview.wss?uid=swg1IV54682

APAR : IV55034
-----------------------
Abstract: UPDATE THE VERSION OF GSKIT USED BY ITNM 3.9
Description: ITNM version 3.9 Fp4 upgraded with GSKIT 8.0.14.33.

APAR : IV55230
-----------------------
Abstact: POLLER SHOULD NOT ATTEMPT TO RECREATE A MONITOR ASSOCIATED WITH
AN UPDATED ENTITY WHOSE IP ADDRESS CHANGED TO NULL
Description: Poller should not attempt to recreate a monitor associated with
an updated entity whose IP address changed to NULL. It has been
shown to be possible to receive a managed status change for an
entity with a new IP address which is NULL resulting in the
poller coring while trying to recreate the snmp target. We
should intercept this immediately upon detecting it, ignore the
update but log a message.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV55230

APAR : IV55465
-----------------------
Abstract: ITNM 3.9 NETWORK DEVICE CONFIGURATION ERRORS DISPLAYS DATA FROM OTHER DOMAINS
Description: When running the Network Device Configuration Errors report in
an environment where there are multiple domains, the section "Connected Devices with Duplex Mismatches"
will combine data from any domain stored in the NCIM database.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV55465

APAR : IV55657
-----------------------
Abstract: EVENT FILTERED VIEWS ARE MISSING THE OPERAND WHEN MORE THAN ONE FILTER IS SPECIFIED
Description: Using multiple filters produces inconsistent results for Maximum Severity.
http://www-01.ibm.com/support/docview.wss?uid=swg1IV55657

APAR : IV56487
-------------
Title : ALCATEL 5529 COLLECTOR TIMEOUT ERROR (APAR=IV56487)
Problem : The Alcatel5529IdmSoapCollector fails with error:
500 Server closed connection without sending any data back at
/opt/IBM/tivoli/netcool/precision/collectors/perlCollectors/Alcatel5529IdmSoap/Alcatel55
29IdmSoapCollector.pm line 225
Due to this timeout, the collector exits and discard all other discovered devices.
http://www.ibm.com/support/docview.wss?uid=swg1IV56487

APAR: IV56672
-------------
Abstract: CREATETRUNKCONNECTIONS STITCHER COULD GENERATE DISCO CORE WHEN DEALING WITH CORRUPTED DATA
Description:Due to failed SNMP or wrong device configuration, the NortelSwitch.retuns table can generate Trunked Vlan information as a 0 lenght list.
http://www.ibm.com/support/docview.wss?uid=swg1IV56672

APAR : IV56717
-----------------------
Abstract: ADDED AOC DEVICE CLASSIFICATIONS IN ITNM FOR THE ARISTA MANUFACTURE AND A SUB-CLASS FOR THE ARISTA VEOS.
Problem: Added new AOC device classifications in ITNM for the Arista manufacture and a
sub-class for the Arista vEOS. Support for those devices once discovered with standard
discovery agents turned on.
http://www.ibm.com/support/docview.wss?uid=swg1IV56717

APAR : IV59327
---------------------
Abstract: DB2 PROVIDES A NUMBER OF STRATEGIES FOR IMPROVING THE AVAILABILITY OF YOUR DATABASE SOLUTION, INCLUDING A FEATURE DB2 (HADR)
Problem: Network Manager allows you to configure the NCIM topology database for high availability,minimizing the impact of computer or network failure.
http://www.ibm.com/support/docview.wss?uid=swg1IV59327

============================================================================