Date: April 1, 2005
Name: 7.1.4-TIV-NVD-{OperatingSystem}-FP0003
Component: IBM Tivoli NetView Version 7.1.4
PTF Number: U802492 (UNIX) or XR23263 (Windows)
Note: Before using this information and the product it supports, read the information in the Notices section at the end of this document.
This section provides general information about this fix pack. It is strongly recommended that you read this entire document before you install this fix pack.
This fix pack consists of five packages, one each for RS6000 AIX(R), SUN SPARC Solaris, Linux Intel(TM), Linux for S/390 (31 bit), and Microsoft(R) Windows(R). You need to use the appropriate package on each platform on which an IBM(R) Tivoli(R) NetView(R) native server or client has been installed.
For Red Hat Enterprise Linux Advance Server 3.0 and SUSE Linux Enterprise Server (SLES) 9 on S/390 (31 bit) systems, a new CD has been added to the base NetView 7.1.4 kit and a standard IBM Tivoli NetView install is required.
For Red Hat Enterprise Linux AS Version 3.0 and SUSE Linux Enterprise Server (SLES) 9 on zSeries (64 bit) systems, a new CD has been added to the base NetView 7.1.4 kit and a standard IBM Tivoli NetView install is required.
This fix pack also contains a set of corrections for the Tivoli NetView Warehouse Enablement Pack used by and contained within the Tivoli Data Warehouse. To apply the corrections to the Tivoli NetView Warehouse Enablement Pack code within the Tivoli Data Warehouse, refer to the 1.1.0.4-TIV-NVD-WEP-FP0003.readme file that is associated with this fix pack.
None.
Because the IBM Tivoli NetView program does not produce patches, no patches are superseded by this fix pack. However, all test fixes for the IBM Tivoli NetView 7.1.4 program that are covered in the APARs corrected in this fix pack section of this readme are superseded.
Please note that the following Windows operating systems are not supported in this release:
The following items are new in IBM Tivoli NetView 7.1.4 Fix Pack 03.
The fix for PJ30038 requires that the user stop recording data into the previous mdb files in the \usr\ov\databases\odbc directory and create new database files using the empty templates that are provided. To do this conversion, perform the following steps:
You can keep old database files for analysis using another tool, such as MS Access, or you can delete them. However, after the above conversion is done, the Tivoli NetView program ignores the old data.
The IBM Tivoli Switch Analyzer program requires the Tivoli NetView program to have a fully discovered path from the NetView server to each managed switch. Switches outside this scope are in what is called remote campuses. The IBM Tivoli Switch Analyzer program does not perform connection discovery on these switches. A typical situation is a link across an unmanaged wide area network (WAN), such as an ISP backbone.
You can now manually create custom links between routers, similar to unnumbered serial links.
To specify and create a custom link, perform the following steps:
where the ifIndex is specified by the '0'.
For example, using an existing IP address on each router, add a statement that is similar to the following example:
To create links from one edge router to several other routers, you must choose a different IP address for each link in the edge router.
The following list describes how a custom link appears in the map:
When IP Map is configured to display interface labels as IP addresses, the default interface label is used for interfaces whose IP address is 0.0.0.0. Formerly, the IP address of 0.0.0.0 was displayed.
The following items were new in IBM Tivoli NetView 7.1.4 Fix Pack 02.
For Red Hat Enterprise Linux AS 3.0 and SUSE Linux Enterprise Server (SLES) 9, if the firewall is turned on, SNMP traps do not pass through. Moreover, the firewall is turned on by default. Therefore, for IBM Tivoli NetView to receive traps from external agents, the firewall must be turned off.
IBM Tivoli NetView provides support for monitoring the Cisco PIX Firewall Failover conditions. NetView can monitor the failover state of the PIX devices during normal status polling. When a failover occurs, a new event is generated announcing whether the active addresses are on the primary or secondary device. In addition, the operator can see on the map when the active addresses are on the secondary device, which indicates that a failover has occurred and action should be taken to prevent a further failure that disables the firewalls. The management interface shows USER2 status, which is usually purple (by default). This propagates to change the PIX device symbol to marginal. When the active addresses are returned to the primary device, an event announces this fact and the status on the map returns to Normal.
Use the netmon.seed file to configure both the PIX Firewall Failover machines and to set and lock the SNMP address.
For more information on the PIX Firewall Failover support see the /usr/OV/doc/PixFailoverReadme.pdf file.
A NetView administrator can now set and lock the SNMP address for any device.
For more information, see the A modified netmon.seed file section.
The netmon daemon now recognizes two new tokens in the seed file.
PIX Firewalls typically swap interfaces between the primary and secondary units on failover. In addition, SNMP access is generally restricted, but access using one address is required. The Tivoli NetView program monitors the PIX machines for failover. For more information, refer to the document, /usr/OV/docs/PixFailoverReadme.pdf.
Use the prefix > followed by the IP address of each machine that is accessible for SNMP. Wildcards and ranges are permitted, but host names are not. Here is an example:
>172.30.100.2
>172.30.100.3
This token automatically enables netmon to poll the Cisco MIB for failover, sets and locks the SNMP address, and enables SNMP status polling for this node.
By default, the Tivoli NetView program attempts to change the SNMP address for a node when that interface is known to be down. However, it would be useful to override this behavior for certain nodes.
Use the prefix = followed by the IP address that should be set and locked for SNMP queries. Wildcards and ranges are permitted, but host names are not.
This token is used to set the SNMP address and lock it so that the Tivoli NetView program does not attempt to change the SNMP address.
When the Tivoli Switch Analyzer program is installed with the Tivoli NetView program, it correlates outages to a layer 2 root cause. When the root cause is a switch, the Tivoli NetView program changes the icon color for the switch to yellow, to indicate a layer 2 problem with the ports. The new Layer2Status object provides a visualization of the layer 2 status by representing it as a new interface object under the switch. The default label for this new object is Layer2Status, and the selection name is <switchhostname>:Layer2Status. The status of the switch icon is based on the standard propagation schemes of the IP interface and the new Layer2Status object, with the caveat that the IP status takes precedence when it is more severe.
The Tivoli Switch Analyzer program is responsible for setting the Layer2Status ovwdb field on the switch when there is a correlated problem, and the Tivoli Switch Analyzer program resets the status when the problem is resolved. The status of the new Layer2Status object reflects the value in the Layer2Status field. When this status changes to Marginal, the netmon daemon issues an Interface Layer2Status marginal event. This event has the same specific ID as the Interface Down event. When the layer 2 status returns to Normal, the standard interface event, Interface UP, is sent for the Layer2Status object.
The existing Layer2Status events for the switch node continue to be sent and are used by ipmap. The new Layer2Status object fields can be seen using the ovobjprint command.
The following messages are displayed in the event browser when new Layer2Status objects are created.
Note: The <switch name> message is displayed for each switch.
The ovtopodump -r command has been modified to list each interface for a given node only once. For a node with more than one interface on the same network, it used to print all the interfaces for every node.
The following changes have been implemented to provide better management of user accounts.
Previously, the only way to change user passwords was by running the Web Console Security application on the NetView server or by having a NetView administrator edit the /usr/OV/www/conf/NetViewRealm.properties file with a text editor.
New files have been created to support password expiration.
Here are instructions for using the new Web Console Security features.
As part of the IBM Tivoli NetView Fix Pack 02 installation, the NetViewRealm.xml file is created. Initially, the NetViewRealm.xml file is set up so that passwords never expire.
<NetViewRealm validFor="-1"/>
When the value of the validFor= attribute is -1, NetViewRealm should be an empty element.
To re-enable the change password feature when the current value for the validFor= attribute is -1, a NetView administrator needs to either:
Note:After running the generateNetViewRealmXML command, all the user accounts are set to expire at the same time based on the validFor= attribute. Users are warned in advance to change their passwords. If users do not change their passwords, their user accounts are deleted. See the Logging on after the password has expired section for more details.
The NetViewRealm.xml file contains the following attributes related to user account password expiration:
Here is an example of an NetViewRealm.xml file with user accounts.
<NetViewRealm validFor="186" expireWarning="21" dateFormat="yyyyMMdd"> <Account name="jane" since="20040701"/> <Account name="john" since="20040701"/> <Account name="maria" since="20040701"/> " ... </NetViewRealm>
To add new Web Console user accounts and keep the NetViewRealm files synchronized, use one of the following methods:
This method picks up the current general password expiration settings from the NetViewRealm.xml file for the new users. For details on how to modify the general password expiration settings, see the Editing NetViewRealm.xml section.
When NetView administrators need to add a large number of users, manually editing the NetViewRealm.properties is easier than using the Web Console Security application.
After updating the NetViewRealm.properties file, instead of manually adding corresponding updates to the NetViewRealm.xml file, an administrator can just add (or remove) Web Console Security user account information to the NetViewRealm.properties file and then run the generateNetViewRealmXML command to update the NetViewRealm.xml file.
Note: After adding users to the NetViewRealm.properties file, the NetView administrator should start the Web Console Security application and specify to save updates. As before, saving updates encrypts the passwords of any users which were manually added.
Note: Always recycle the Web Server after adding or deleting Web Console user accounts.
A Change Password (disabled) button has been added to the NetView Web Console logon screen.
After a user enters the username and password in the Web Console logon screen, the user can click Change Password.
An authentication check request is sent to server. The cursor changes to an hour-glass until a response is received.
If authentication succeeds, the logon dialog box closes. Then the user should follow the steps under the Changing the Password section.
If the passwords do not match, the following error message is displayed:
The passwords do not match.
Both password fields are cleared, and focus moves to the first password field. The user should try again.
In either case, focus moves to the OK button.
The password change request is sent to the Web Server. The cursor changes to an hour-glass until a response is received. If the password change succeeds, the logon dialog box is closed and the Web Console is started.
A user's new password cannot be the same as the old password.
Possible errors:
A dialog box is displayed with the following message:
Password file is busy. Try again?
If the user presses the Enter key or clicks Yes, the dialog box closes. The user can repeat the password change request.
If the user clicks No, the dialog box is closed. The logon dialog box is also closed, and the Web Console is started.
A dialog box is displayed with the following message:
Realm property file access time-out. Please try again later.
Only the OK button is enabled. If the user presses the Enter key or clicks OK, the dialog box closes. Repeat the change password step to resend the password change request. If the user gets this error message after several tries, the administrator might be using the Web Console Security application. Web Console Security locks the password files, and the user cannot change the password while the files are locked. The user should press the button to close the Change Password dialog and start the Web Console. The user should contact the Administrator about the access timeout.
Immediately after successfully logging on and before the logon dialog box is closed, the user is notified that the password is about to expire and is prompted to change it. The new password takes effect immediately.
If a user's password has expired, the user cannot log on to the Web Console and must contact the NetView administrator for a new password.
The NetView administrator must use Web Console Security to reset the user's password or restore the user's account if the daily task which removes accounts with expired passwords (from NetViewRealm.properties) has removed the account, or both.
This daily task removes accounts with expired passwords and logs information about the removed accounts in the ${jetty.home}/logs/expiredpasswords.log file. A NetView administrator can use this information to reinstate the accounts.
A NetView administrator can manually edit the NetViewRealm.xml file to update the various attributes related to password expiration.
When password expiration has been enabled (see the Initializing Web Console Security settings section), a NetView administrator can disable it again by setting the value of the validFor= attribute in NetViewRealm.xml to -1 and running the generateNetViewRealmXML command. In this case, the generateNetViewRealmXML command deletes all the Account name entries leaving NetViewRealm as an empty element.
A NetView administrator can change the length of time passwords are valid by changing the value of the validFor= attribute in the NetViewRealm.xml file.
To change when users are notified that their passwords are about to expire, change the value of the expireWarning= attribute. The default value is 21 days. When a user logs on within 21 days of the expiration date for the user's password, a dialog box is displayed that contains the following message:
Your password will expire in 21 [20, ...] days. Do you want to change it now?
The NetView administrator should only change the values of the expireWarning= and validFor= attributes. Administrators should not change the Account elements.
<NetViewRealm validFor="30" expireWarning="7" dateFormat="yyyyMMdd"> <Account name="jane" since="20040701"/> <Account name="john" since="20040628"/> <Account name="maria" since="20040627"/> ... </NetViewRealm>
In this example, the maximum password age is 30 days. If today is July 28, 2004, Jane must change her password on or before July 31. When Jane signs on to the Web Console today, she is told that her password will expire in 3 days and is asked if she wants to change it. John must change his password today. When John signs on to the Web Console today, he is told that his password will expire today and asked if he wants to change it. It is possible that his password might expire while he is using the Web Console. Since Maria's password has already expired, she has to get a NetView administrator to change her password for her. Maria's account will probably already have been removed from the file by the daily expired password removal task.
Note: The since= attribute is the date the password was last changed. The administrator can change the value of the validFor= attribute without needing to recalculate the since= attribute values.
Note: After editing the NetViewRealm files, run the generateNetViewRealmXML command and recycle the Web Server.
The global Acknowledge/Unacknowledge function in the IBM Tivoli NetView for UNIX program now works for user-created non-IP map objects.
On both UNIX and Windows, it is now possible to acknowledge map objects whose object status is Normal.
The trapd daemon on Windows can now receive external SNMPv2 traps. Incoming SNMPv2 traps are translated to their SNMPv1 equivalent. The trapd daemon also acknowledges SNMPV2 INFORMs.
To receive SNMPv2 traps, the trapd daemon must be reconfigured to listen for traps directly on the incoming trap port (usually 162). It cannot receive SNMPv2 traps through the Microsoft(R) Windows Trap Service. To configure trapd to listen for traps on port 162, follow these steps:
> ovstop > nvservice -stop
HKEY_LOCAL_MACHINE\SOFTWARE\Tivoli\NetView\CurrentVersion\trapdSharePort162
HKEY_LOCAL_MACHINE\SOFTWARE\Tivoli\NetView\CurrentVersion\trapdTrapReceptionPort
> net stop snmptrap
> nvservice -remove > nvservice -install -username .\NetView -password <password>
Note that incoming SNMPv2 traps that contain 64-bit values are dropped from the translated SNMPv1 traps. If tracing is enabled for the trapd daemon, the dropped value is logged to the trace file.
The nvmaputil manage-interface and nvmaputil unmanage-interface subcommands now accept either an IP address or a selection name. If the IP address is used and it is either not found or there are multiple objects with the same IP address, the nvmaputil command exits and prompts the user to enter the selection name.
Note: AutoTrace exists only on UNIX.
The new /usr/OV/service/nvAT script controls the function of the AutoTrace subsystem without requiring a system reboot. This script incorporates the steps outlined in previous release notes for controlling AutoTrace.
Usage statement:
Usage: nvAT [start|status|stop]
nvAT [/h | /?]
start : Initialize the AutoTrace subsystem and start the NetView channel
status : Display the status of the AutoTrace subsystem
stop : Stop the NetView channel and the AutoTrace subsystem
/h : Help
/? : Display the usage statement
The modified Web Console Switch Management Views show the VLAN on which every port is located. This is done through community string indexing and entails using the community string <community name>@<port number> where <community name> is the community string for the device and <port number> is the VLAN port number from the VLAN table for the switch.
These changes apply only to switches that support the Cisco BRIDGE_MIB, in effect, Cisco Catalyst switches. Other switches are displayed without VLAN information.
Cisco lists the following as providing support:
See: http://www.cisco.com/en/US/tech/tk648/tk362/technologies_configuration_example09186a008015773e.shtml
Displaying VLAN information:
Limitations in this implementation:
When configuring a NetView client using the Add/Change Server option in the Server Setup tool, an ICMP echo request (ping) is used to test the connection to the proposed NetView server. In past releases, if the ping did not give a valid response, then the Tivoli NetView program assumed that the chosen NetView server machine was either not valid or unreachable, and it did not allow the client to be configured using that server. Now, if the ping fails, the Server Setup tool displays a warning and continues configuration.
In the Server Setup tool, the option Maintain -> Clear databases -> Clear object/topology/map databases, remove customizations now re-creates the default SmartSets. In previous releases, the default SmartSets were removed by this option.
The new /usr/OV/service/nvPerms script restricts the permissions on the Tivoli NetView files and directories to tighten security.
The nvPerms script resets the permissions on the Tivoli NetView files and directories so that:
Note: The nvPerms script cannot be used in the NetView native client/server environment.
APAR IY53703 introduces a new environment variable for use by the trapd daemon to help the user better control the formatting of variables that contain hexadecimal data in strings. Currently, the trapd daemon tries to interpret this data as printable to make the output in the trapd.log file and the event windows as readable as possible, unless more than half the string contains unprintable characters. However, sometimes users know that the entire string should be printed as hexadecimal data, such as when it is really a MAC address. Users who want to have the trapd daemon more strictly interpret hexadecimal data and dump an entire string to hexadecimal unless all of it is printable, should set the TRAPD_STRICT_HEX_FORMAT environment variable.
On UNIX, create a /usr/OV/bin/netnmrc.pre file with the following line, or add the following line to the existing usr/OV/bin/netnmrc.pre file:
export TRAPD_STRICT_HEX_FORMAT=TRUE
Restart all the daemons so the daemons inherit this variable. Either reboot the NetView machine or use the ovstop nvsecd command to stop all the daemons. Then use the /etc/netnmrc command (AIX) or the /etc/init.d/netnmrc command (Solaris or Linux) to enable this change.
On Windows, follow these steps to set the environment variable using the control panel:
Control Panel --> System Environment --> Advanced --> Environment Variables System --> Add Variable Name Value TRAPD_STRICT_HEX_FORMAT TRUE
Reboot the NetView machine to enable the change.
Due to APAR IY56563, a new option, "Add collected data to the NetView Warehouse?," was added to the daemon options for the UNIX snmpCollect daemon in serversetup.
This new option is set to yes, by default, which causes the snmpCollect daemon to start with the -w option, which enables adding collected data to the NetView Warehouse. Setting this option to no disables adding collected data to the NetView Warehouse.
Note: This new option does not show up in the UNIX smit path to setting options for the snmpCollect daemon in this fix pack, only in the serversetup path. The workaround is to use serversetup instead of smit.
As part of APAR IY58514, the following two snmpCollect options were added:
Disconnects the snmpCollect daemon from the NetView Tivoli Data Warehouse database, but allows the snmpCollect daemon to continue to run. This option is not available on Linux on zSeries.
Causes the snmpCollect daemon to connect to the NetView Tivoli Data Warehouse database without having to recycle the snmpCollect daemon. This option is not available on Linux on zSeries.
A new trapd.conf file is shipped with this fix pack for UNIX. Changes to the trapd.conf file fix problems related to APAR IY57432 with sending interface lists to the Tivoli Enterprise Console on events, as well problems on Linux related to APAR IY52871 with launching the MIB browser from the Events Window on Linux.
If the trapd.conf file in the /usr/OV/conf/C directory (not the obsolete trapd.conf file in /usr/OV/conf) has not been customized, replace it with the new trapd.conf file from /usr/OV/newconfig/OVSNMP-RUN directory. If the original trapd.conf file has been modified, manually add the changes into the original file. For details, see the sections on Sending interface lists to the Tivoli Enterprise Console on events and Changes to trapd.conf for launching the MIB browser sections.
Note: See the New in Fix Pack 01 section for information on the trapd.conf file for Windows.
As mentioned in the A new trapd.conf file section, the new trapd.conf file that is shipped with this fix pack fixes the problem described by APAR IY52871 with launching the MIB browser from the Events Window on Linux.
If the original trapd.conf file has been modified, manually add the changes in the new trapd.conf into the original trapd.conf file. Otherwise, simply use the new trapd.conf.
The suggested method is to use the xnmtrap application to edit the trapd.conf file because doing so causes the trapd daemon to automatically reload it when you are finished. You must make the edit as root.
/usr/OV/bin/xnmbrowser -node "`awk '{if (NR==1)print $10}'`
to this:
/usr/OV/bin/xnmbrowser -node "`awk '{if (NR==1)print $10}'` "
You can also make this change using another editor. Locate the following line in the trapd.conf file:
ACTION 0 "Browse MIB" /usr/OV/bin/xnmbrowser -node "`awk '{if (NR==1)print $10}'`
and change it to this:
ACTION 0 "Browse MIB" /usr/OV/bin/xnmbrowser -node "`awk '{if (NR==1)print $10}'` "
Save the file and either restart the trapd daemon or run the /usr/OV/bin/event -e FMTCHG command to cause the trapd daemon to re-read the file.
APARs IY57432 (UNIX) PJ29821 (Windows) deal with certain events failing to parse correctly when sent to the Tivoli Enterprise Console(R) server. This is because some nodes and routers have so many interfaces (in excess of 30) in some environments that the interface list, sent as varbind 8, is too large for the Tivoli Enterprise Console server to parse correctly. This problem is addressed by no longer sending the interface list to the Tivoli Enterprise Console server as part of those events.
In the Tivoli NetView for UNIX program, this is controlled by the slot map entries in the trapd.conf file. The fix is provided in a new trapd.conf file in the /usr/OV/newconfig/OVSNMP-RUN directory. This file does not automatically replace the /usr/OV/conf/C/trapd.conf file that is installed. If you have not customized the /usr/OV/conf/C/trapd.conf file, replace it with the new trapd.conf file that is provided. If you have customized the /usr/OV/conf/C/trapd.conf, follow these steps to manually add the entries to the customized trapd.conf file:
Using the xnmtrap application, modify each of the following entries in the trapd.conf file.
IBM_NVNM_EV 50790400 (Node Marginal)
IBM_NVNUP_EV 58916864 (Node Up)
IBM_NVNDWN_EV 58916865 (Node Down)
ROUTDOWN_EV 58916971 (Router Down)
ROUTUNREACH_EV 58916972 (Router Unreachable)
ROUTERUP_EV 58916973 (Router Up)
ROUTMARG_EV 58916974 (Router Marginal)
Follow these steps for each entry:
In the Tivoli NetView for Windows program, this is controlled by the entries in the \usr\ov\conf\tecad_nv6k.cds file. The fix is provided in a new tecad_nv6k.cds file in \usr\ov\newconfig\tecad directory. This file does not automatically replace the \usr\ov\conf\tecad_nv6k.cds file that is installed. If you have not customized the \usr\ov\conf\tecad_nv6k.cds file, replace it with the new tecad_nv6k.cds file that is provided. If you have customized the \usr\ov\conf\tecad_nv6k.cds file, follow these steps to manually add the entries to the customized tecad_nv6k.cds file:
6: ATTR(=, "VB_8") ;
iflist = $V6 ;
The following example shows the Node Up entry as shipped.
CLASS TEC_ITS_NODE_STATUS
SELECT
1: ATTR(=,$ENTERPRISE) , VALUE(PREFIX, "1.3.6.1.4.1.2.6.3" ) ;
2: $SPECIFIC = 58916864 ;
3: ATTR(=, "nvObject" ) ;
4: ATTR(=, "nvEventDescr" ) ;
5: ATTR(=, "nvApplNbr" ) ;
6: ATTR(=, "VB_8") ;
FETCH
1: IPADDR($V3);
MAP
origin = $F1 ;
hostname = $V3 ;
msg = $V4 ;
category = $V5 ;
nodestatus = 1 ; # UP
iflist = $V6 ;
nvhostname = $ADAPTER_IP ; # Required for ALL TEC_ITS events
END
The following example shows the modified Node Up entry
CLASS TEC_ITS_NODE_STATUS
SELECT
1: ATTR(=,$ENTERPRISE) , VALUE(PREFIX, "1.3.6.1.4.1.2.6.3" ) ;
2: $SPECIFIC = 58916864 ;
3: ATTR(=, "nvObject" ) ;
4: ATTR(=, "nvEventDescr" ) ;
5: ATTR(=, "nvApplNbr" ) ;
FETCH
1: IPADDR($V3);
MAP
origin = $F1 ;
hostname = $V3 ;
msg = $V4 ;
category = $V5 ;
nodestatus = 1 ; # UP
nvhostname = $ADAPTER_IP ; # Required for ALL TEC_ITS events
END
Tivoli NetView for Windows users can implement the new overridestatus_user1.rs rule set, which replaces the broken setstatusseverity.rs rule set. The \usr\ov\doc\Rulesets_on_Windows.readme document describes the rule set samples for use with the nvcord correlation daemon. Delete the setstatusseverity.rs rule set. It will not be shipped in future releases.
The following new options have been added to the xnmsnmpconf command. Note: The characters in brackets ([]) are optional.
Removes all entries from the SNMP configuration nodedb files (nodedb.dir and nodedb.pag) that contain entries for specific nodes.
Removes all entries from the SNMP configuration wilddb files (wilddb.dir and wilddb.pag) that contain entries for IP address wildcards.
Removes all entries from the SNMP configuration colldb files (colldb.dir and colldb.pag) that contain entries for SmartSets.
Note: The SNMP configuration files are in the /usr/OV/conf/ovsnmp.conf_db directory.
Using these new options and the xnmsnmpconf -clearCache command, you can entirely clear out the SNMP configuration database by following these steps:
xnmsnmpconf -clearnode xnmsnmpconf -clearwilddb xnmsnmpconf -clearcolldb xnmsnmpconf -clearCache
Note: Prior to clearing out the SNMP configuration database, save and make note of any xnmsnmpconf customizations (such as, the default community name and IP address wildcards.) After clearing out the SNMP configuration database, restore these customizations using either the xnmsnmpconf -import command or the xnmsnmpconf GUI.
In conjunction with APARs IY56279 and IY60528, three new options have been added to the tecint.conf file on UNIX to control the behavior of the internal Tivoli Enterprise Console adapter in the nvserverd daemon (for both Tivoli and non-Tivoli environments). By default, these options are commented out, but the user can uncomment them by removing the pound sign (#) that precedes the entries to change the behavior of the nvserverd daemon as described below. A new man page for the tecint.conf file has been provided to document these options.
When this option is set to Yes or YES, this flag causes the nvserverd daemon to create a log of events being sent to the Tivoli Enterprise Console server in the /usr/OV/log/nvserverd.log file. The events are logged prior to being processed by the event adapter. The log also records the startup parameters of the internal adapter itself as well as error messages concerning the internal adapter should any be recorded. This error information was previously available only in the nettl logs. Using this flag in the tecint.conf file removes the need to export the NVSERVERD_TRACE_TEC_EVENTS=YES environment variable in the /usr/OV/bin/netnmrc.pre file to control the logging function. Moreover, it overrides that variable if present.
If the NvserverdTraceTecEvents= flag is not present in the tecint.conf file or if it is commented out, the default behavior is that no nvserverd.log file is created.
When this option is set to No or NO, this flag causes the nvserverd daemon to open a session with the nvcorrd correlation daemon, requesting not to be primed with recent events, as event windows are. The default behavior is that the Tivoli Enterprise Console rule set is primed. Priming it increases the probability that duplicate events are sent to the Tivoli Enterprise Console server when the nvserverd daemon is stopped and restarted while the nvcorrd daemon remains active. Using this entry removes the need to export the NVSERVERD_PRIME_TEC_EVENTS=NO environment variable in the /usr/OV/bin/netnmrc.pre file to control the priming function. Moreover, it overrides that variable if present.
When this option is set to Yes or YES, this flag causes the nvserverd daemon to revert to its behavior priot to version 7.1.4 and to send the user-configurable Tivoli NetView severity on events sent to the Tivoli Enterprise Console server. Event severity was initially dropped in version 7.1.4 because it interfered with the working of new Tivoli Enterprise Console rules that are shipped in IBM Tivoli Enterprise Console Version 3.9 program (and in fix packs for versions 3.8 and 3.7.1). These new Tivoli Enterprise Console rules automatically set the event severity of Tivoli NetView internal events upon receipt at the Tivoli Enterprise Console server, and escalated de-escalated the severity based on event correlation and time open at the Tivoli Enterprise Console server. However, many users are not ready to migrate to using these new rules and require that Tivoli NetView severity still be sent for their own custom rules to work. The new NvserverdSendSeverityTecEvents= flag is, therefore, provided as a migration aid. Using this option removes the need to export the NVSERVERD_SEND_SEVERITY_TEC_EVENTS=YES environment variable in the /usr/OV/bin/netnmrc.pre file to control the sending of severity. Moreover, it overrides that variable if present. If the NvserverdSendSeverityTecEvents= flag is not present in the tecint.conf file, or if it is commented out, the default behavior is that no severity field is sent on default events. Explicitly coded severity in a slotmap field is not affected by these changes.
The following items were new in IBM Tivoli NetView 7.1.4 Fix Pack 01.
APAR IY48379 described a problem where the ovstop command does not always stop the tdwdaemon. A modified tdwdaemon.lrf file is provided to correct the problem. To install this correction on a UNIX platform, replace the tdwdaemon.lrf file in the installation using the following steps:
To back this correction out, follow the above directions, replacing the two copy steps with a single copy of the saved lrf file back to its original location.
The location sensitive topology functionality (the location.conf file) has been enhanced to allow IP address ranges and wildcards in gateway entries. Ranges or wildcards can be used in any part of the gateway IP address. Wildcards cannot be used in a gateway name.
The following change (which should be manually applied to the comment section of the location.conf file) describes the changes to the gateway name or address field:
The format of a gateway entry is:
<location name> <gateway name or address>
Where location name is the name of the location under which this gateway should be placed, and gateway name or address is the name of the gateway or the IP address of one of its interfaces in dotted-decimal notation. Wildcards and ranges are allowed in gateway IP addresses (meaning, 5.25-30.2.8 or 3.5.*.*). Wildcards are not allowed in gateway names (myrouter.company.com).
Here are some examples of gateway entries:,/p>
If a gateway matches more than one gateway entry, one entry is arbitrarily chosen. To help in resolving overlapping gateway entries, warning messages about gateway interfaces that match more than one entry are displayed in the location.log file. These are only warning messages. They do not indicate any problem with gateway or router placement.
A modified netmon.seed file for APAR IY47145 is shipped with this fix pack. This file documents that to explicitly discover a node and make it SNMP-managed there must be two entries in the netmon.seed file: one to discover the node and one for SNMP management. This fix pack places the modified netmon.seed file in the /usr/OV/newconfig/IBMVA-RUN/ subdirectory on UNIX and the \usr\ov\newconfig\netmon subdirectory on Windows.
By default, up to seven alternate community names can be specified in communityNames.conf file. You can modify the maximum number of alternative community names allowed by setting the MaximumCommunityNames environment variable in the netmon.conf file to a value between 0 and 100.
A new trapd.conf file is shipped with this fix pack. This trapd.conf file no longer accidentally truncates several values in the HPCIFxxx traps in the trapd.conf file for the Tivoli NetView for Windows program. This file is located in \usr\ov\newconfig\ovsnmp-run\trapd.conf cirectory. If you want to use this file and you have not customized the \usr\ov\conf\trapd.conf file, replace the trapd.conf file in \usr\ov\conf\trapd.conf directory. If the \usr\ov\conf\trapd.conf file has been modified, you must merge the two files.
If you are having trouble finding services when using the servmon daemon to do ITM queries, you might need to enable secondary IP address retrieval for servmon initialization.
To enable secondary IP address retrieval for the servmon daemon during servmon initialization, add a line in the servmon.conf file that consists of just SECONDARY_IPADDRS. By default, secondary IP addresses are not retrieved during servmon initialization.
However, enabling this function is very performance intensive. It would be best to allow the Tivoli NetView program to stabilize before making the change. Then, when the Tivoli NetView program is running, follow these steps:
Since version 5.1 of the Tivoli NetView for UNIX program, the Action node rule set added three additional variables to the end of every trap passed through it. The variables,in order, are:
In version 6.0.3 of the Tivoli NetView program, the new NVATTR_COUNT attribute was added to tell the user how many variables the trap now contains, and it permitted easier reference of the these last three variables:
These additional varbinds are now exported into the following easy-to-reference variables:
mib2trap mib_file addtrap_output_file [baroc_file] [base_event_class_name]
If you want to continue using the old Nvserverd_Event base class, you can do so by specifying it as the fourth (optional) parameter, in the following way because the command now supports any user-defined base event class:
mib2trap mib_file addtrap_output_file baroc_file Nvserverd_Event
The origin field of an event sent to the Tivoli Enterprise Console server by the internal Tivoli Enterprise Console adapter in nvserverd always contains the IP address of the NetView server. The adapter_host field contains either the fully qualified domain name of the NetView server or the same IP address as the origin field. Short names can no longer be used for the adapter host field to facilitate event responses and forwarding at the Tivoli Enterprise Console server.
The Java(TM) MIB Loader no longer puts MIBs that failed to load into the list of loaded MIBs. It also now warns you if loading a MIB is skipped and reminds you that the only valid extensions for MIB files are .mib, .my, and .def for the Java MIB loader.
The /etc/snmp/conf/mgragentd.rsrc file installed on Solaris has been changed so that the command field reads just 'command = "" ' (a null command entry) from its former value, 'command = /usr/bin/echo mgragentd registered with snmpdx'. The purpose of this change is to avoid any possibility of a defunct process being created when snmpdx is restarted. However, the file is installed only once, during the initial installation of NetView, and the current copy on your system is not updated by this fix pack. Therefore, if you want to have this change before the next release, you must manually insert this change into the /etc/snmp/conf/mgragentd.rsrc file.
As of version 7.1.2 of the Tivoli NetView program, the Tivoli Enterprise Console class definitions used by the Tivoli NetView program were upgraded. The previous Tivoli Enterprise Console event class definitions used by the Tivoli NetView program are no longer supported, and you are expected to use the new class definitions because future migration will not support the old class definitions.
Depending on your system configuration and operating system level, you might need to upgrade to a newer version of the JRE when a Tivoli NetView version 7.1.4 fix pack is installed. If the NetView Web Console causes a core dump after a fix pack has been installed, you can request the JRE upgrade from IBM Customer Support by asking for fix IY45265 for the appropriate platforms and installing it on your systems.
If the snmpCollect daemon is deferring more collections than seem appropriate, the problem might be because the default wait time of 3 seconds is not enough time for the snmpCollect daemon to receive the reply.
To allow more time for collections, follow these steps:
When Tivoli NetView security is enabled, the fully qualified host name of the Tivoli NetView server must be the first host name for the entry for the server in the /etc/hosts file for the server. Otherwise, the credentials of the Tivoli NetView client cannot be verified, and when a message is sent from the server in the Security Administration dialog, Error(35) is reported.
When using a Tivoli NetView native server/client configuration on UNIX, the native NetView client must be stopped and restarted if the platform on which the Tivoli NetView server resides is rebooted. The reboot and subsequent restart of the Tivoli NetView server causes the loss of session information regarding the native client, and the client must be stopped and restarted to restore that information.
The wildcard range in the seed file limits the discovery to be in the range for all managed networks. It does not force the netmon daemon to discover any new node in that range, and it does not force the netmon daemon to create a new network. However, for the seeded entry (single IP or host name), the netmon daemon always attempts to discover the entry and to create a network if it is needed.
To enforce the discovery of an "island-network" (that is, a network with no router connection to the Tivoli NetView server), you must add one explicit (seeded) node entry in that network into the seed file to force the node discovery and network creation. In addition, to expand discovery on other devices in that network, you must ensure the seeded entry is SNMP enabled.
In the SNMP Configuration Panel of the xnmsnmpconf application, the Retry Count field indicates the total attempts on a ping or SNMP request. The count includes the initial poll, so a value of 1 indicates one attempt should be made, a value of "2" indicates an initial attempt and one retry if the attempt fails, and so forth.
If the daily Web Server log files are taking up too much space on your system, in addition to lowering the number of days that these request logs are retained (as described in detail in the the IBM Tivoli NetView 7.1.4 for UNIX Release Notes), you can also take the following actions:
The query interval determines how often Web Consoles send a request to check whether the Web server is still active.
Normally, you do not need customize this parameter. However, if the daily servlet request logs (yyyy_mm_dd.request.log in the /usr/OV/www/logs directory for UNIX or the \usr\ov\www\logs directory for Windows) are routinely much too large, you can increase the query interval so that fewer EventChannel query log entries are created. However, this change also increases the lag time before Web Consoles realize that the Web server has gone down.
You can configure this parameter in the queryIntervalSec parameter for the EventChannel Servlet in the /usr/OV/www/webapps/netview/WEB-INF/web.xml file for UNIX or the \usr\ov\www\webapps\netview\WEB-INF\web.xml file for Windows.
The Client Refresh rate determines how often Web Consoles ask the Web Server to refresh the events for their event browsers.
Normally, you do not need to customize this parameter. However, if the daily servlet request logs (yyyy_mm_dd.request.log in /usr/OV/www/logs directory for UNIX or the \usr\ov\www\logs directory for Windows) are routinely much too large, you can increase the client refresh rate so that fewer EventBrowser event request log entries are created. However, this change also decreases the frequency with which the EventBrowser data is updated in the Web Consoles.
You can configure this parameter in the clientRefreshRate parameter for the EventBrowser servlet in the /usr/OV/www/webapps/netview/WEB-INF/web.xml file for UNIX or the \usr\ov\www\webapps\netview\WEB-INF\web.xml file for Windows.
Both of these actions, as well as lowering the number of days the logs are retained, can be performed on a NetView Web Server running on either a UNIX or Windows system.
Version 7.1.4 of the Tivoli NetView program requires the latest Tivoli Enterprise Console releases (version 3.8 with fix pack 01 and version 3.9) if event forwarding to the Tivoli Enterprise Console server is enabled. This is due to changes in the trap parameters that are sent from the Tivoli NetView program to the Tivoli Enterprise Console server, which is only known to the latest Tivoli Enterprise Console releases.
When configuring a client/server system for the first time on the Solaris operating system, you might see the folloiwng error message during the "Add Client Access" stage:
!-> error message "Can't open /etc/dfs/sharetab" (3 times)-> Operation completed
This is because the /etc/dfs/sharetab file does not exist on Solaris until it is created. The client setup creates the file after failing to access it, so the error messages does not indicate the failure.
The Security Audit Configuration dialog box has two panes.
The upper pane contains the Select Audit Categories field. Use the On and Off buttons in this field to select the categories of audit records that should be stored. Refer to the IBM Tivoli NetView for UNIX Administrator's Guide for information about the events that are recorded under each category.
The lower pane contains the following fields, which can be used to control the size and location of audit log files:
Neither the IBM Tivoli NetView for UNIX Administrator's Guide or the IBM Tivoli NetView for Windows User's Guide nor the comments in the itm_attributes.conf and the servmon.conf files nor the comments in the man pages for the servmon daemon correctly explain how and when the servmon daemon creates Service SmartSets.
The servmon daemon does not automatically create Service SmartSets for service attribute entries (see both the itm_attributes.conf and the servmon.conf files). If you would like to have a Service SmartSet, you should add an entry to the servmon.conf file for that service (or uncomment the commented-out line, if there is already one for that service, making any necessary changes). Refer to the servmon.conf file for details. Then stop and restart the servmon daemon to incorporate the changes.
Alternatively, you can manually create the SmartSet on your own. For Windows, the status for nodes under the SmartSet submap depends only on the service status. However, for User SmartSets, the status for nodes depends on both the IP status and the service status. For UNIX, although Service SmartSets are supported, service icons are not created and the service status does not contribute to the overall status of a node.
Additional Notes on the servmon daemon:
On UNIX, the /usr/OV/bin/netview.pre file is a script a user can create to set environment variables that pertain to the NetView Console. The /usr/OV/bin/netview.pre script is run when the user brings up the NetView Console. Like the /usr/OV/bin/netnmrc.pre script, which is used for daemon startup environment variables, the /usr/OV/bin/netview.pre script is automatically backed up and migrated during an update installation or when users select the /usr/OV/bin.USER category during migration.
The netview.pre script can be used to set the NVMAPGLOBALACK environment variable. Setting the NVMAPGLOBALACK environment variable equal to one activates the global-based mode for map aknowledgments.
The following customer-reported problems are addressed in the Tivoli NetView version 7.1.4 program by this fix pack:
| APAR # | Platform | Description |
|---|---|---|
| IY56398 | Windows UNIX |
Incorrect compound port object status after uninstalling the Tivoli Switch Analyzer program. |
| IY56765 | UNIX |
Demand poll fails on SNMP V3 for the Linux platform. |
| IY56873 | UNIX |
The netmon daemon cores after running xnmsnmpconf -clear<xx>db. |
| IY57383.1 | UNIX |
Tivoli version of the nvtecia -reload command in the Tivoli NetView version 7.1.4 program does not work (update to IY57383). |
| IY58223 | UNIX |
Status updates for open topology objects stop after 15 minutes. |
| IY60881 | UNIX |
Applications registered with -restart are not actually restarted. |
| IY60946 | UNIX |
High CPU usage of ovw_binary under certain conditions. |
| IY61505 | Windows UNIX |
In the submap explorer of the Web Console, the nodes displayed in the left and right panes do not always match. |
| IY61588 | UNIX |
If the entry "!@oid 0" is added to the netmon.seed file after initial discovery, non-SNMP-supported devices are deleted after the daily configuration check or demand poll. |
| IY61634 | UNIX |
The trap varbinds (nv_var1 ... nv_var<n>) are all sent to the Tivoli Enterprise Console server as long as no explicit slot mapping is configured using the xnmtrap (the trapd.conf file). As soon as a slot is defined (such as, severity), only the severity is sent. The man pages for the trapd.conf file will be updated with this information. |
| IY62044 | UNIX |
The nvcorrd man page needs to be updated to indicate that nvcorrd caches the last 100 events, not just the last 25. |
| IY62164 | Windows UNIX |
Users are getting thousands of nonexistent segments added to their database. |
| IY62217 | UNIX |
On Solaris, if 22 or more native GUIs are opened, ovwdb starts to grow rapidly until it gets to 2 GB, then cores. |
| IY62939 | Windows UNIX |
The data mart ETL of the Tivoli NetView Warehouse Enablement Pack is failing in steps anm_m05_s010_metric and in step anm_m05_s020_fact with FOREIGN KEY errors. The problem is that some nodes are not being added to the ANM.D_L3NODES table. |
| IY63318 | UNIX |
The 7.1.4 Release Notes do not mention that as well as not being included on Linux Intel Platforms, certain functions are also not inlcuded for zSeries or S/390 platforms either. That fact will be documented in this readme. |
| IY63336 | Windows UNIX |
The mib2trap command makes the addtrap command silent, which suppresses warning messages that users should see. |
| IY63370 | Windows UNIX |
In the NetView Web Console, the Open Map table is unnecessarily opened for only maps whose name is not default. |
| IY63655 | UNIX |
Periodic ovw_binary cores when pinging nodes. |
| IY63686 | Windows UNIX |
When a new object that belongs to a SmartSet is discovered while the SmartSet is open on the NetView console, the new object is displayed with a generic computer symbol. Closing the SmartSet and reopening it displays the proper symbol. This workaround is documented in this readme under the Known defects and limitations section. |
| IY64107 | Windows UNIX |
*.*.*.* should be allowed as an IP address range for SNMP locking in the netmon.seed file. |
| IY64224 | Windows UNIX |
SNMP "locked" addresses are moving to other SNMP "locked" addresses. |
| IY64226 | Windows UNIX |
In the NetView Web Console, the find and search functions are case sensitive on UNIX, but not on Windows. |
| IY64241 | UNIX |
During the installation of Tivoli NetView 7.1.4 fix pack 2 on Solaris, worrisome but irrelevant messages are displayed (such as, "cp: cannot access .SunOS.V714"). |
| IY64272 | Windows UNIX |
If a user customizes a column within a column view, for example, the IP Address view on the Web Client, the column size is reset to its original size if the user changes the submap or presses the refresh button. |
| IY64366 | UNIX |
The snmpCollect Unix APAR, IY50954, allowed the user to specify IP addresses. However, name resolution was always used. Now the user has the option of not doing name resolution on the IP address. |
| IY64466 | Windows UNIX |
A Node marginal event that is forwarded to the Tivoli Enterprise Console server contains the name of the NetView machine in the origin slot, not the name of the affected node. |
| IY64564 | UNIX |
After creating a SmartSet that contains intefaces, the interface labels in the SmartSet for newly discovered interfaces take the default topology label, rather than the ipmap label which is based on the current IP Map "How should interface labels be shown" configuration parameter. |
| IY65255 | Windows UNIX |
The nvserverd daemon does not create the FQHOSTNAME slot. |
| IY65381 | Windows |
The Tivoli NetView data mart ETL, step ANM_M05_S010_METRIC, fails with a foreign key error when inserting into the ANM.F_3TRS_HOUR table. |
| IY65723 | Windows UNIX |
The readme description of using Secure Socket Layer (SSL) with the Web Console was incomplete. It does not mention the jetty.xml file. |
| IY66025 | Windows UNIX |
The NetView Web Console event viewer rule sets need to be sorted in lexigraphical order. |
| IY66032 | Windows UNIX |
When sparse MIB tables are detected, the Java MIB Browser does not warn users that the returned data might be displayed incorrectly. |
| IY66096 | UNIX |
The /usr/OV/bin/nvhotbackup command with the df -k sometimes fails. |
| IY66382 | UNIX |
After using the loadhosts -a option to load an un-pingable device, the device goes down. However, the interface status change is not propagated to the segment or to the network in the database. However, the status does show up correctly on the map. |
| IY66383 | Windows UNIX |
The nvmaputil.sh and nvmaputil.bat --unmanage-interface command does not work correctly. |
| IY66397 | UNIX |
DebuggingToolsV2 does not install on a system without the tools already installed. |
| IY66440 | UNIX |
Unsecure version of the ovobjprint command (not shipped with the fix pack) |
| IY67304 | UNIX |
The output displayed by rnetstat -an is corrupt. |
| IY67766 | UNIX |
Document the fact that on Solaris only about 24 IBM Tivoli NetView GUIs can be open at once. |
| IY67784 | UNIX |
netmon cores when unmanaging/managing a compoundPort object while deleting or unmanaging a network. | PJ29147 | Windows |
On Intel Windows machines, incrementing dummy addresses for serial links was performed in the wrong byte order meaning that unnumbered serial links that were not discovered before the first restart of the netmon daemon were never discovered and the compound port status object could not be created. |
| PJ29866 | Windows |
Nodes added using the loadhosts command never appeared on the NetView GUI. |
| PJ29983 | Windows UNIX |
A rule set queries a SmartSet and causes an "NVCOLLECTION API not initialized" error. |
| PJ30029 | Windows |
netmon loses the connection to the ICMP service due to socket problems. |
| PJ30038 | Windows |
Using a custom filter in tess and the Tivoli NetView for Windows event browser, the browser window disappears after a while, with this text shown in the nv.log file: [tess] OnTimer exception at requery with hstmt:0. (-1) State:S1000,Native:-1045,Origin: Microsoft ODBC Microsoft Access Driver. Not a valid bookmark. |
| PJ30082 | Windows |
Using the Tivoli NetView for Windows event browser, garbage or data from the previous event is shown on the Event Detail panel in tess. |
| PJ30090 | UNIX |
Modify wording for AIX V4 and V5 in the oid_to_type file. |
| PJ30107 | Windows |
NetViewD at Windows takes always the map "default". |
| PJ30154 | Windows |
Netmon dies on dual CPU boxes. |
The following customer-reported problems were fixed in the Tivoli NetView version 7.1.4 program by Fix Pack 02:
| APAR # | Platform | Description |
|---|---|---|
| IY49659 | UNIX |
For AIX 5.1, commands launched by the ovactiond daemon are not killed when wait time is exceeded. |
| IY50486 | Windows UNIX |
The DB2 Availability database for the Tivoli NetView program has duplicate records (see IY53390 and IY55671). |
| IY51398 | Windows UNIX |
nvdbformat performance improvement |
| IY52871 | UNIX |
Additional actions of browse mib and sorting events in the Tivoli NetView events application fail with shared library load errors. |
| IY53390 | Windows UNIX |
The central data warehouse ETL results in duplication of node (comp_id) and measurement values pairs when inserting availability data into TWH_CDW. |
| IY53554 | UNIX |
The mibtable command gets stuck looping when attempting to retrieve status tables from the Mid-Level Manager MIBs. |
| IY53580 | UNIX |
APM data collection is returning the same value continuously. |
| IY53703 | Windows UNIX |
The trapd daemon does not format hexadecimal data in traps as expected. |
| IY54430 | UNIX |
The netviewd daemon fails to start. The message in the ovstatus output stanza is "Exiting for unknown reason." The workaround for this is documented in this fix pack. |
| IY54434 | Windows UNIX |
For complex SmartSets, the nvUtil evaluation command does not always yield the correct results, especially if the right-hand side of the rule ended with a NOT operator (for example, !(xxxx)) and had multiple && conditions. |
| IY54452 | UNIX |
Backup the /usr/OV/ITSL2 file during migration of the Tivoli NetView program. |
| IY54575 | UNIX |
On Linux, the nvcorrd daemon cores when delete [] is called for an array that was not dynamically allocated. |
| IY54590 | UNIX |
The ovtopofix command should delete empty networks. |
| IY55011 | UNIX |
The nvUtil exits when it comes across an object with no Selection Name in the list of objects that is retrieved from the ovw database, issuing the following message: "GetValues failed - OVwDbGetFieldValuesByObjects returned NULL values. Verify that the ovwdb cache size is greater than the number of objects in the database." |
| IY55025 | UNIX |
The netview_root.log file gets filled with font errors on AIX 5.2 because the "/usr/OV/app-defaults" directory contains ibm850 fonts, which are not supported with LANG=en_US. |
| IY55219 | Windows UNIX |
The netmon daemon cores on Solaris if DEMANDPOLL and PING STATUS POLL happen at the same time. |
| IY55256 | UNIX |
The nvsecd daemon cores launching nvsec_admin if subdirectories exist in /usr/OV/security/C/Domains/* directory. |
| IY55354 | UNIX |
HSRP interface gets deleted and recreated every minute. |
| IY55559 | UNIX |
The addition of the AutoTrace facility in version 7.1.4 caused major performance degradation to the snmpwalk command. |
| IY55671 | Windows UNIX |
The tdwdaemon is adding multiple records to the Availability database for the same status event because of the way the netmon daemon is generating events. |
| IY55689 | Windows UNIX |
Layer 2 status does not propagate correctly to switch icons on SmartSet or custom maps. |
| IY55803 | UNIX |
The online books and helps for the Tivoli NetView program have Netscape hard-coded as the browser to use. |
| IY55831 | UNIX |
Spurious ovwdb hang during communications with the ovspmd daemon. |
| IY55966 | UNIX |
Holding the spacebar down causes ovw_binary to core in about 20 seconds. |
| IY55972 | UNIX |
$AGENT_ADDR does not resolve the host name in events that are forwarded to the Tivoli Enterprise Console server. |
| IY56095 | UNIX |
Two NODE UNMANAGED traps for one UNMANAGE action. |
| IY56157 | UNIX |
Slow performance when closing multiple NetView windows causing the maps to grey out and hang around for a while. |
| IY56279 | UNIX |
Duplicate events might be sent to the Tivoli Enterprise Console server if users run ovstop nvserverd or ovstart nvserverd commands without stopping the nvcorrd daemon as well. |
| IY56323 | UNIX |
The NMAVAILABILITY=TRUE flag in the netmon.conf file does not work. Regardless of the setting in the netmon.conf file, no available data was created in the /usr/OV/databases/availData_yyymmdd file. |
| IY56338 | Windows UNIX |
The netmon daemon does not resolve seed file names that start with a digit. |
| IY56522 | UNIX |
On Linux, the netcheck TCP routine displays an incorrect error message: "ERROR: Could not connect to port 5888: Connection refused." |
| IY56534 | UNIX |
The UNIX -R option (remove empty networks and segments) for the ovtopofix now works by itself, not just in conjunction with the -A and -a options. |
| IY56563 | Windows UNIX |
The Server Setup application does not include the ability to set -w flag for the snmpCollect daemon. This results in the Tivoli Data Warehouse function being disabled. |
| IY56564 | Windows UNIX |
The -w flag for the snmpCollect daemon is not documented in the IBM Tivoli NetView Warehouse Enablement Pack Implementation Guide or in the snmpCollect man page. |
| IY56565 | UNIX |
No DB2 activity tracing messages or options related to snmpCollect writing data to the Tivoli NetView warehouse database. |
| IY56637 | Windows UNIX |
The netmon -a 160 command does not dump the value of the TDWAVAILABILITY flag. |
| IY56671 | Windows UNIX |
Enumerations are lost from the Tivoli NetView database after running the ovstop, ovstop nvsecd, or the /etc/netnmrc command. |
| IY56958 | Windows UNIX |
Eliminate major memory leak in the nvcold daemon that is triggered by running the nvUtil evaluate commands. |
| IY57192 | UNIX |
Zooming on submaps does not always work. |
| IY57383 | UNIX |
The nvtecia -reload command does not work with tme_nvserverd. The process dies. |
| IY57388 | UNIX |
On Linux, rule sets have problems with nodes that do not have names. |
| IY57432 | UNIX |
Parsing failure at the Tivoli Enterprise Console server for Tivoli NetView events with big interface lists (varbind 8). |
| IY57468 | UNIX |
Turning off "Poll for Status" does not disable SnmpStatus poll, but it does disable ICMP ping. |
| IY57469 | UNIX |
When all settings from the Topology/Status Polling Configuration are turned off, explicit seed file entries are not discovered. |
| IY57579 | UNIX |
APAR IY4909 can cause performance problems during startup of the netmon daemon. |
| IY57654 | UNIX |
The ovtopmd daemon cores when there are incomplete objects left in the database. It occurs only in rare cases. |
| IY58119 | UNIX |
Users used the SmartSet editor (collectioned) dialog to create SmartSets where the first definition was a "two word" attribute and the second definition was an IN_COLLECTION test. Creating the SmartSet was successful. However, when the user subsequently tried to modify that SmartSet using the SmartSet editor dialog, the SmartSet editor said "SmartSet cannot be correctly expressed in dialog. Text editor used instead." |
| IY58244 | Windows UNIX |
The nvdbformat command gave the wrong results with some selectrule definitions. |
| IY58305 | UNIX |
ovw_binary cores when resizing window. |
| IY58392 | UNIX |
The trapgend daemon fails to properly parse CORE_DUMP on AIX 5.2. |
| IY58404 | Windows UNIX |
If there are problems with the first address, snmp_addr in the netmon cache should not be changed to be the second address if the second address also fails. |
| IY58428 | UNIX |
The netmon daemon exits with a signal 13 after the Ctrl-C key combination is issued to stop a nmdemandpoll. |
| IY58512 | UNIX |
The snmpCollect daemon is not closing DB2 database connections, which results in many connections still listed as active even though they are not being used. |
| IY58514 | Windows UNIX |
Using the Server Setup application for the tdwdaemon to create the DB2 NetView warehouse database unnecessarily issues db2stop and db2start commands. |
| IY58551 | UNIX |
The mib2trap command does not add slotmap embedded comments into the baroc file. |
| IY58586 | UNIX |
The user brings up the SmartSet editor (collectioned) either from the native NetView console or from the command line. The user then clicks Add -> Modify and then switches the Definition type to Subnet. The user then clicks Calculate, types in an IP address and Subnet Mask, and then clicks Calculate. collectioned then cores with a Segmentation Fault. |
| IY58772 | UNIX |
The sqlstate and sqlcode values are not being written to the snmpCol.trace file during failures between snmpCollect-UNIX and DB2. |
| IY58780 | Windows UNIX |
Icon size in the submapexplorer is sometimes calculated incorrectly. |
| IY59196 | Windows UNIX |
The ovtopmd daemon cores due to memory leaks. |
| IY59201 | Windows UNIX |
The ovtopmd daemon grows in size when objects are deleted from the map. |
| IY59424 | Windows UNIX |
Status is reported incorrectly on SnmpStatus poll whenever there is a timeout on the request. |
| IY59537 | Windows UNIX |
The servmon daemon does not clear the SmartSet ovw database entry when a node is deleted. |
| IY60528 | UNIX |
Tivoli NetView event severity is no longer sent as part of Tivoli Enterprise Console events in version 7.1.4. Add an environment variable to the nvseverd daemon to cause it to put severity back on but omit the severity by default. |
| IY60701 | UNIX |
The netmon daemon reports interface down events for the rest of the interfaces after encountering an "Snmp error: no such name" on one of the SNMP status requests (req = SNMPStatus). |
| IY60978 | UNIX |
Global acknowledging does not work for location objects. |
| IY61174 | Windows UNIX |
Error in the /usr/OV/conf/tdwdaemon.log file: "Wrong number of parameters". |
| PJ29495 | Windows |
The setstatusseverity.rs ruleset is broken. |
| PJ29522 | Windows UNIX |
The OVwFieldNameToFieldId API call fails if the Tivoli NetView program is recycled and the client application is not. A new ovw API, OVwDbDone(), was added so that applications can recover from a lost connection to ovwdb. |
| PJ29525 | Windows |
User created backgrounds for submaps are not being displayed. |
| PJ29701 | Windows UNIX |
A Tivoli NetView SmartSet is not created for ITM services. |
| PJ29726 | Windows UNIX |
Restrict SNMP status polling to discovered interfaces. |
| PJ29734 | UNIX |
When the UNIX xnmgraph View -> Statistics is saved to a file, the output line field is limited to 20 characters in the output file. |
| PJ29812 | Windows |
The Windows Tivoli Enterprise Console event forwarding daemon (tecad_nv6k) stops working after consuming lots of memory (determined by looking at the Task Manager Window). |
| PJ29814 | Windows |
"Configure Data Export to DB2 for use in TEDW" does not add the -w flag to snmpcollect so that it writes records to DB2. Moreover, nvsetup.exe configuration for snmpcollect does not update the ovsuf and snmpCol.lrf files with the Tivoli Data Warehouse "-w" flag. |
| PJ29821 | Windows |
Parsing failure at the Tivoli Enterprise Console server for Tivoli NetView events with big interface lists (varbind 8). |
| PJ29910 | Windows UNIX |
SQL statements are missing single quotes around zeroes in an SQL command file. |
The following customer-reported problems were fixed in the Tivoli NetView version 7.1.4 program by Fix Pack 01:
| APAR # | Platform | Description | |
|---|---|---|---|
| IY41085 | Windows UNIX |
ISDN status monitoring does not work on Cisco routers due to the organization of MIB data. | |
| IY42473 | UNIX |
The Tivoli NetView program does not restart applications that are registered with the "-restart" parameter. | |
| IY47145 | Windows UNIX |
Documentation request to explicitly state requirements on the entries in the netmon seed file to discover and SNMP manage nodes. (See the entry for this APAR in the Product Notes section.) | |
| IY47892 | Windows UNIX |
The ovesmd daemon is making DNS requests on networks and network segments, unnecessarily degrading performance of the DNS server. | |
| IY48379 | Windows UNIX |
The ovstop nvsecd command does not stop the tdwdaemon. (See the Product Notes section to implement this correction,) | |
| IY48447 | Windows |
Error in the help message panel for object status for specific traps. | |
| IY48542 | UNIX |
The trapd daemon causes a core dump with a buffer overflow. | |
| IY49032 | UNIX |
The Server Setup application on Solaris and Linux contains AIX-specific terminology and options. | |
| IY49405 | UNIX |
CNAT data is not migrated from earlier versions of the Tivoli NetView version 7.1.x program to version 7.1.4. | |
| IY49428 | Windows UNIX |
The nvdbformat application "TopM Interface List" is always prepended by a carriage return. | |
| IY49532 | Windows UNIX |
The nvdbformat application does not complete and gives incorrect output. | |
| IY49546 | UNIX |
The printtool application causes a core dump on the Tivoli NetView version 7.1.3 program on Linux. | |
| IY49787 | UNIX |
Tivoli NetView version 7.1.3 ovw_binary core. | |
| IY49953 | UNIX |
The nvauth executable cores with a segmentation fault when bad data is given. | |
| IY49954 | UNIX |
An overflow condition occurs on the command line parameters of the nvauth executable. | |
| IY49955 | UNIX |
The nvauth executable causes a core dump when the password command line argument is longer than expected. | |
| IY49975 | UNIX |
The ipmap executable never ends synchronization and causes a core dump when GLOBAL ACKNOWLEDGE is enabled. | |
| IY49985 | UNIX |
The pmd daemon causes a core dump when its port is scanned by third-party software. | |
| IY50013 | UNIX |
The nvlockd daemon causes a core dump when its port is scanned by third-party software. | |
| IY50133 | UNIX |
Installation of the Tivoli NetView version 7.1.4 program on AIX 4.3.3 ML 10 fails to start the nvsecd daemon, issuing the error message "ovstart: unable to contact ovspmd: No such file or directory." | |
| IY50196 | Windows UNIX |
Request for additional alternative community names. | |
| IY50227 | Windows |
Unnecessary trace messages in the servmon.log file. | |
| IY50265 | UNIX |
The nvcorrd daemon causes a core dump when bad data is written to the nvcorrd socket. | |
| IY50266 | UNIX |
The actionsvr daemon causes a core dump when bad data is written to its socket. | |
| IY50394 | Windows UNIX |
The IPMAP application causes a core dump after migrating from the Tivoli NetView version 7.1.2 program to version 7.1.3. | |
| IY50461 | Windows UNIX |
Provide support in the location.conf file for wildcard IP addresses for routers. | |
| IY50464 | Windows UNIX |
The netmon daemon is discovering several devices and setting them as unmanaged even though it is configured to manage newly discovered nodes. | |
| IY50663 | UNIX |
The ovtopmd daemon and the ovtopofix command run much more slowly when AutoTrace is turned on. | |
| IY50702 | Windows UNIX |
A switch managed by the Tivoli Switch Analyzer program might not have the correct symbol status shown in ipmap. | |
| IY50954 | UNIX |
The snmpCollect daemon does not collect on an IP address if the address resolves to a host name. | |
| IY50990 | Windows UNIX |
The netmon daemon does not wait the configured amount of time to perform status polling after a restart. | |
| IY51064 | Windows UNIX |
Query Database Field node does not work for an interface selection name in a rule set. | |
| IY51125 | Windows UNIX |
Errors using SNMP wildcards in the netmon seed file in the Tivoli NetView version 7.1.3 program. | |
| IY51162 | UNIX |
Trap slot values with long names are truncated to 15 characters when it is saved. | |
| IY51181 | UNIX |
The right scrollbar becomes non-functional in the seed file editor. | |
| IY51516 | Windows UNIX |
The nvrsEdit application does not display an error if it cannot handle a ruleset correctly at open. | |
| IY51641 | UNIX |
The nvserverd daemon causes a core dump due to a buffer being too small to properly format a very large trap. | |
| IY51700 | Windows UNIX |
The Tivoli NetView version 7.1.4 program requires the latest Tivoli Enterprise Console releases if event forwarding to the Tivoli Enterprise Console server is enabled. | |
| IY51778 | Windows UNIX |
Behavior of a range of addresses in netmon seed file differs from the previous release (see the Product Notes section). | |
| IY51844 | Windows UNIX |
Rediscovery of routers does not work. | |
| IY52086 | UNIX |
Need to clarify the meaning of the "retries" parameter for the netmon daemon (see the Product Notes section). | |
| IY52227 | UNIX |
Need to be able to acknowledge layer 2 marginal status on a switch that has an IP status of normal. | |
| IY52295 | UNIX |
Need to add ping count control to non-AIX Tivoli NetView UNIX platforms. It is already on AIX. | |
| IY52234 | UNIX |
Upgrading to version 7.1.4 results in incorrect IP topology counts in the relational database when the ipgettopoinfo command is run. | |
| IY52320 | UNIX |
The SNMP configuration look up routine is now returning NetView's default value instead of user's setting in Global Default. | |
| IY52341 | Windows UNIX |
The ipmap executable is coring on Tivoli NetView 7.1.3 Fix Pack 1 on AIX 5.1. | |
| IY52351 | Windows UNIX |
Several traps are causing DNS requests for network and segment addresses. | |
| IY52394 | Windows UNIX |
The nvcorrd daemon causes a core dump in function setstatednode with signal 11. | |
| IY52429 | UNIX |
In the Tivoli NetView version 7.1.4 program, a trap slot with host name in a Tivoli Enterprise Console event is left blank if the host name contains a hyphen (-). | |
| IY52758 | UNIX |
The xnmgraph application is calculating wrong y-axis values. | |
| IY52773 | Windows UNIX |
The netmon daemon causes a core dump with a signal 11 every 10 minutes. | |
| IY52820 | UNIX |
The trapd daemon is coring on a regular basis in the Tivoli NetView version 7.1.4 program on AIX 5.2. | |
| IY53038 | UNIX |
On the Tivoli NetView version 7.1.3 and 7.1.4 program for Linux, non-root users are not allowed to start the GUI due to a missing line in the startup script. | |
| IY53188 | Windows UNIX |
Document the fact that users can modify the maximum number of community names by setting the following variable in netmon.conf file: MaximumCommunityNames=x. The value of x can be from 0 to 100. | |
| IY53205 | Windows UNIX |
Attempt to add status monitoring of an IBM Tivoli Monitoring resource model to the servmon daemon did not work. | |
| IY53244 | UNIX |
Garbage printed in the command usage for the nvgethost command. | |
| IY53271 | Windows UNIX |
The trapd daemon stops when toggling tracing or logging on Linux on zSeries. | |
| IY53400 | UNIX |
The xnmloadmib2 and xnmbrowser2 applications cannot run if NetView security is enabled. No security registration files exist. | |
| IY53554 | UNIX |
The mibtable command loops on Mid-Level Manager MIBs. | |
| IY53666 | UNIX |
The trapd daemon stops with a signal 13 when the trapd.log file reaches its maximum size. | |
| IY53704 | UNIX |
Problem editing a Set State node in the NVRSEdit application. | |
| IY53727 | Windows UNIX |
An incorrect symbol is showing up on the map for several interfaces. | |
| IY53780 | UNIX |
The netmon daemon pings unwanted interfaces. | |
| IY53804 | Windows UNIX |
The nvdbformat application is coring after applying the Tivoli NetView version 7.1.3 Fix Pack 2. | |
| IY53928 | Windows UNIX |
A servmon daemon performance problem exists in the Tivoli NetView version 7.1.4 program. | |
| IY54371 | Windows UNIX |
"Interface Unreachable" events occur on every polling cycle when router fault isolation is enabled. | |
| IY54644 | Windows UNIX |
The NetView Web Console is hanging. | |
| IY54964 | Windows UNIX |
The netmon daemon is trying to poll thousands of interfaces simultaneously on large networks. | |
| IY55374 | UNIX |
Unclear documentation for using the snmpd64v1 agent on AIX 5.2 | |
| PJ29055 | Windows |
The output of the origin tecad default slot values for certain traps is incorrect. | |
| PJ29416 | Windows |
The hostname field is being left blank when forwarding events to Tivoli Enterprise Console server. | |
| PJ29421 | Windows UNIX |
The ovobjprint command is not displaying all appropriate information on large databases. | |
| PJ29460 | Windows |
Issue with icon selection name and label when the netmon daemondiscovers a router whose first interface is not in DNS. | |
| PJ29480 | Windows |
Node status is not properly updated after a service symbol is deleted based on the Service Down interval. | |
| PJ29481 | Windows |
Node status in a SmartSet might be incorrect when node comes back up. | |
| PJ29503 | Windows UNIX |
Deleted service symbols might propagate incorrect status. | |
| PJ29517 | Windows |
The nvcord daemons exchanges a "$A" value incorrectly for an event that passed through an activated rule set. | |
| PJ29580 | Windows |
The "filter by" option on the Collected Data objects properties dialog box does not reflect data if the "<" or ">" options are used. | |
| PJ29586 | Windows UNIX |
Cannot get or set fields that contain a comma (,) the nvdbformat and nvdbimport commands. | |
| PJ29588 | Windows UNIX |
Excessive number of truncation errors logged when using the Tivoli NetView ETLs. | |
| PJ29644 | Windows |
Tivoli NetView for Windows trapd daemon exception on null IP address from Microsoft Trap Service. | |
| PJ29651 | Windows UNIX |
Tivoli NetView Warehouse Enablement Pack fails to insert the destination data because of string truncation errors." | |
| PJ29653 | Windows UNIX |
The Tivoli NetView Warehouse Enablement Pack is missing historical data for the reports that show percent outage for both month and yearly reports. | |
| PJ29654 | Windows UNIX |
The Tivoli Netview data mart ETL is not calculating availability accurately in all cases. | |
| PJ29666 | Windows |
The snmpCollect daemon repeats snmpget requests when the number of objects to collect exceeds 16. | |
| PJ29671 | Windows |
The graph function is not working for previously defined data collection. |
To install IBM Tivoli NetView V7.1.4 Fix Pack 03 on a system, you must meet the following prerequisites:
You must download the appropriate package for that platform and make it available to the system on which it is to be installed.
You must be logged into the system with the same user account that performed the original Tivoli NetView version 7.1.4 installation (the root account on UNIX).
The system must already contain a Tivoli NetView Version 7.1.4 installation.
The system on which the fix pack is to be installed must have enough free space to do the following:
Installations that contain a previous Tivoli NetView fix pack might recover most of the space used by that fix pack by deleting the /usr/OV/service/V714/Fix PackX directory on UNIX or the \usr\ov\service\v714\Fix PackX directory on Windows prior to installing this fix pack. Please note, however, that doing so removes the ability to back out the previous fix pack, as well as the ability to install code fixes that require the previous fix pack for installation.
The following table describes the amount of space that is required on each platform for each of the space requirements described above.
Operating System |
Space Required During Installation1 |
Space Required for Upgraded Product2 |
|---|---|---|
RS6000 AIX |
625 Mb |
400 Mb |
SunOS Solaris |
740 Mb |
485 Mb |
Linux Intel |
605 Mb |
360 Mb |
Linux 390 |
570 Mb |
330 Mb |
Windows |
115 Mb |
65 Mb |
1This space is required in whatever location the user places the fix pack installation package. 2This space is required in /usr/OV on UNIX or in \usr\ov on Windows. |
||
Most of the space required to contain the fix pack contents in the /usr/OV or \usr\ov directory (the space described in the second column) can be returned to the system by deleting the contents of the /usr/OV/service/V714/FixPack3 directory (on UNIX) or the \usr\ov\service\V714\FixPack3 directory (on Windows). However, doing so removes the ability to back out the fix pack. Do not delete the FixPack3 subdirectory itself. This subdirectory must exist to install fixes that have this fix pack as a prerequisite.
For Red Hat Enterprise Linux AS Version 3.0 and SUSE Linux Enterprise Server (SLES) 8.0 and 9, a new Tivoli NetView 7.1.4 Linux for zSeries (64 bit) CD has been added to the base Tivoli NetView 7.1.4 kit, and a standard new Tivoli NetView installation is required. This new CD is ONLY for new Tivoli NetView installations on Red Hat Enterprise Linux AS Version 3.0 and SUSE Linux Enterprise Server (SLES) 8.0 and 9 for zSeries (64 bit).
See Software requirements for the new full image Linux for S/390 (31 bit) CD for more details on Linux packages that must be installed for the Tivoli NetView program to work correctly.
For Red Hat Enterprise Linux Advance Server 3.0 i386 (32 bit) and S/390 (31 bit) and SUSE Linux Enterprise Server (SLES) 9 i386 (32 bit) and S/390 (31 bit), a new NetView 7.1.4 Linux CD has been added to the base NetView 7.1.4 kit and a standard new Tivoli NetView installation is required. This new CD is ONLY for new Tivoli NetView installations on Red Hat Enterprise Linux Advance Server 3.0 i386 (32 bit) and S/390 (31 bit) and SUSE Linux Enterprise Server (SLES) 9 i386 (32 bit) and S/390 (31 bit).
For the Tivoli NetView installation on Red Hat Enterprise Linux Advance Server 3.0 i386 (32 bit) and S/390 (31 bit) and SUSE Linux Enterprise Server (SLES) 9 i386 (32 bit) and S/390 (31 bit) to work correctly, the following Linux packages must be installed first.
The user must install the optional Red Hat i386 package, compat-libstdc++-7.3-2.96.122, or the S/390 package, compat-libstdc++-7.2-2.95.3.78. This package can be found on the Red Hat Enterprise Linux Advance Server 3.0 Update 1 or 2 CD set.
To correct ifIndex and ipAdEntIfIndex mismatches in the ifIndex reference by IP-MIB, the user needs to update the net-snmp RPMs to the following levels:
rpm -i --force pdksh-5.2.14.8.i386.rpm (or pdksh-5.2.14-197.s390.rpm)
Verify the installation using the following command:
rpm -q pdksh
Note: On SLES 9, the Xvfb package is XFree86-Xvfb, rather than xextra.
During a new installation, if at least one package in each group is not found, then installation prerequisite checking fails and installation ends.
For Red Hat Enterprise Linux Advance Server 3.0 i386 (32 bit) and S/390 (31 bit) and SUSE Linux Enterprise Server (SLES) 9 i386 (32 bit) and S/390 (31 bit), get the new Tivoli NetView 7.1.4 Linux CD from IBM and follow the existing directions for a standard install.
For Red Hat Enterprise Linux AS Version 3.0 and SUSE Linux Enterprise Server (SLES) 9 for zSeries (64 bit), get the new full image Linux for zSeries (64 bit) CD from IBM and follow the existing directions for a standard install.
Note: The new Tivoli NetView CDs can be ordered through normal IBM Passport Advantage channels.
For all other platforms, perform the following steps to install the fix pack:
Based on whether the Web Console has been customized or not, users of the Secure Socket Layer (SSL) protocol need to do one of the following:
If you are installing the fix pack on a Tivoli NetView for UNIX client, ensure that the client is disconnected from the Tivoli NetView server by completing the following steps:
Open a new window, copy the package into a local subdirectory on the operating system on which the fix pack is to be installed, and change directories to that subdirectory.
Prior to copying the package, ensure that the space available in this subdirectory is larger than the space required for installation of the IBM Tivoli NetView 7.1.4 Fix Pack 03 on this operating system.
Extract the installation script by doing one of the following:
zcat 7.1.4-TIV-NVD-{OperatingSystem}-FP0003.tar.Z | tar -xvf - install
7.1.4-TIV-NVD-Win32-FP0003.exe
Run the installation script:
./install
install
install noversioncheck
When installing the fix pack on a native NetView client installation, the installation process reports a set of errors that certain files (such as, nvexportd.jar, nvexportd.lrf, and notfoundhandler.class) could not be copied. These files are not necessary for a client installation, and you can ignore the errors.
If you encounter any other errors during the installation (regardless if the installation process itself reports success), complete the following steps:
Restart the Tivoli NetView daemons on UNIX either by rebooting the machine or by running the netnmrc script (/etc/netnmrc on AIX, /etc/init.d/netnmrc on Solaris or Linux). Restart the Tivoli NetView daemons on Windows by rebooting the machine and either running the netview command or by clicking on the Tivoli NetView icon.
Note: If the Tivoli NetView installation is using a SQL database on Windows, restart the SQL server before you restart the Tivoli NetView daemons.
Reinstall any Tivoli NetView Web Console installations from the upgraded native NetView server installation. You can download the Web Console from the following Web site, after the NetView server has been upgraded:
http://server:8080/download
For the Web Console on Windows, the following files contain the fix pack updates. If you have been using the thin client, apply the nvwc_thin.zip file. If you have been using the full Web Console, apply the nvwcinstall.exe executable.
nvwc_thin.zip (thin Web Console for Windows) nvwcinstall.exe (self-extracting installation)
For the Web Console on UNIX, the following file contains the fix pack updates:
nvwc.tar (Web Console for UNIX)
Note that when upgrading a NetView Web Console from a NetView fix pack UNIX server, this version of the NetView Web Console package should be used. The full version (nvwc_xxx_jre_1.3.tar) of the NetView Web Console does not contain the Fix Pack updates. After the Fix Pack has been installed on a NetView server, the nvwc.tar package must be applied on top of the full package whenever the full package is installed.
As a part of the fix pack installation, a copy of each file that is replaced during the installation is saved in the /usr/OV/service/V714/FixPack3 sub directory on UNIX or the \usr\ov\service\V714\FixPack3 subdirectory on Windows. If you want, you can uninstall the fix pack by running the backout script that is contained in that directory, and the system returns to the state that it was in prior to the installation of the fix pack.
Note: For Windows, you must run the backup script from the \usr\ov\service\V714\FixPack3 subdirectory. Similarly, for UNIX, you must run he backout script from the /usr/OV/service/V714/FixPack3 subdirectory.
Any Tivoli NetView fixes installed on the system after the fix pack is installed must be backed out prior to backing out the fix pack.
Any modifications to the NetViewRealm.xml file to customize the NetView Web Console user password function provided in NetView 7.1.4 Fix Pack 2 will be overwritten when this Fix Pack is installed. The customizations can be restored by copying /usr/OV/service/V714/FixPack3/NetViewRealm.xml.0 (\usr\ov\service\V714\fixpack3\NetViewRealm.xml.0 on Windows) back to /usr/OV/www/conf/NetViewRealm.xml (\usr\ov\www\conf\NetViewRealm.xml on Wndows).
A problem has been found in the State Correlation Engine of the NetView/TEC Integration adapter on the zLinux (System 390) versions of Red Hat Enterprise Linux Advance Server 3.0 and SuSE Linux Enterprise Server (SLES) 9.0. The nvserverd daemon will core on those platforms if the tecint.conf file specifies UseStateCorrelation=YES due to an unresolved internal conflict between the State Correlation Engine and the X11 libraries used for the events display. For this reason, the State Correlation Engine is being disabled (by changing the value of the UseStateCorrelation statement to "NO" in the tecint.conf file) during NetView 7.1.4 FixPack 3 installation on zLinux versions of Red Hat 3.0 and SLES 9.0; events sent to TEC from these platforms will not contain an fqhostname field, nor will these platforms produce and send Service Impact events. Earlier versions of zLinux and all other platforms supporting NetView 7.1.4 FixPack 3 are not affected.
Please note that if Server Setup is used to set up or modify NetView/TEC Integration after NetView 7.1.4 FixPack 3 is installed, the user will need to manually edit /usr/OV/conf/tecint.conf to set the UseStateCorrelation value to "NO" on zLinux Red Hat 3.0 and SLES 9.0 systems.
When installing a standalone NetView Language Pack on a NetView UNIX installation, the NVenvironment script must be sourced using the command
$ . /usr/OV/bin/NVenvironmentin the command window in which the installation will be performed.
The Solaris operating system has a limitation of 256 open file descriptors for a process that has been compiled in 32 bit mode. This limitation is in the libc library for Solaris. Due to this limitation, only about 24 Tivoli NetView GUIs can be open at the same time.
Regular expressions provide a way to search for specific strings from a set of character strings. For the Web Console on UNIX systems, the search is case-sensitive. On Windows systems, the search is not case-sensitive.
The NetView console SNMP view shows RS/6000s with AIX version 5 as being at AIX version 4.1.
The fix pack installs an updated version of the oid_to_type file under the /usr/OV/newconfig/OVIPMAP directory. To pick up the update, a user can copy the new oid_to_type file to the/usr/OV/config directory (or migrate the change to their current oid_to_type file). With the change, RS/6000s with AIX Version 5 will indicate that the AIX version is V4 or V5.
For installations with large enterprise networks, change the nvsetup->Discovery->Discover Using Ping Spray option from Yes to No. Choosing to "Discover Using Ping Spray" seriously degrades the performance of discovery for large enterprise networks.
Use the netmon -n flag to receive critical notifications about failures in an unmanaged Wide Area Network (WAN).
Assume that there is an outsourced MPLS WAN that connects two sites. Normally, when a failure in the WAN occurs preventing access to the remote site, the Tivoli NetView program reports that the entire remote site is Unreachable and issues Network and Router Unreachable events. Because the events are non-critical events, the problem does not get the attention it deserves from the operators.
You can configure the Tivoli NetView program to issue a critical event for the remote border router, which will draw the attention of the operators to the WAN. To do this, follow these steps:
Because unmanaging a network prevents other end node devices from being managed, it is better to unmanage a subnet that represents a serial link. Unmanaging a network does not affect the managed state of the router interfaces.
Here is an explanation of what is happening with router fault isolation to make this work. The goal is to make the remote MPLS border router go Critical instead of Unreachable when the MPLS route fails.
Router fault isolation judges the state of an interface by two factors:
The state of the router itself is based on the propagation of all the interface states.
For a router to be unreachable, every interface must be down AND each network must be unreachable or unmanaged. If the netmon -n switch is set, then an unmanaged network is treated as reachable. For a router to be down, no interfaces can be up AND at least one interface must be down in a reachable network. The other interfaces can be unmanaged and, therefore, are ignored.
Now when the route through the WAN fails, the router interface is down and its unmanaged network is considered reachable. This is enough for router fault isolation to issue a Router Down event for this router.
When a new object (for example an interface, network, segment, or node) that belongs to a SmartSet is discovered while the SmartSet is open on the NetView console, the new object is displayed with a generic computer symbol. Closing the SmartSet and reopening it displays the proper symbol.
Note: If you run the ovstop followed by running the ovstart or /etc/init.d/netmrc command, the nslookup might core again.
To determine whether a /usr/OV/PD/cores/ovspmd/core is due to this problem, run the following command on the core file:
strings core | moreIf the string nslookup appears in the output, you can ignore the core.
TDWAVAILABILITY=FALSE
The Tivoli NetView program might have discovery and status updating problems for machines that are managed by the Tivoli NetView program and are running SNMPv3 agent daemons. These problems can be eliminated by ensuring that the SNMP daemon of the managed machines are linked to an SNMPv1 daemon and not linked to an SNMPv3 daemon.
UNIX users who use the Web Console with a large NetView database, follow these steps:
On AIX, the netviewd daemon fails to start if the correct X11 Xvfb is not running.
If the netviewd daemon fails to start on AIX because the correct X11 xvfb is not running (APARs IY54430 and IY58569), the user needs to add grep ":2" | into the /etc/netnmrc file as shown in the following example.
# Start Xvfb Xserver
#
pid=`/bin/ps -ef | /bin/grep vfb | /bin/grep -v grep | grep ":2" | /bin/awk '{print $2}'`
Nodes with "Discover Node(s) Managed" unchecked using the xnmsnmpconf application (UNIX) or due to the settings in the oid_to_type file that are discovered when the native NetView console is running, show up on the map as Unknown rather than as Unmanaged.
The workaround (documented in APAR IY58520) is to make sure the native NetView console is not running when these nodes are being discovered or to unmanage the nodes manually from the GUI when they show up on the map as Unknown. In addition, if the user wants to manage the node, the user must first unmanage it and then manage it.
Manipulation of the Korn shell history file from a .kshrc file is not recommended. Several of the installation mechanisms used by the Tivoli NetView program expect the standard history file to be set. If it is not set, the prerequisite checking and the actual Tivoli NetView installation fails.
Some users have encountered difficulties when using the stand-alone Tivoli NetView installation on a system that has the Tivoli Management Framework installed in a customized location. To avoid these difficulties, it is recommended that if you want to retain the Tivoli Management Framework in the environment, install the Tivoli NetView program using the standard Tivoli Management Framework installation. If the Tivoli Management Framework was installed to only support the Tivoli NetView program and is no longer wanted, complete the following steps:
When you attempt to run the nvColToSQL command, you might get the following error message:
Database Server is not present or not enough disk space for log files, temp files and rollback segment or table does not exists in the databaseThis message can occur if there are too many snmpCollect files to process (usually more than 500). If you get this message, you can use the following script to process the files. This script processes each file one at a time and saves the output to the log file.
#!/bin/sh LOG=/usr/OV/log/nvColToSQL.log date > $LOG # CREATE NEW LOG EACH TIME # Loop all files that do not have ! or archive in name for file in `ls /usr/OV/databases/snmpCollect | grep -v "\!" | grep -v "archive"` donvColToSQL -f $file -m 64 >> $LOG 2>> $LOG RC=$ ? echo "* FILE <$file> Return Code <$RC>" >> $LOGdone
The help for the SNMPV2 command on the Tivoli NetView for Windows version 7.1 program, found by opening the Help menu item, selecting Help Topics, selecting the Find tab, entering snmp, selecting SNMPV2 from the list, and selecting Tools -> MIB -> SNMPV2 -> Load Command is no longer valid and should be ignored.
Due to a quirk in the Linux operating system and the formatting of the Tivoli NetView for UNIX man pages, the man pages do not display properly on Linux systems unless the Linux terminal window is greater than 80 characters wide. To correct the formatting, enlarge the terminal window by using your mouse to drag the right border of the window farther to the right.
In the Tivoli NetView for Windows program, admin down interfaces (status USER1) can incorrectly change to a critical status when the interface is unmanaged and then managed again due to the use of the ICMP ping by the netmon daemon. The admin down status of the interface is not picked up again until the next configuration check or demand poll. To prevent the admin down interfaces from being reported as critical in this situation, configure the router with the admin down interfaces to use SnmpStatus poll instead of ICMP poll in the netmon seed file.
On the Tivoli NetView for Windows program on Windows 2003, attempting to send a message to a client or server (Server Setup -> Client/Server -> Send Message tab) might result in a pop-up with a "NO MESSAGE" error. This might be due to the Windows Messenger service not being enabled. The default is to leave the service disabled after installation on Windows 2003. To enable the Messenger Service, complete the following steps:
When a SmartSet submap is opened in the Tivoli NetView for UNIX program, the collmap application overrides any user-customized labels for generic symbols with the selection name for the object. A workaround is to configure and use a nongeneric symbol for the object.
On some Linux platforms, the operating system might not properly shut down some of the threads of the Tivoli NetView for UNIX Java daemons, occasionally resulting in problems restarting the NetView daemons. A "force" option has been added to the nvKill command to shut down all rogue Tivoli NetView daemons and threads and clear out the sockets. This enables the daemons to be restarted properly if this situation arises. If the Tivoli NetView daemons appear to be hung (the restart has lasted longer than 10 minutes), enter the following command at the command prompt:
Then run the netnmrc or the netview command to restart the daemons in the usual manner.
If a Tivoli NetView for Windows installation was migrated from a previous version, then the NetView.exe executable locks up when using "Find by Other Properties" under the simple tab of the find dialog due to the /usr/ov/conf/c/finddialog.conf file having a <CR><LF> changed to just a <CR>.
To fix the problem, load the finddialog.conf file into the Windows Wordpad program and then save the file. WordPad changes a <CR> to <CR><LF>. Note: The Windows Notepad program does not work.
When installing and backing out the Tivoli NetView for 7.1.4 Fix Pack 03 on an installation of the Tivoli NetView program with the Japanese language pack on a Solaris system, the NetView GUI failed to start with an error message of "Error retrieving mapdb field info: Null field value." To recover from this error when it occurs, perform the following steps:
The Windows and AIX Web Consoles do not connect to a Solaris NetView server when double-byte characters are used in the user ID. However, a Solaris 8 Web Console can connect to a Solaris NetView server when double-byte characters are used in the user ID.
When using the Japanese language pack, the default value of a map snapshot name (such as, Snapshot 1), is not displayed in the snapshot menu of the Server Setup Window. The created snapshot is displayed when you select File -> Map Snapshot -> Open. The snapshot can also be viewed when you run the ovmapsnap command.
Tivoli NetView for Windows users should implement the new overridestatus_user1.rs rule set, instead of the setstatusseverity.rs rule set, which is broken. The setstatusseverity.rs rule set puts the correlation daemon into a loop, continually sending bogus status change events to the trapd daemon, until both the correlation daemon and the Tivoli Enterprise Console adapter daemon are overwhelmed. This rule set is removed from the product in a future release, but it cannot be removed in a maintenance release.
Note: When the fix pack has been installed, a user can refer to the new \usr\ov\doc\Rulesets_on_Windows.readme for more information on rule sets.
Updated Tivoli NetView version 7.1.4 documents can be found at the following Web site:
If you have a problem with any Tivoli product, refer to the following IBM Software Support Web site: http://www.ibm.com/software/sysmgmt/products/support
If you want to contact IBM Software Support, refer to the IBM Software Support Guide at the following Web site: http://techsupport.services.ibm.com/guides/handbook.html
The guide provides information about how to contact IBM Software Support, depending on the severity of your problem, and it provides the following information:
This information was developed for products and services offered in the U.S.A. IBM might not offer the products, services, or features discussed in this document in other countries. Consult your local IBM representative for information on the products and services currently available in your area. Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM product, program, or service might be used. Any functionally equivalent product, program, or service that does not infringe any IBM intellectual property right may be used instead. However, it is the user's responsibility to evaluate and verify the operation of any non-IBM product, program, or service.
IBM may have patents or pending patent applications covering subject matter described in this document. The furnishing of this document does not give you any license to these patents. You can send license inquiries, in writing, to:
IBM Director of Licensing
IBM Corporation
North Castle Drive
Armonk, NY 10504-1785 U.S.A.
For license inquiries regarding double-byte (DBCS) information, contact the IBM Intellectual Property Department in your country or send inquiries, in writing, to:I
IBM World Trade Asia Corporation
Licensing2-31 Roppongi 3-chome, Minato-ku
Tokyo 106, Japan
The following paragraph does not apply to the United Kingdom or any other country where such provisions are inconsistent with local law:
INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Some states do not allow disclaimer of express or implied warranties in certain transactions, therefore, this statement might not apply to you.
This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. IBM may make improvements and/or changes in the products and/or the programs described in this publication at any time without notice.
Any references in this information to non-IBM Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the materials for this IBM product and use of those Web sites is at your own risk.
IBM may use or distribute any of the information you supply in any way it believes appropriate without incurring any obligation to you.
Licensees of this program who wish to have information about it for the purpose of enabling: (i) the exchange of information between independently created programs and other programs (including this one) and (ii) the mutual use of the information which has been exchanged, should contact:
IBM Corporation
2Z4A/10111400 Burnet Road
Austin, TX 78758 U.S.A.
Such information may be available, subject to appropriate terms and conditions, including in some cases payment of a fee.
The licensed program described in this document and all licensed material available for it are provided by IBM under terms of the IBM Customer Agreement, IBM International Program License Agreement or any equivalent agreement between us.
Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products.
All statements regarding IBM's future direction or intent are subject to change or withdrawal without notice, and represent goals and objectives only.
IBM, AIX, S/390, NetView, zSeries, Tivoli, and Tivoli Enterprise Console are trademarks of International Business Machines Corporation in the United States, other countries or both.
Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both.
Java and all Java-based trademarks and logos are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both.
UNIX is a registered trademark of The Open Group in the United States and other countries.
Linux is a trademark of Linus Torvalds in the United States, other countries, or both.
Intel, Intel Inside (logos), MMX, and Pentium are trademarks of Intel Corporation in the United States, other countries, or both.
Other company, product, and service names may be trademarks or service marks of others.