Guest

Cisco Element Manager System

Cisco EMF Deployment Release Notes, Release 3.2 Patch 1

  • Viewing Options

  • PDF (446.0 KB)
  • Feedback
Deployment Release Notes for Cisco Element Management Framework v3.2, Patch 1

Table Of Contents

Deployment Release Notes for Cisco Element Management Framework v3.2, Patch 1

Summary of Patch

Other Bugs

Compatibility

Additional Notes

Known Issues With This Patch Release

Solaris Issues Affecting Cisco EMF

Bugs Fixed in Cisco EMF 3.2 Patch 1

Bulk File Transfer

Other Bugs

Patch Installation

System Requirements

Installing the Patch

Deinstalling the Patch

Documentation Updates

Upgrading Cisco EMF v3.0.4 or v3.1 and any installed EMs from Solaris, Release 2.6, to Cisco EMF v3.2 on Solaris 8

Using a Backup Machine

Using One Machine

Subnet Masks in the Auto Discovery Process

Installing and Deinstalling Element Managers - Close All cemf sessions

Problems When Starting Cisco EMF - Feedback

alarmDeleter Process Usage Guidelines

AgeOfAlarmsInDays/Hours/Minutes

Viewing and Opening Object Configuration

Errors

Query Editor Severity Sorting Options

Solaris Patches

Solaris 2.6 Patches

Solaris 8 Patches

Related Documentation


Deployment Release Notes for Cisco Element Management Framework v3.2, Patch 1



Caution This patch release note details improvements made to Cisco EMF 3.2 since release. It is important to note that this is a mandatory patch and any Cisco EMF 3.2 based Element Management Systems (EMS) deployment must be made on at least this patch release, or later. Deployment of Cisco EMF 3.2 without this patch (or later patch) is strictly not supported.

If you have a Patch 1 CD then this should be installed on top of Cisco EMF v3.2. If you do not have a Patch 1 CD then you can download Patch 1 from the following link http://www.cisco.com/kobayashi/sw-center/netmgmt/cems/cemf-sp-planner.shtml


Note If Cisco EMF 3.2 Patch 1 FCS version is being installed on a system which previously had the early field trial version installed (or any previous Cisco EMF 3.2 Patch 1 version installed) then this earlier version of Cisco EMF 3.2 and Patch 1 must be manually deinstalled. To deinstall Cisco EMF, refer to http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/install/deinstal.htm. Refer to Deinstalling the Patch for Patch deinstallation instructions. After successful manual deinstallation of this previous version, the FCS version can be installed.

If Cisco EMF 3.2 Patch 1 is being installed on a clean system then this special note does not apply.


These Release Notes include:

Summary of Patch

Compatibility

Additional Notes

Known Issues With This Patch Release

Solaris Issues Affecting Cisco EMF

Bugs Fixed in Cisco EMF 3.2 Patch 1

Bulk File Transfer

Other Bugs

Patch Installation

Installing the Patch

Deinstalling the Patch

Documentation Updates

Related Documentation

Summary of Patch

This patch includes a range of improvements in quality and performance, including:

Fixes to enable Cisco EMF 3.1 EMS binary compatibly - that is, to enable Cisco EMF EMSs built, released and supported on Cisco EMF 3.1/Solaris 2.6 can be run on Cisco EMF 3.2/Solaris 2.6.

Fixes to improve the the Bulk File Transfer method (as used by some EMSs) of information retrieval from network elements

A number of installation and startup issues, including misleading error messages, have been resolved.

Alarm propagation issues related to re-parented objects have been resolved.

Other Bugs

This patch has been issued to resolve other problems as detailed in Bugs Fixed in Cisco EMF 3.2 Patch 1.

This patch contains both server and client patches and options are available for installing each.

Compatibility

The following products can be installed and operated with this release:

1. Cisco EMF 3.2

2. Cisco EMF SDK 1.3.1

3. Cisco EMF CORBA Gateway Developer Toolkit v2.0


Note Cisco EMF is supported on Solaris 2.6 and Solaris 8.


This patch is a "jumbo" patch.

Additional Notes

Instructions for installing this patch are detailed in Patch Installation.

If you have any problems carrying out this upgrade, please contact our Technical Assistance Centre (TAC) on http://www.cisco.com/public/support/tac/home.shtml

Refer to Solaris 2.6 Patches and Solaris 8 Patches for lists of Solaris patches against which Cisco EMF 3.2. Patch 1 has been tested and is approved for use.

Known Issues With This Patch Release

Solaris Issues Affecting Cisco EMF

Cisco has found a Solaris related performance issue that impacts the overall performance of Cisco EMF v3.2 EMSs running on Solaris 8. A bug in the Solaris 8 virtual memory management results in a significant performance degradation when running Solaris 6 applications in compatibility mode. Sun Microsystems Inc are currently working on a patch for this problem. In the meantime, Sun have recommended the following workaround.

To avoid any degradation in performance when Cisco EMF v3.2 is used on Solaris 8, the superuser (su) must edit the /etc/system file as follows:


Step 1 Login as superuser (su).

Step 2 Add the following line to the /etc/system file
set free_pages = 0


Note This line can be added anywhere in the /etc/system file.


Step 3 You must reboot your system, enter:
reboot -- -r


Note This change forces Solaris 8 to use the memory management model used in Solaris 6, thus avoiding the performance issue bug. Sun Microsystems Inc will resolve this bug in the next Kernel Patch.



Bugs Fixed in Cisco EMF 3.2 Patch 1

Bulk File Transfer

The following DDTs are fixed with Cisco EMF 3.2 Patch 1 and relate to Bulk File Transfer Functionality.


Note The following IOS images have been successful in polling devices for Bulk File Transfer:
c5800-p4-mz-v122_2_xb_throttle.1.1.0
c5850-p9-mz-v122_2_xb_throttle.1.4.0
c5400-js-mz.122-2.XB_0735
c5300-js-mz.122-2.XB


CSCdu68514

BulkDownload class doesnt provide an API to change remote filename with the RNE as usual

CSCdv16261

infrastructureCEMF Bulk File enhancments to 3.2 with the RBE as usual

Other Bugs

CSCat08697

Newly created profile should be selected if in ObjectList

Symptom: When a new profile is created from within an Element Manager dialog, the list of profiles shown in that dialog will update to show the new profile. However, the new profile will not always be automatically selected, and the user will have to manually select the new profile.

Conditions: This happens when the new profile is not the first profile in the updated list.

Workaround: The workaround is for the user to manually select the new profile, once the object list has updated to show it.

CSCat13797

participation broken handling requires fixes

Symptom: Deployment of Managed Objects fails.

Conditions: If one of the participants in the deployment cycle dissapears or fails to respond, then the deployment cycle is aborted immediately.

Workaround: This is no workaround at present.

CSCat13799

termination of initiators needs handled better

Symptom: The deployment of Managed Objects fails.

Conditions: If the deployment initiator dissapears or stops responding after the deployment cycle has started, the deployment is aborted immediately.

Workaround: There is no workaround at present.

CSCds44704

Map viewer icons overlap

Symptom: A map icon overlaps an existing icon.

Conditions: This happens when objects are deleted and then recreated.

Workaround: For a transient map, close the tree view and re-open it or select another view and then re-select the original tree view.

For a persistent map (i.e. you have edited the map), edit the map and move the overlapping object to another location in the map.

CSCdu22747

coordinator has its own logging level

Symptom: The Cisco EMF coordinator.log file appears to contain insufficient logging information

Conditions: A user has changed the logging level in <CEMFROOT>/config/init/loggercommon.include.

Workaround: The logging level should also be changed in <CEMFROOT>/config/init/coordinator.ini.

CSCdu33387

nbr. of children incorrect on BAMS object after rediscovery

Symptom: When new objects have been created in the Map Viewer sometimes the parent object will display an incorrect count for the number of children it has, in the hierarchy.

Conditions: Common when the parent object is not expanded in the hierarchy.

Workaround: Expand the parent object.

CSCdu51912

Changing ip and hostname on CEMF server

Symptom: After the hostname of the CEMF workstation is changed and the update -M command is run, the CGW daemons do not startup.

Conditions: This happens after the user changes the hostname of the CEMF workstation

Workaround: The user needs to rerun the Orbix2000 configuration.

As root from a CEMF shell run:

$BASEDIR/Orbix2000/orbix_art/1.2/bin/sc42/configure -no_prompt -nolog -useprep $BASEDIR/Orbix2000/prep/cemf_orbix2000.prep

where $BASEDIR is the directory CEMF is installed

CSCdu51972

Sub-dialogs launch with incorrect list population

Symptom: Child dialog launches with previous object list's objects.

Conditions: This happens when the user changes the selection on a parent dialog's object list prior to launching the child dialog.

Workaround: Do not change the selection on the parent dialog prior to launching the child dialog.

CSCdu58089

Alarm sync not clearing some alarms

Symptom: The Event Browser Application may occasionally show "No Name Available" instead of an Object Name, for Events that exist in the system against valid objects (i.e. Objects which are not orphaned).

Conditions: This situation occurs with the use of Clear Correlation when an Event Browser is first opened, and it has never displayed an event against an object.

If that object has an Event raised and Cleared by Clear Correlation in quick succession, and an Event is then subsequently raised against that object, then the Event Browser will fail to display a name for that object.

Workaround: Refreshing the Event Browser Application will display all Events correctly.

CSCdu63213

1st time launch of EvBrw from LaunchPad with CDM too long

Symptom: The first time the user launches the Event Browser from the Launch Pad, it may take at least 30 seconds for the query editor to display.

Conditions: The length to time it takes the query editor to display depends upon the internal complexity of the element manager you are using, and can vary between element managers. This only affects the first time launch, subsequent launching of the Event Browser, will result in it being displayed in the region of 5 seconds.

Workaround: There is currently no workaround.

CSCdu68514

BulkDownload class doesnt provide an API to change remote filena

Symptom: The user does not have the ability to change the destination filename of an ftp transfer once a bulk downloader has been triggered.

Conditions: This will occur under any condition as the functionality is not provided.

Workaround: No workaround currently identified.

CSCdu70783

All Loch Ness calls fail due to missing expect

Symptom: Calls to the Async IOS data repository fail.

Conditions: Any call which will attempt communication with a device.

Workaround: Manually install the "tcl" and "expect" packages.

CSCdv13051

Closing rename dialog before rename complete causes map exit

Symptom: Map Viewer exits unexpectedly when the Rename Object dialog which can be opened by accessing the Tools->View Manipulation->Rename Object ... menu is closed before the rename operation has completed.

Conditions: Select the Rename Object menu option from the View Manipulation Menu. Perform an operation that loads the CEMF manager server (e.g. Commissioning a number of objects). Change the name of the object in the Rename Object Dialog then quickly press Apply and then Close. The Map Viewer will exit unexpectedly.

Workaround: When the Apply button has been pressed, wait until the Map Viewer updates with the renamed object before selecting close.

CSCdv16261

CEMF Bulk File enhancments to 3.2

Symptom: Current bulk file transfer mechanism does not support the added fuctionality that is provided in the new CISCO-BULK-FILE-MIB and CISCO-FTP-CLIENT-MIB mibs. Effectively this means that it does not provide support for incremental downloads and notifications.

Conditions: All previous download functionality will work with the new mibs however there are no conditions under which incremental downloads and notifications will work.

Workaround: No workaround currently identified.

CSCdv31848

EM screen closed rapidly after opening may cause process to exit

Symptom: If the user closes an element manager screen very quickly after he opens it, then the process may unexpectedly exit.

Conditions: This will only occur if it is the first time the element manager screen is opened after it is installed. In subsequent Cisco EMF sessions, the dialog can be closed quickly and there will be no adverse affect.

Workaround: No workaround currently identified.

CSCdv46283

BULK FILE fails after CEMF reset

Symptom: Bulk File Transfer modules fail.

Conditions: This will occur on a "cemf start" following a "cemf reset" command.

Workaround: There are 2 possible workarounds, as outlined below. Either one may be used, but not both.

1. After installation of the patch, perform the following command:

mv <CEMF_ROOT>/config/objectType/* <CEMF_ROOT>/config/objectTypes/

Once this command has been issued, all future cemf reset and start cycles will work as expected.

2. After performing "cemf reset" and "cemf start" commands, performing the following command:

<CEMF_ROOT>/bin/objectTypeFileParser <CEMF_ROOT>/config/objectType/S12bulkFTP.types

This command will have to be issued after every cemf reset and start cycle.

CSCdv47775

Bulk File support for redundant data fails with mixed table types

Symptom: Bulk file transfer from an snmp agent fails.

Conditions: This problem will occur when both a scalar or lexicaltype and a nexttype are retrieved from the agent at the same time.

Workaround: No workaround currently identified.

CSCdv74389

Process corbaActionLauncher hangs on CPU

Symptom: Corba process corbaActionLauncher hang on CPU. No corba command will work without killing the process first.

Conditions: A unix work station that has been already commissioned or decommissioned is commissioned or decommissioned again.

Workaround: Kill the process corbaActionLauncher.

CSCuk23650

Coordinator should handle an initiator dying thats a participant

Symptom: Deployment of objects fails and a "participation broken" error is produced.

Conditions: The process that initiated the deployment exits unexpectedly, and is also involved in the creation of objects.

Workaround: There is no workaround at present.

CSCuk24022

Alarm propagation is wrong in mapViewer after reparenting objects

Symptom: In the map viewer application, the alarm propagation circle is grey when it should be green or it it is green when it should be grey.

Conditions: This problem will occur when objects are reparented either from a propagating view to a non-propagating view or from a non-propagating view to a propagating view.

Workaround: The problem can be resolved by reopening the map viewer application.

CSCuk24394

license file wont parse if command is run within cemf shell

Symptom: When attempting to parse in the Cisco EMF license file from a UNIX command-line session in which the 'cemf shell' command has been executed, an error is produced and the license file is not parsed.

Conditions: This occurs when trying to parse in a license file from within a cemf shell using the 'cemf license' command.

Workaround: A workaround to the problem is to parse in the license file outwith a cemf shell.

e.g. Open a new terminal

cd <CEMFROOT>/bin

Become root

Run the command 'cemf license -file <full path to license file>'

CSCuk24605

Modal dialogs end up behind parent window

Symptom: Sometimes a dialog will appear to have frozen with the busy cursor displayed, as if the dialog is waiting to do something.

Conditions: This occurs when a dialog has opened a further dialog and become busy itself until the new dialog is dealt with. It is possible that this new dialog can be hidden behind the original parent dialog giving the impression that the original dialog is frozen, whereas it is actually waiting for you to do something with the (hidden) new dialog.

For Example, launch the EventBrowser from the Launchpad. EventBrowser opens with QueryEditor on top. Click on the title bar of the EventBrowser window. The QueryEditor goes behind the EventBrowser and the EventBrowser remains busied on.

Workaround: Minimize or move the busy dialog, and the hidden dialog will be visible.

CSCuk24656

Objectgui core dump on object deployment

Symptom: The Map Viewer exits unexpectedly.

Conditions: This may happen when deploying objects.

Workaround: There is currently no workaround.

CSCuk24753

Race conditions in ObjectGui. DA vs Events. Grey smartie.

Symptom: The mapViewer displays a grey propagated status and no plus sign to indicate that a tree (object) view has children.

Conditions: This problem happens when the system is very busy and the children of a tree view are being changed rapidly, whilst the tree view is being expanded at the same time. For example, when the system is very busy and objects are being deployed under a particular tree view whilst that view is being expanded.

Workaround: Close and re-open the viewer.

CSCuk25543

OG fails to be (re-)populated because query times-out,but hangs

Symptom: When attempting to create an object group based upon an attribute value over a large number of objects, the object group may incorrectly be empty.

Conditions: This occurs when the query that is used by object group is based upon the value of an attribute, and is over an object scope containing a very large number of objects.

Workaround: No workaround currently identified.

CSCuk26358

Containment isLocked calls shouldnt go to appsServer

Symptom: After restoring a backup, objects appear with different names or under different parents.

Conditions: If while performing a backup, the user attempts to rename or reparent objects, these changes may potentially be unsaved.

Workaround: No workaround currently identified.

CSCuk26432

Popup menu not available for ObjectGroup on Client machine

Symptom: Sometimes when right-clicking on the objectGroups item within Object Group Manager the resultant menu is empty, and only displays the menu title. "Create Object Group" and "Create Folder" are expected to appear, but they do not.

Conditions: This can happen on an installation where there are several Element Managers installed, and relates to timing. It can occur on both CEMF client and manager server installations, but may be more likely on a client.

Workaround: The following steps should be followed to work around this problem -

1) Close your current session

2) Start a new session, and open the MapViewer

3) From any object within the MapViewer choose the Get object group membership option from the Tools menu.

4) You can now start the Object Group Manager from the launchpad (the MapViewer can be closed if no longer required.)

CSCuk26567

Notifications database corruption

Symptom: The notificationServer process will exit with an ObjectStore exception saying that it was out of PSR.

The list of pending scripts will be created in and remain in the wrong segment.

Conditions: Lots of notification profiles are setup to run scripts.

The list of pending scripts will always be created in the wrong segment.

Workaround: The workaround is to delete some notification profiles. There is no workaround for the list of pending scripts being in the wrong segment.

CSCuk26632

corba event channel will not work after first time start

Symptom: No events are received on Corba event channels.

Conditions: It is the first time the Cisco EMF server has been started, using 'cemf start'.

Workaround: After first time start-up of the Cisco EMF server has completed, stop and restart the CEMF server by running 'cemf stop; cemf start'.

CSCuk26637

Licensing problems

Symptom: When an @ symbol exists in the LM_LICENSE_FILE environment variable the following error message occurs:

CEMF Manager License Installer...

CEMF Manager is configured for a remote license.

The license installer cannot be used.

when attempting to parse in a local license file using the following command :

[CEMFROOT]/bin/cemf license -f <license_file_pathname>

Conditions: This occurs when the LM_LICENSE_FILE environment variable is set up to point at a remote license (i.e. @ exists in variable ), however the user is attempting to use a local license file which requires to be parsed in.

Workaround: Explicitly set up the value of the LM_LICENSE_FILE environment variable using the unix 'setenv' command prior to attempting to parse the local license file. {Note: This should be performed from within a cemf shell}

CSCuk26662

Chassis Deployment: Objects fail to deploy in physical view

Symptom: Objects are not deployed under the Physical view. Only the parent object is displayed in the Physical view.

Conditions: This occurs when an object is deployed either manually or automatically with autodiscovery.

Workaround: Expand the parent object in a different view in order to see the child in the Physical view.

OR

Close the MapViewer and re-launch it.

CSCuk26892

SNMP has FMR error, which could cause SNMP request failures

Symptom: The Cisco EMF process 'asyncSnmpDataRepository' may not fetch a table attribute correctly from a device, possibly resulting in it exiting unexpectedly.

Conditions: An SNMP table is being retreived.

Workaround: There is currently no workaround.

CSCuk26967

EM screen exit if screen closed before object list population end

Symptom: Some EM dialogs that use Object Lists which are populated when the Dialog is launched may exit unexpectedly when that particular dialog is closed.

Conditions: This occurs when a user launches a dialog and then immediately closes it by double clicking on the top left of the Window, before the Dialog has been fully populated.

Workaround: There is currently no workaround.

CSCuk27081

updateName reports cant find directory error

Symptom: The following error message is displayed after a 'cemf restore' operation:-

Can't open <CEMFROOT>/config/env/licenses/*.lic : No such file or directory

Conditions: During a 'restore' operation of databases from one system to another system, where the following conditions exist:-

the hostname of the backed-up system is different from the system to which the databases are being restored.

- a remote license server is being used (this is only configurable at CEMF installation).

Workaround: The message can be safely ignored.

CSCuk27174

PollerServerHelper does not suppress events

$$IGNORE

This is an internal message between subsystems, and does not directly affect standard behaviour.

CSCuk27422

Backwards Compatibility issue found with ATM QoS Profiles

Problem: Attempting to apply a profile can fail, because the underlying system cannot retrieve the profile.

Symptom: This occurs when code built against CEMF version 3.1 is rerun against CEMF version 3.2 (Odin). The ProfileMonitor class contains virtual methods which will not be called properly unless code is rebuilt.

Workaround: The workaround is to recompile any code which uses the ProfileMonitor API.

CSCuk27436

Backwards Compatibility isue found with ATM Upload

Symptom: When an Element Manager which was compiled for Cisco EMF 3.1 is run on a Cisco EMF 3.2 system, query-related functionality may not appear to give the correct results.

Conditions: A query is being performed which checks for the presence of a particular attribute or attributes against one or more objects.

Workaround: If the Element Manager is recompiled for Cisco EMF 3.2, this problem will not occur.

CSCuk27942

BulkDownloaderCommonImpl::addObjectIdList is broken

Symptom: When calling the method BulkDownloaderInterface::triggerDownload to perform a Bulk File Transfer the download fails with the following ERROR in the log file:

ERROR: Invalid GetType specified in BulkDownloadSpec for object <objId>, The object will not be added for download.

Conditions: This can happen when a list of object ids have been added to the downloader using the method BulkDownloaderInterface::addObjectIdList and these objects have not subsequently had a BulkDownloadSpec applied to them using the BulkDownloaderIncremental:: applyDownloadSpecs method.

Workaround: A code based workaround is to add each object to the downloader seperately using the BulkDownloaderInterface::addObject method, rather than adding the whole list at once.

CSCuk27948

BulkDownloader may not delete all files on destruction

Symptom: When a user creates a BulkDownloader or BulkDownloaderIncremental object and requests that the destination files are removed on destruction some files are left behind and not removed.

Conditions: This happens if one or more files are transfered and then the destination file name is changed to be something else.

Workaround: The user can remove the files manually once the BulkDownloader object has been deleted.

CSCuk27952

Certain operations fail in downloadComplete

Symptom: Performing certain BulkDownloader operations while processing the BulkDownloaderUser:: downloadComplete callback returns failure with a log message indicating that the operation cannot be performed while a download is in progress, even though the download has completed.

Conditions: This can happen when BulkDownloaderUser::downloadComplete is called with a failure status.

Workaround: A code based workaround is to return from the BulkDownloaderUser::downloadComplete call before performing the operation in question, i.e perform the operation on a mainloop callback.

CSCuk28767

coordinator core-dump because commshandle ptr used after its del.

Symptom: In Cisco EMF object creations or deletions will fail. There will be a 'core' file in the <CEMF_ROOT>/bin/.coordinator.sysmgr directory.

Conditions: Under some circumstances when object creations or deletions take place.

Workaround: Currently there is no workaround.

CSCuk29337

No security on controller operations (action part)

Symptom: A user can perform operations on their machine that should only be available to users with root permission, even if that user doesn't have root permission.

Conditions: This occurs when the user creates actions that perform root operations.

Workaround: There is currently no workaround.

Patch Installation

The Patch installation instructions detail how to install and deinstall a patch to a Cisco EMF Server or a Cisco EMF Client.

System Requirements

Refer to Compatibility for details of products which must be installed and operated with this release.


Note You will need 250MB of free disk space available to install a Cisco EMF patch. The patch will not install if there is insufficient disk space for the installation.


Installing the Patch

The patch must be installed for the Server and all Clients on the network.


Note Backups must be made of all databases. Refer to the Cisco EMF Installation Instructions for complete details.


To install the patch, proceed as follows:


Step 1 Login as the superuser.

Step 2 Insert the Cisco EMF patch CD into the CD-ROM device.

Step 3 Change directory into the cdrom area. At the command line prompt, enter:

cd /cdrom/cdrom0

Step 4 Run the install script ./cemfinstall to start the patch installation. The Install Package screen is displayed, for example when a Cisco EMF Server is installed, the following menu is displayed:

INSTALL PACKAGE
===============
1) CEMF Server 3.2 Patch 170001-18
2) CEMF Client 3.2 Patch 170101-18 - Not Installed "CSCOcemfc"
q) Quit
Which package do you wish to install? (Def:1) [?,q]

Note The patch number identifies the type and version of the patch.


The first two digits indicate the product, for example Patch 170001-18 is for Cisco EMF.

The second pair of digits indicate that the product is for a Server (00) or a Client (01).

The third pair of digits identifies the patch number, for example 01 is Patch 1. The final two digits identify the build number of the patch, for example 18 shows this is the eighteenth build of a patch.

Step 5 Select the relevant package to install.


Note If you have a Cisco EMF Server installation you are only able to install the Server patch. If you have a Cisco EMF Client installation you are only able to install the Client patch.


The installation process will now start. Cisco EMF processes will be stopped automatically prior to the patch installation starting. You will see a message which tells you that the Cisco EMF Patch installation was completed satisfactorily.

Step 6 Once the installation is complete, cd out of the directory /cdrom/cdrom0 and enter eject.

Step 7 Remove the Cisco EMF product CD from the CD-ROM drive.

Step 8 Start Cisco EMF. From the command line, type

cd <CEMF_ROOT>/bin/cemf start

where <CEMF_ROOT> is the directory in which Cisco EMF is installed. Cisco EMF processes start.


Deinstalling the Patch

The patch must be deinstalled from the Server and all Clients on the network.

To deinstall the patch, proceed as follows:


Step 1 Log onto the machine with Cisco EMF installed.

Step 2 Log in as the superuser.

Step 3 Go to <CEMF_ROOT>/bin where <CEMF_ROOT> is the Cisco EMF root directory.

Step 4 Deinstall Cisco EMF as follows:

<CEMF_ROOT>/bin/cemf remove

Step 5 The De-Install Package screen is displayed, for example where the Cisco EMF Server is installed and the patch number is 170005-21, the following menu is displayed:

REMOVE PACKAGE
===============
1) Cisco EMF Server Package
2) CEMF Server 3.1 Patch 170001-18
q) Quit
Which package do you wish to remove? (Def:1) [?,q]

From the displayed list select the patch that has to be deinstalled. You will be notified that the deinstallation is about to take place.

If you proceed with the deinstallation the screen will scroll through the software being deleted and you will see the following message:

Removal of <package> was successful.


Documentation Updates

This section of the Release Notes includes the following updates to the Cisco EMF documentation set:

Upgrading Cisco EMF v3.0.4 or v3.1 and any installed EMs from Solaris, Release 2.6, to Cisco EMF v3.2 on Solaris 8

Using a Backup Machine

Using One Machine

Subnet Masks in the Auto Discovery Process

Installing and Deinstalling Element Managers - Close All cemf sessions

Problems When Starting Cisco EMF - Feedback

alarmDeleter Process Usage Guidelines

Viewing and Opening Object Configuration

Errors

Upgrading Cisco EMF v3.0.4 or v3.1 and any installed EMs from Solaris, Release 2.6, to Cisco EMF v3.2 on Solaris 8

Cisco EMF v3.2 can be deployed in a number of different configurations, from a single hardware system to a multi-site management system, using Sun Solaris/SPARC hardware with Solaris, Release 2.6 or Solaris 8.

This Cisco EMF upgrade procedure can be used for various scenarios:

Upgrading from Cisco EMF 3.0.4 on Solaris 2.6


Note The minimum patch level must be Cisco EMF 3.0.4 Patch 14.


Upgrading from Cisco EMF 3.1 on Solaris 2.6


Note The minimum patch level must be Cisco EMF 3.1 Patch 4.


Multiple Element Managers are deployed on Cisco EMF v3.x on Solaris 2.6, and the system is operational

The Element Managers have been released on Cisco EMF v3.2


Note Check your Element Manager Installation Guide for EM specific upgrade information, for example database upgrades.


Using a Backup Machine

To upgrade Cisco EMF v3.x and any installed Element Managers from Solaris, Release 2.6, to Cisco EMF v3.2 on Solaris 8, it is recommended that you use a backup machine. This will avoid any disruption to management of the network, and keep management down-time to a minimum.

Proceed as follows:


Step 1 Install and configure the backup machine with Solaris 8. The Solaris 8 patches against which Cisco EMF v3.2 has been tested and is approved for use are listed in http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/notes/rn_3_2.htm

Step 2 Install Cisco EMF v3.2, refer to http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/install/install.htm

Step 3 On the backup machine, start Cisco EMF, from the command line, enter <CEMF_ROOT>/bin/cemf start where <CEMF_ROOT> is the directory in which Cisco EMF is installed.

Step 4 On the backup machine, stop Cisco EMF, from the command line prompt, enter <CEMF_ROOT>/bin/cemf stop

Step 5 On the production machine where Cisco EMF is installed on Solaris 2.6, remove the g.histCriteria file, enter rm <CEMF_ROOT>/config/selfManagement/actions/g.histCriteria

Step 6 Backup the databases, enter ./cemf backup

Step 7 Copy the upgradePackages files to a temporary location so that they can be restored on to the upgraded system before startup. Enter cp -r <CEMF_ROOT>/config/selfManagement/upgradePackages /opt/

Step 8 FTP the backed-up databases taken in Step 6 and the upgradePackages files from Step 7 from the production machine to the backup machine.

Step 9 On the backup machine, restore the databases that were transferred from the production machine, enter ./cemf restore -l <backup path> -u


Note The -u flag implements the correct required behavior by preventing the configuration files from the backup being copied across to the new installation.

You may see the message Backup system catalog file is missing. You can ignore the message and choose y.


Step 10 Remove the attribute history server databases, enter

rm <CEMF_ROOT>/db/attributeHistoryCollector.db
rm <CEMF_ROOT>/db/attributeHistoryServer.db


Note This is necessary as it is not possible to guarantee the integrity of any existing performance data.


Step 11 On the backup machine, restore the upgradePackages files copied in Step 7, enter cp /opt/upgradePackages/*<CEMF_ROOT>/config/selfManagement/upgradePackages/

Step 12 Start Cisco EMF v3.2 on the backup machine, enter ./cemf startupgrade

Step 13 Now install the v3.2 Element Managers that are installed on the production machine. For each Element Manager, enter ./cemf install <path> where <path> is the path of the Element Manager, then choose the option you require from the menu.

Step 14 Verify that the backup machine is now successfully managing the required network elements, in other words it is now a temporary production machine. If management is successful then stop the production machine.

Step 15 On the production system, deinstall the Element Managers. For each Element Manager, enter ./cemf install -remove then choose the option you require from the menu.

Step 16 On the production system, deinstall Cisco EMF, refer to http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/install/deinstal.htm

Step 17 Install and configure the production machine with Solaris 8. The Solaris 8 patches are listed in Solaris 8 Patches

Step 18 Install Cisco EMF and your Element Managers.

Step 19 Backup the databases on the backup machine, enter ./cemf backup -l <backup path>

Step 20 FTP the latest backup from the backup machine to the production machine.

Step 21 On the production machine, restore the databases that were transferred from the backup machine, either interactively (cemf restoreDataset) or manually (cemf restore). Refer to http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/install/backup.htm for full details.

Step 22 Start Cisco EMF on the production machine, enter ./cemf start

Step 23 Verify that the production machine is successfully managing the network elements again.

Step 24 Take the backup machine off-line.


Using One Machine

To upgrade Cisco EMF v3.x and any installed Element Managers from Solaris, Release 2.6, to Cisco EMF v3.2 on Solaris 8, proceed as follows:


Step 1 Login to the machine where Cisco EMF is installed as the superuser (su).

Step 2 Change directory to the Cisco EMF bin directory, enter:

cd <CEMF_ROOT>/bin

Step 3 As a precautionary measure in case there are problems with the subsequent upgrade, backup all databases, and copy these to a non-default location, using the cp command. If the upgrade is successful, this backup will not be required.

It is necessary to have sufficient disk space for the backups.

For backup instructions for the version of Cisco EMF you are upgrading from, refer to the "Cisco EMF Database Backup and Restore" chapter in the Cisco EMF v3.x Installation and Administration Guide.


Caution The directory used for this backup should not be a directory that will be over-written by the installation of Solaris 8. The backup should be on a separate disk that will be protected during Solaris 8 installation and not over-written, or in a directory on a remote file server.

Step 4 Remove the g.histCriteria file, enter:

rm <CEMF_ROOT>/config/selfManagement/actions/g.histCriteria

Step 5 To display a list of all Cisco EMF packages, enter:

./cemf install -show

This will list the version of Cisco EMF and the package type and all Element Managers and their package types.

Step 6 Deinstall the Element Managers for upgrade. For each Element Manager, enter:

./cemf install -remove -o -upgrade

then choose the option you require from the menu.

Step 7 Backup the databases, enter:

./cemf backup

For backup instructions for the version of Cisco EMF you are upgrading from, refer to the "Cisco EMF Database Backup and Restore" chapter in the Cisco EMF v3.x Installation and Administration Guide.


Caution The directory used for backup should not be a directory that will be over-written by the installation of Solaris 8. The backup should be on a separate disk that will be protected during Solaris 8 installation and not over-written, or in a directory on a remote file server.


Caution These are the backups that will be restored later.

Step 8 Copy the upgradePackages files to a temporary location so that they can be restored on to the upgraded system before startup. Enter:

cp -r <CEMF_ROOT>/config/selfManagement/upgradePackages/* /opt/upgradePackages/


Caution These files can be copied to any temporary location except /tmp/ (/opt/ is used in the example).
Do not use /tmp as the reboot ( Step 11) will remove any files in this area.

Step 9 Stop Cisco EMF, enter:

./cemf stop

Step 10 Remove the current Cisco EMF installation and Cisco EMF patches, enter:

./cemf install -remove

then choose the option you require from the menu.


Caution The order in which components are removed is important. You must deinstall the patch first, then Cisco EMF itself.

Step 11 Cisco EMF v3.x and its Element Managers are now removed, and you must reboot your system.


Note A reboot is necessary to ensure there are no processes remaining which may interfere with successful operation of Cisco EMF v3.2.


Step 12 Install and configure Solaris 8. The Solaris 8 patches against which Cisco EMF v3.2 has been tested and is approved for use are listed in http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/notes/rn_3_2.htm


Caution It is recommended that you choose Initial Install of Solaris, not Upgrade.

Step 13 Cisco EMF v3.2 can now be installed. Insert the Cisco EMF product CD into the CD-ROM device.

Step 14 Change directory into the cdrom area. At the command line prompt, enter:

cd /cdrom/cdrom0

Step 15 Install Cisco EMF v3.2, enter:

./cemfinstall

from the Cisco EMF v3.2 CD-ROM.

Step 16 Change directory to the Cisco EMF bin directory, enter:

cd <CEMF_ROOT>/bin

Step 17 Start Cisco EMF, enter:

./cemf start

Cisco EMF processes start and Cisco EMF is initialized.

Step 18 Stop Cisco EMF, enter:

./cemf stop

Step 19 Restore the databases backed up in Step 7, enter:

./cemf restore -t <date of backup> -u

The <date of backup> value is the date the backups were created. The date must always be specified in US format, that is mm-dd-yyyy.


Note The -u flag implements the correct required behavior by preventing the configuration files from the backup being copied across to the new installation.

You may see the message Backup system catalog file is missing. You can ignore the message and choose y.


Step 20 Remove the atttribute history server databases, enter:

rm <CEMF_ROOT>/db/attributeHistoryCollector.db
rm <CEMF_ROOT>/db/attributeHistoryServer.db


Note This is necessary as it is not possible to guarantee the integrity of any existing performance data.


Step 21 Restore the files copied in Step 8, enter:

cp /opt/upgradePackages/* <CEMF_ROOT>/config/selfManagement/upgradePackages/

Step 22 Start Cisco EMF v3.2, enter:

./cemf startupgrade

Step 23 Now install the v3.2 Element Managers. For each Element Manager, enter:

./cemf install <path>

where <path> is the path of the Element Manager, then choose the option you require from the menu.

Step 24 The backups created at Step 3 should only be removed after the upgraded Cisco EMF and Element Manager installation has been fully tested and verified.


Subnet Masks in the Auto Discovery Process

The Auto Discovery application allows you to examine the network for IP and SNMP devices and create a managed object for each new device discovered. Refer to http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/user/discover.htm for full details.

The device name, device address and device subnet mask are displayed in the Discovery Configuration panel on the Discover Network Devices window. The subnet mask is used to extract network and subnetwork information to refine the discovery range.

Auto Discovery between subnetworks is transparent. Internally, Cisco EMF recognizes any IP addresses that are destined for a subnetwork, and sends those packets to the gateway for that subnetwork. If a discovered object knows its own subnet mask, that is it has an entry for the home network in the ipAddrTable, then that subnet mask is used.

If the user does not specify the correct subnet mask, it is possible for a subnet to be discovered and deployed as an IP device. If a child device of that subnet is subsequently discovered then the subnet is correctly deployed as a device.


Caution If the wrong subnet mask is entered, the representation of the network in Cisco EMF will be incorrect.

Installing and Deinstalling Element Managers - Close All cemf sessions

Cisco EMF should already be installed and running on the system, before installing or deinstalling Element Managers. The procedures are detailed in the Cisco EMF v3.2 Installation and Administration Guide, http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/install/eminstal.htm


Note To prevent any loss of information, it is recommended that you close all cemf sessions before installing or deinstalling Element Managers.


The EM installation or deinstallation process forcefully closes all cemf sessions. A dialog is displayed which states that communications with the server have been lost. Any cemf session is then closed.

Problems When Starting Cisco EMF - Feedback

If you start Cisco EMF using <CEMF_ROOT>/bin/cemf start, you are now informed about any processes that fail to start. The message Failed to start core server(s) followed by the name of each process that failed to start is displayed. Examine the sysMgr.log file and the log file of the failed process to start the troubleshooting process. For example, if the message identifies that attributeHistoryServer failed to start, look at the attributeHistoryServer.log file.

Troubleshooting sections are provided in the Cisco EMF Installation and Administration Guide and the Cisco EMF User Guide. The Cisco EMF Knowledge Base and the Communications Software Group Knowledge Base are sources of additional technical information on Cisco EMF.

alarmDeleter Process Usage Guidelines

The alarmDeleter process is responsible for deleting alarms which have become redundant and whose associated data is no longer of interest to users. Refer to http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/install/adeleter.htm

The alarm database has a finite size and its size can affect performance. Removal of alarms is an essential task.

The two main criteria involved when deciding if a particular alarm should be deleted or not are:

1. the state of the alarm, or more specifically whether it is active or cleared

2. the age of the alarm, that is, how long has it been in the system

All configuration is done via the alarmDeleter's initialisation file: alarmDeleter.ini. The following example shows the relevant section from the file.

[AlarmDeleter]
databaseName= /opt/cemf/db/alarm.db
segmentDeletionInterval = 15
ageOfAlarmsInDays= 1
ageOfAlarmsInHours= 0
ageOfAlarmsInMinutes= 0
deleteAllAlarms = 0
maxDeletionChunkSize = 1000

AgeOfAlarmsInDays/Hours/Minutes

These values specify the period after which alarms are candidates to be deleted. Remember that an alarm has to satisfy both the age and the state criteria before it will be deleted.

The default value is set to delete those alarms that are in the cleared state (state criteria) for more than 1 day (age criteria). In previous Cisco EMF version this was seven days. This value assumes a daily rate of 25,000 traps/alarms, which would give a total of 50,000 alarms in the database after two days, at which point the first 25,000 cleared alarms would be deleted.

If the incoming rate is lower than 25,000 per day then the deletion age value can be increased to give a longer alarm history. This means cleared alarms will remain in the database longer. However, the new value should be set such that there will be a maximum of ~50,000 alarms in the database.

Viewing and Opening Object Configuration

All managed objects have a configuration that is particular to the network element which it represents.

You can view and then edit this configuration, refer to http://www.cisco.com/univercd/cc/td/doc/product/rtrmgmt/cemf/3_2/user/services.htm for details. The Object Configuration window includes an Object Types drop down list.


Note Long object type names in the drop down list may not be fully viewable. To view all the details, highlight the object type name and move the pointer left or right to scroll through the name.


Errors

Query Editor Severity Sorting Options

In the User Guide, p6-10, the list of severities should be as follows:

Critical

Major

Moderate

Minor

Warning

Normal

Informational

Solaris Patches

For information, note that Cisco EMF 3.2 was tested with the following Solaris patches, Solaris 2.6 Patches and Solaris 8 Patches, which were current when Cisco EMF 3.2 was released. None of these patches is known to be mandatory, though we would recommend having this or higher revisions of any recommended Solaris patch.

The current Sun recommended patch cluster can be downloaded from http://sunsolve.Sun.COM. Please note that higher revisions of the patches may introduce dependencies on other patches which are not reflected in the lists above. You should follow the patch dependencies as recommended by Sun Microsystems, Inc.

Solaris 2.6 Patches

Solaris 8 Patches

Related Documentation

Cisco Element Management Framework Link to Release Note

Cisco Element Management Framework Release Notes, 78-12543-01

Cisco Element Management Framework Installation and Administration Guide v3.2, 78-12539-01

Cisco Element Management Framework User Guide v3.2, 78-12536-01