Build number: 7.0.8641

Monitor Solution helps you ensure your server and network up-time and availability and ultimately reduce the costs of network management in your organization.
You can do the following with Monitor Solution:

• Identify the health of your environment by collecting detailed data from servers, applications, and network devices.
• Analyze trends and isolate recurring issues by collecting comprehensive real-time and historical performance data.
• Pinpoint problems, define their cause, and take automated actions to resolve them.

See the following sections for information about this release:

• New features
• Supported client operating systems
• Upgrade issues
• Fixed issues
• Things to know
• Known issues

Back to top

New features

The following are features of this release:

Back to top

Supported client operating systems

This section lists the client computer operating systems that are supported by Monitor Solution.

• AIX IBM Advanced Interactive eXecutive 5.2, 5.3, and 6.1
• HP-UX Hewlett Packard UniX 11i v1 (PA-RISC)
• HP-UX Hewlett Packard UniX 11i v2 and v3 (IA64 and PA-RISC)
• Microsoft Windows Server 2003 (32- and 64-bit) -excludes Microsoft Windows Server 2003 R2
• Microsoft Windows Server 2008 (32- and 64-bit)
• Microsoft Windows Server 2008 R2 (32- and 64-bit)
• Microsoft Windows Server 2008 Core Edition (32-bit)
• Microsoft Windows Server 2008 R2 Core Edition (32-bit)
• Red Hat Enterprise Linux 4 (32- and 64-bit)
• Red Hat Enterprise Linux 5 (32- and 64-bit)
• Red Hat Enterprise Linux 5.1 (32- and 64-bit)
• Red Hat Enterprise Linux 5.2 and 5.3 (32- and 64-bit)
• Red Hat Enterprise Linux 5.4 (32- and 64-bit)
• Solaris 9 on SPARC
• Solaris 10 on SPARC and x86
• SUSE Linux Enterprise Server 10 (32- and 64-bit)
• SUSE Linux Enterprise Server 11 (32- and 64-bit)
• VMware ESX 3.0.1, 3.0.2, 3.5

Back to top

Upgrade issues

The following are issues and things to know about upgrading Monitor Solution 7.0.

Back to top

Fixed issues

The following are previous issues that were fixed in this release. If additional information about an issue is available, click the Article ID link.

Back to top

Things to know

The following are things to know about in this release. If additional information about an issue is available, click the Article ID link.

Monitor Solution 7.0 cannot be hosted on a Windows Server 2008

Currently, the Symantec Management Platform, which Monitor Solution is installed on, does not yet support Windows Server 2008 as a host; this support will be added in a future release.

The Monitor Agent plug-in supports Windows 64-bit platforms

The Monitor Agent includes 64-bit support for both Windows 2003 and Windows 2008 as a managed device.

You may need to install sysstat on your monitored Linux computers

You must install sysstat on your targeted Linux computers to use the following monitor policies:

• Processor
• Disk I/O
• Memory

If you install the Remote Monitoring Server on a dedicated computer, the PPA agent and the Credential Manager are required

If you install the Remote Monitoring Server on a dedicated computer, the PPA agent and the Credential Manager are required.

Network Discovery found devices may not be listed in the targeted filters of the default agentless policy

Agentless monitor policies target filters of computers and devices. By default, the target only included unmanaged devices running Windows 2003 or 2008. When computers are discovered, the local operating system is a property of that computer resource. If during discovery, a computer's operating system cannot be determined, an operating system specific filter will not recognize that computer.

This is a possible cause if an agentless monitor policy is not running on a computer as expected.
Note: The operating system version is correctly discovered for target computers that have SNMP enabled.

Activated Monitor Policies web part only shows policies that the Monitor Agent has activated

A Monitor Agent must first activate a monitor policy before it is displayed in the Activated Monitor Policies web part of the Monitoring and Alerting home page. If you have enabled a monitor policy but it is not yet displayed in the web part, this may be because the web part is populated from inventory data that is received from the agent.

Purging Log Event and Sys Log data

Both Log Event and Sys Log data purging is controlled with the single String metric data purging schedule.
To schedule string metric data purging:

1. In the Symantec Management Console, on the Home menu, click Monitoring and Alerting.
2. In the left pane, click Monitoring and Alerting > Monitor > Settings > Monitor Server Settings.
3. On the Monitor Server Settings page, click the Purge Maintenance tab.
4. Under Non-numeric Data, set the value for String metric data.
5. On the Monitor Server Settings page, click Save changes.

The Poll Metric on Demand task cannot poll agent-based metrics on a computer that does not have the Monitor Agent

The Poll Metric on Demand task can poll two different types of metrics: agent-based and agentless. It is supported on two different types of targets: with and without Monitor Agent.

If the Poll Metric on Demand Task is executed on an agentless target (the client computer without the Monitor Agent installed on it), then it can only poll agentless metrics on the client computer. Agent-based metrics are not polled on an agentless client computer as agent-based metrics require the Monitor Agent to run.

A failed heartbeat may not raise an alert if the Check for heartbeats interval is less than the Send Heartbeat interval with zero retry attempts

If you configure the Monitor Server Setting's > Heartbeat tab's > Retry attempts value to "0" and the Check for heartbeats every interval to be less than or equal to the Monitor Agent Configuration Setting's Send heartbeat every interval, then a failed heartbeat may not raise an alert in the Event Console due to network latency. In this case, a Monitor Agent may appear to occasionally go down only to come right back up again. However, the uptime data will not be affected in this case.

There are time drifting issues when monitoring heartbeats on Linux computers that are hosted on VMware

When monitoring heartbeats on Linux computers that are hosted on VMware, there are issues with time drifting.
This issue can cause heartbeats to be received by Notification Server after a very long delay and an alert to be raised in the Event Console.
To avoid issues with unnecessary alerts in the Event Console, make sure that your heartbeat time settings are synchronized to account for possible time drifting.

HTTP metrics do not support virtual hosts

The HTTP metric cannot be used with virtual hosting. The HTTP metric only supports the use of the current hostname of a physical computer that reports basic inventory to the Notification Server computer.

There is a specific setup required to enable Smart Metrics using SNMP

To enable Smart Metrics using SNMP:

• The SNMP service must be installed on the monitored computer.
• The community name must be configured in the SNMP service on the monitored computer. The community name must match the community string in the SNMP protocol settings for your connection profile.
• SNMP packets must be accepted from any host in the SNMP properties on the monitored computer.

If two WMI metrics are configured to parse the same command output and they both have different polling intervals, these metrics will read data with the lowest polling interval. The WMI metrics combine into a similar query for optimization. The query runs at the lowest interval setting.

When "Altiris" log type is selected, the Log Event metric only supports Windows platforms

When the "Altiris" log type is selected, the Log Event metric uses the Notification Server log file as source. This type of log is supported for Windows platforms only.

The reports "Alert History" and "Most Active Hosts" are located in the Event Console Reports folder

The 6.x reports:

• Alert History by Monitor Agent
• Alerts Over Last 'N' Days
• Monitor Agents with Most Alerts

Are now replaced with:

• Alert History
• Most Active Hosts

These reports are located under the Event Console Reports folder on the Monitoring and Alerting portal page.

There is a delay in the Computers Lists for the Real-time and Historical Performance Viewers

Computers are not populated in the Computers List for the Historical and Performance Viewers until data is received from the target computer by Notification Server.

Metrics do not work if Credential Manager is installed by pull to Remote Monitor Server

Some metrics do not work if Credential Manager is installed by a pull to the Remote Monitor Server. If Credential Manager is pushed to the Remote Monitor Server, there are no problems.

The repeat count feature specifies the number of times the rule must trigger in a row before an action is taken

After the repeat count specified in a rule has been reached, the rule resets to "normal" on the next value, even if that value crosses the threshold. If the value crosses the threshold, the rule restarts the repeat count for the next evaluation.

Example:

Rule count definition - if CPU > 10 for 3 times

If the values are 11, 12 , 13, 11, the rule will trigger on the third value and then reset to normal on the fourth value. The fourth value then becomes the first increment for the next check.

Agentless monitoring uses Pluggable Protocol Architecture (PPA) connection profiles

The Remote Monitoring Server (RMS) agent leverages PPA to make connections to monitored remote resources to obtain metric data. PPA references the Connection Profiles used during Network Discovery. The Network Discovery Connection Profiles define which protocols are enabled and any applicable credentials and settings for each protocol.

When the RMS agent attempts to monitor an agentless resource, it queries the database and uses the connection profile that was used during Network Discovery. If the RMS agent cannot find a reference to a connection profile, the Default Connection Profile is used. To select a different connection profile, run a new Network Discovery and select the desired connection profile for remote resources.

Timed rules are no longer supported

Timed rules are no longer supported in Monitor Solution. You can now use the heartbeat feature and Task Server instead.

How WMI Metric handles WMI property arrays

In some cases WMI property values are stored as arrays of values. In this case, each property in the array is represented as a different instance\value pair for the metric. The name for instance will be "Name of the instance (n)" where n is the index. For example if % Disk Space had multiple properties for disk space:
"C: (1)" = 23, "C: (2)" = 43, "D: (1)" = 54, "D: (2)" = 7

Issues with agentless monitoring

• If you first discover a computer, and then change either its computer name or its IP address, then attempting to monitor that device will fail.
• It is possible that an agentless monitoring policy will not stop monitoring, even after the policy is disabled.
• The Remote Monitoring Server cannot be installed on 64-bit platforms; therefore, agentless monitoring cannot be performed from a 64-bit computer.
• In the Event Console, the Remote Monitoring Server will be shown as the source of an alert, even if the alert is raised by a monitored agentless resource.
• The monitored computer must be specifically set-up to enable SNMP agentless metrics:
  • The SNMP service must be installed on the monitored computer.
  • The community name must be configured in the SNMP service on the monitored computer. The community name should match the community string in the connection profile SNMP protocol settings. (From the Symantec Management Console, click Settings > All Settings > Solution Settings > Protocol Management > Manage Connection Profiles).
  • The monitored computer's SNMP properties must be configured to accept SNMP packets from any host.

Unmanaged computers are not applied to agentless monitor policies when using filters

If a discovered computer is unmanaged (does not have the Altiris Agent installed), then it cannot be applied to agentless monitor policies with a filter. There is an issue that prevents the unmanaged computers from being targeted correctly, even if the computer is discovered and is contained in a filter.

To apply unmanaged computers to agentless monitor policies, you must manually add the computers using the resource selector. Managed computers are unaffected and can be added to agentless monitor policies using the filter.

Agentless monitoring is not available for targets that have been discovered through an Active Directory Import

You cannot apply agentless monitor policies to targets that have been discovered through an Active Directory Import. To work around this issue, use a Network Discovery or a WINS import instead of an Active Directory Import.

When the Remote Monitoring Server computer’s password is about to expire, agentless monitoring is unavailable

When a Remote Monitoring Server computer has either of the following:

• A password has expired or is within 14 days of expiration
• The option for a user to change the password on first start has been enabled

Then the PPA agent/plug-in is unable to connect through WMI. In this case, the Remote Monitoring Server is then unavailable because the Remote Monitoring Server depends on the PPA agent/plug-in for communications.
To work around this issue you must ensure that the Remote Monitoring Server computer’s password does not come within 14 days of expiration. If the password does expire, then resetting the password resolves the issue until the next time the password comes within 14 days of expiration.
As a possible means of prevention, you can set the Remote Monitoring Server computer’s Windows password to never expire.

Known Issue when Exporting a monitor pack to Windows Vista and Windows Server 2008 when IE Protected Mode is On

When all of the following are true:

• You are exporting a monitor pack to a Windows Vista or a Windows Server 2008 computer
• Internet Explorer protected mode is set to ON
• The Symantec Management Console is open

Then the monitor pack's .XML file is copied to the \temp\ folder instead of the intended path. In this case, completing the import process from this location would fail. You must copy the .XML file manually to the correct location first.

The Process Control task only supports 32-bit systems

At this time, Monitor Solution 7.0 does not support the controlling of 64-bit Windows processes.

Agentless Smart WMI Metrics do not work on Windows 2000 Server target computers

Agentless Smart WMI Metrics do not work on Windows 2000 Server target computers.

Log Event metric supports ANSI-only

The Log Event metric supports ANSI-only log files. When parsing, the Log Event metric assumes a Windows Latin-1 encoding.

If characters that require UTF-8 encoding (for example, when using non-English languages) are contained within either a Log Event metric or within the log, then the metric fails.

The Log Event metric fails to open client side Altiris log files

The Log Event metric is only capable of parsing Notification Server’s a.log file and, therefore, can only monitor the local Notification Server computer’s log. The Log Event Metric fails to open Altiris log files on remote client computers that contain the Monitor Agent.

ASP.NET metric enters retry state when ASP.NET is installed but not registered

This issue occurs when ASP.NET is installed but not registered. To work around the issue, register ASP.NET by running the following command on the target computer:

aspnet_regiis.exe -i

Known Issues with reports

• Reports are run first using default parameters. To receive reports with adjusted parameters, wait until the default parameter graph is displayed, change the parameters, and then run the report again.
• After drilling down into a report's details, you cannot change parameters and see new graph results. To return to the initial view of the report, right-click the report in the left pane, and click Open.

A Monitor Solution report’s customized view is not preserved after the report refreshes

After customizing the view of a Monitor Solution report, once the report is refreshed the customizations to the view are lost and the default view is displayed.

Agent uptime via Heartbeat report not reporting the correct uptime percentage after crash

For the "Agent uptime via Heartbeat" reports to accurately reflect uptime, the Monitor Agent service must be started again after a critical system crash. If the agent is not started again, this period will be erroneously reported as uptime instead of downtime. If you receive an alert that an agent heartbeat has failed, make sure that there is connectivity from the Altiris Agent to Notification Server, and make sure that the Monitor Agent service is running on the monitored computer.

Issues with IIS Memory Performance Report

When viewing the IIS Memory Performance report, some metrics are not displayed, and an exception is logged in the NS log.

The following metrics are not displayed:

• Internet Information Services Global - File Cache Flushes (Avg)
• Internet Information Services Global - File Cache Flushes (Max)
• Internet Information Services Global - File Cache Hits % (Avg)
• Internet Information Services Global - File Cache Hits % (Max)
• Page Reads per Second (Avg)
• Page Reads per Second (Max)

To work around this issue, either view these metrics in the Table View of the report or view them from within the Historical Performance Viewer.

Issues with Disk Paging Activity Report

The Disk Paging Activity report displays incomplete data when it is viewed in the Chart View.
To work around this issue and view all data values, use the Grid View to display this report.

You must reconnect the Real-time Performance Viewer if metric data becomes unavailable

When using the Real-time Performance Viewer, if a metric becomes unavailable and then becomes available again, updated data is not displayed. The viewer continues to display the data value that was last received before the metric became unavailable. The Real-time Performance Viewer must be re-connected to the target computer before updated data is reflected in the viewer.

When using the Historical Performance Viewer, the value displayed at the mouse pointer may not update

This issue is known to occur when the chart area of the Historical Performance Viewer does not have the current cursor focus, for example when the "refresh" button has been clicked. To work around this issue, click the mouse cursor within the chart, and then hover the mouse cursor over the desired data point.

Limited monitoring of ESX Servers

Monitoring of ESX Servers is limited when using the WS-MAN Agentless metric because of limitations in Pluggable Protocol Architecture (PPA)

Only the CIMV2 namespace can be used for monitoring ESX servers. The following vendor namespaces are not currently supported:

• OMC=http://schema.omc-project.org/wbem/wscim/l/cim-schema/2
• VMware=http://schemas.vmware.com/wbem/wscin/l/cim-schema/2
• PRS

 This issue is caused by limitations of PPA. Only namespaces defined in the PPA repository can be used.

Performance issues with a resource target that is running Windows Server 2003

If a resource target is running Windows Server 2003 and has the Monitor Agent installed, the resource target runs slowly.

Event Console purges all events during purging maintenance

During purging maintenance, Event Console purges all alerts including active events.

Issues with “UNIX and Linux servers requiring Monitor Agent upgrade” filter

Error occurs when attempting to open the "UNIX and Linux servers requiring Monitor Agent upgrade” filter.

Custom filters do not update if applied to an agentless based Monitor policy

If a custom filter is applied to an agentless based Monitor policy, the filter does not update when the asset status of a resource changes.

Power Monitoring stops working after 4 weeks

After 4 weeks, Power Monitoring does not correctly report the state of the servers it is monitoring.  

Error with Historical Performance Viewer

When viewing Historical Performance Viewer from a workstation, the following error occurs: "There was an error in the callback." When viewing historical data for the same computer directly from the Notification Server, the following error occurs: "column 'instancekey' is constrained to be unique. value '13932' is already present". 

Monitor Agent can crash on service stop/restart when NT Event Log monitoring is performed

When large amount of NT events are parsed by metric provider, the Monitor Agent can crash on service stop/restart and create gaps of monitored data in Historical Performance Viewer.

Metric provider does not properly parse NT events on Windows 2008 SP1 Server

Metric provider does not properly parse description of NT events according to configured NT Event Rule.

App detector does not run on a scheduled interval

On a Linux server, when app detector is not able to pass any detection method due to a locked RPM database, all packs are set to inactive state and the monitor agent will not invoke app detector again until it receives a new configuration.

Percent Disk Utilization metric returns incorrect values

Percent Disk Utilization metric is returning incorrect values causing
incorrect rule trigger.

Symantec Management Agent log does not contain monitor's messages

By default, Monitor Agent does not log warning and error messages to the Symantec Management Agent log.

Actual state of a device and the state reflected in the console are sometimes out of sync

When monitoring 500 or more devices using agentless monitoring, the state of the actual device and the state reflected in the console are sometimes out of sync. The actual device might be in a critical state and the console would indicate that it is in a normal state. 

Windows Service metric based rule always reverts to "Is not equal to" with query type of "Installed"

A rule which references a Windows Service metric that has a query type of
"Installed" always reverts to a condition of "Is not equal to". If you modify the rule's condition to "Equal to installed", save it, then modify it again, it reverts back to "Is not equal to installed". 

Metric provider crashes on AIX computers

Metric provider crashes on AIX computers and leaves a core dump error of "SOFTWARE PROGRAM ABNORMALLY TERMINATED". The Monitor Agent then restarts on its own after a period of time.  

Command metric with incorrect command specified never gets into Retry/Unavailable mode

A Command metric never gets into Retry/Unavailable mode if an incorrect command is specified. This metric is also shown as available in Real-
time Viewer and always return 0. 

Cannot open the "Monitor Table Space Used by Monitored Resources" report

After monitoring a group of computers for a period of time, you cannot open the "Monitor Table Space Used by Monitored Resources" report. When you try to open the report, a timeout error message appears.

Incorrect heartbeat alert behavior on a Virutal Machine

A heartbeat alert may be raised in Event Console from a monitored resource, which runs on Virtual Machine, even if it is online and metric provider is running. This incorrect behavior is caused becase of a known issue with VMware virtual machines. The following link describes this issue and how to resolve this problem:  http://www.vmware.com/pdf/vmware_timekeeping.pdf

1982360

A single ping rule in two different agentless Monitor policies only works in the first policy

If two different agentless Monitor policies contain the same ping rule, only the first occurrence of the rule will trigger and generate alerts.

A second Monitor Solution is created when you upgrade to Monitor Solution 7.0 SP5

If you have an off-box Remote Montioring Server and you upgrade to Monitor Solution 7.0 SP5, a second Monitor Service is created on the Notification Server. To resolve this issue after the upgrade, do the following:

1. In the Symantec Management Console, on the Settings menu, click Notification Server > Site Server Settings.
2. In the left pane, under Site Management, click Site Servers > Notification Server.
3. In the right pane, in the Monitor Service section, click Install/remove services.
4. In the Add/Remove Services dialog box, uncheck Monitor Service.
5. You must then manually install the SP5 version of the PPA Agent on the Remote Monitoring Server where the SP4 version of the PPA Agent is installed.

After you install the SP5 version of the PPA Agent SP5 on the remote server, the PPA inventory is sent to the Notification Server and the upgraded off-box Remote Monitoring Server will be shown as correctly installed on the Site Management Page.

Unable to manually resolve agent-based legacy alerts in Event Console after upgrade from 7.0 SP4 HF1

After you upgrade from 7.0 SP4 HF1, legacy alerts that were raised from agent-based resources could not be properly resolved from the Event Console when using manual resolve. This issue doesn't affect agentless alerts or alerts resolved by a metric value change.

The workaround is to reset all agent-based monitored resources after the server and clients are upgraded to the 7.0 SP5 version.
To reset the agent-based monitored resources, create a new "Reset Monitored Resource" task. Add a rule to the target that excludes resources that are not in the filter "Computers with Monitor Agent installed." The task will then not target Remote Monitoring Servers so agentless alerts will not be reset. After you run the task, all agent-based alerts will be resolved both in Event Console and on the client, and you will be able to  resolve new alerts that are raised raised in the Event Console.