Microsoft Lync Server 2010Management Pack Guide

Microsoft Corporation

Published: November, 8th 2010

Send suggestions and comments about this document to . Please include the management pack guide name with your feedback.

Copyright

Information in this document, including URL and other Internet Web site references, is subject to change without notice. Unless otherwise noted, the companies, organizations, products, domain names, e-mail addresses, logos, people, places, and events depicted in examples herein are fictitious. No association with any real company, organization, product, domain name, e-mail address, logo, person, place, or event is intended or should be inferred. Complying with all applicable copyright laws is the responsibility of the user. Without limiting the rights under copyright, no part of this document may be reproduced, stored in or introduced into a retrieval system, or transmitted in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise), or for any purpose, without the express written permission of Microsoft Corporation.

Microsoft may have patents, patent applications, trademarks, copyrights, or other intellectual property rights covering subject matter in this document. Except as expressly provided in any written license agreement from Microsoft, the furnishing of this document does not give you any license to these patents, trademarks, copyrights, or other intellectual property.

© 2010 Microsoft Corporation. All rights reserved.

Microsoft, and MS-DOS, Windows, Windows PowerShell, Windows Server, Active Directory, and Lync are trademarks of the Microsoft group of companies. All other trademarks are property of their respective owners.

Contents

Contents

Microsoft Lync Server 2010 Management Pack Guide

Introduction to the Microsoft Lync Server 2010 Management Pack

Changes in This Update

What's New

Central discovery

Reduced Alert Noise:

Prioritized alerts:

Stateful alerting:

Transient handling:

Synthetic transactions

Call reliability monitoring

Media Quality monitoring

Port monitoring

Simple URL monitoring

Supported Configurations

Getting Started

Before You Import the Management Pack

Deploying Microsoft Lync Server 2010 Management Pack

Pre-Requisites

Enable Agent Proxy settings in Operations Manager Console

How to Import the Lync Server 2010 Management Pack

Import Management Pack

How Central Discovery works

Verify Component monitoring works

Troubleshooting Central Discovery failures

Coexistence with Lync Server 2013 Management Packs

Deploying Call Reliability & Media Quality Monitoring

Prerequisite

Deployment steps for QoE Monitoring

Verify Media Quality monitoring works

Verify Call Reliability monitoring works

Configure alerting thresholds for Media Quality monitoring

Deploying Synthetic transactions (STs)

Setting up Synthetic transaction watcher node

Capacity and OS Requirements:

Deploying synthetic transactions watcher node

Verify that Synthetic Transactions are working

Troubleshooting ST Watcher issues

Configure Port monitoring for your deployment

Configure URL monitoring for your deployment

How Health Rolls Up

Health Rollup Diagram

Maintenance

Pool Maintenance

Server maintenance

Configuring Notifications

Create a New Management Pack for Customizations

Optional Configuration

Enabling Performance Threshold and Collection Rules

Slow WAN Links or Large Branch Office Deployments

Security Considerations

Low-Privilege Environments

Run As Profiles

Management Pack problems and resolutions

Microsoft Lync Server 2010Management Pack Guide

Document Version

This guide was written based on the 4.0.7577.0version of the Microsoft®Lync™ Server 2010 communications software Management Pack.

Revision History

Release Date / Changes
November, 8th 2010 / Original release of this guide
December, 3rd, 2012 / Co-existence support with Lync Server 2013

Introduction to the Microsoft Lync Server 2010Management Pack

The Lync Server 2010 Management Pack (MP) is your monitoring solution of choice for monitoring any Lync Server 2010 deployment.

The management pack implements traditional Event Log and Performance counter based instrumentation is utilized, as well as enabling newly available instrumentation in Lync Server 2010, such as pair events (failure/success) for several Key Health Indicators, as well as fully utilizing the new Synthetic Transactions (Test-Cs* Windows PowerShell™command-line interface cmdlets).

Changes in This Update

Version 4.0.7577.203 of the Lync Server 2010 Management Pack includes the following changes:

  • Added functionality to support co-existence of management packs during migration from Lync Server 2010 to Lync Server 2013. For more information, see Coexistence with Lync Server 2013 Management Packs.
  • Fixed an issue that caused alerts from non-Windows computers that were not used by Lync.

Getting the Latest Management Pack and Documentation

You can find the Lync Server 2010 Management Pack in the System Center Operations Manager 2007 Catalog (

What's New

The following features are new in this release of the Lync Server 2010 Management Pack, compared to previous iterations (Microsoft®Office Communications Server 2007 and 2007 R2 Management Packs).

Central discovery

Discovery of roles and components that need to be monitored is automatically completed based on a central discovery script that reads the topology document published in Central Management Database.

Reduced Alert Noise:

This management pack features significantly improved alert to noise ratio compared to previous versions of Lync Server 2010 management packs. This investment is based on applying health models for each component in the product and driving alerting based on the health model.

Prioritized alerts:

Alerts have been classified out-of-the-box into:

High Priority alerts:

These are indicative of service unavailability for multiple users. For example, a Component failure on a single machine is NOT a high priority alert. This is because Lync Server 2010 product has built-in high availability features (for example, multiple Front-end servers behind load balancers). A high-priority alert indicates an outage as observed from a synthetic transaction. Alerts with this priority are serious enough “to wake up administrators at night.”

Medium Priority alerts:

These are indicative of component failures. For example, Conferencing service not operational on a particular server. Alerts in this category are stateful and indicate the current status of the issue. In this example, if conferencing service has recovered, the alert would be auto-resolved in Microsoft® System Center Operations Manager. The expectation for these alerts is that an administrator will look at them on the same business day.

Other Alerts:

These are alerts from components that might affect a specific user/ subset of users. For example, the Address Book service could not parse the Active Directory® Domain Services (AD DS) entry for user: . The expectation for these alerts is that administrators will get to them when they have time available.

Stateful alerting:

Alerts are auto-resolved when a failure condition is no longer applicable. Component alerts (Medium priority only) are driven by Operations Manager monitors that automatically turn back to “healthy” state when the problem is no longer observed. (These alerts are still available in “Closed Alerts” view for administrators to look at). The benefit for administrators is to know the current status of a problem/alert by looking at monitor state.

Note: “Other Alerts” category of component alerts (see above) and Call failure alerts are NOT stateful.

Transient handling:

This management pack defines monitors that turn to unhealthy state only when there are consecutive failures. For example, a “High Priority” alert is generated only when a synthetic transaction fails twice consecutively. (The first failure is treated as a warning state for the monitor).

Synthetic transactions

This new feature in Microsoft Lync Server 2010 management pack provides alerting based on end-to-end monitoring scenarios.

Synthetic transactions are Windows PowerShell cmdlets that are integrated into Operations Manager management pack and test end-to-end user scenarios. Once an administrator has designated a server to execute synthetic transactions, these cmdlets are triggered periodically by the management pack. Failures from a synthetic transaction generate a stateful alert.Here are supported synthetic transactions for Microsoft Lync Server 2010:

No / Scenario Name
1 / Peer to Peer Instant Messaging
2 / Peer to Peer audio call
3 / Group Instant Messaging
4 / Audio/Video Conference
5 / Rich Presence
6 / Registration (User Login)
7 / Dial in conferencing
8 / Address book service (file download)
9 / Address book web query
10 / Peer to Peer PSTN Call

Call reliability monitoring

This new feature in Lync Server 2010 management pack provides administrators with real-time alerting based on call reliability metrics.

Every call failure is reported to the Lync Server 2010 call detail recording (CDR) database.Lync Server 2010management pack integration queries this database (CDR) periodically to look at call failures. When call failures are detected, they are aggregated on “Diagnostic Codes.” Alerts are generated when a “Diagnostic Code” exceeds a specified threshold for call failures.

Media Quality monitoring

This feature in Microsoft Lync Server 2010 management pack provides real-time alerting based on network conditions. After every session is completed, all Lync Server 2010 endpoints generate a “Quality report” that is delivered to the Quality of Experience (QoE) database (Monitoring Server). The System Center Operations Manager integration for Media quality alerts is driven out of this data. Administrators receive alerts based on threedifferent categories. The first category is network locations such ascalls between regions, Sites, or subnets. The second category is servers such as A/V Conferencing Server. The last category is call legs such as Gateway (Mediation Server Bypass), Gateway and Mediation Server Leg, or Mediation Server and Client Endpoint Leg.

This integration features dynamic discovery of faulty instances and generating alerts for them. Alerts are stateful and get auto-resolved when a problem no longer occurs.

Port monitoring

This new feature in Microsoft Lync Server 2010 management pack provides automatic monitoring for ports that are used by Microsoft Lync Server 2010 product inside the corporate firewall. Any port that cannot be pinged results in an alert that is auto-resolved when the port is working again.

Simple URL monitoring

This new feature in Microsoft Lync Server 2010 management pack provides automatic monitoring for URLs that are used by Microsoft Lync Server 2010 for conference join scenarios. Any URL that cannot be pinged results in an alert that is auto-resolved when that URL is working again.

Supported Configurations

System Center Operations Manager 2007 server support

  • System Center Operations Manager 2007 R2 (32-bit and 64-bit)
  • Not supported: System Center Operations Manager 2007 RTM, or 2007 SP1*

* Lync Server 2010 Management Pack requires the Windows PowerShell provider support which had been added with Operations Manager 2007 R2, and cannot be deployed on Operations Manager 2007 RTM.

System Center Operations Manager 2007 agent server support

  • System Center Operations Manager 2007 R2 Agent, 64-bit only
  • Not supported: System Center Operations Manager 2007 Agent RTM, or 2007 SP1
  • Not supported: System Center Operations Manager 2007 Agent R2 32-bit running in WoW64

Lync Server 2010 supported configurations

Configuration / Support
WindowsServer® operating system2008R2 / Yes, all editions.
WindowsServer®operating system 2008 SP1 / Yes, all editions, 64-bit only, SP1 and higher.
Clustered servers / Not supported.
Agentless monitoring / Not supported.
Virtual environment / Yes, see Lync Server 2010 virtualization documentation for details.
Domain Joined Server Roles / All internal Lync Server 2010 server roles have to be domain joined.
Standalone Server Roles / Lync Server 2010 Edge is not required to be domain joined.
Topology Limitations / All server roles in a deployment must be monitored from the same Operations Manager 2007 R2 Management Group.
Survivable Branch Office Appliance (SBA) support / SBA monitoring is supported(additional configuration required).
Synthetic Transactions (ST) Watcher Node / ST Watcher Node monitoring is supported (additional configuration required).

Getting Started

Before You Import the Management Pack

Note: Read this section carefully, as additional steps are required to enable full monitoring of your Lync Server 2010 deployment.

Deploying Microsoft Lync Server 2010 Management Pack

Pre-Requisites

  • Operations Manager 2007 R2 installation is already completed.
  • Agent proxy enabled for all Front-end servers in Front-end pool where Central Management Server is deployed. This is usually the first pool in your deployment. Agent Proxy also needs to be enabled for the Synthetic transaction watcher node machine (see details below).
  • Lync Server 2010 is already installed.
  • Operations Manager Agent has been deployed on all servers that are part of the Lync Server 2010 deployment (including Edge Servers, if Edge monitoring is desired).

Enable Agent Proxy settings in Operations Manager Console

  • Make sure Agent Proxy setting is enabled for all Front End Servers in the Pool that contains Central Management Server and also for the server that will be used as synthetic transaction watcher node where synthetic transactions will run from.
  • To enable agent proxy, follow these steps:
  1. Go to the server that is running System Center Operations Manager. Open Operations Manager Console.
  2. Navigate to Administration tab.
  3. Select Agent Managed.
  4. From the detail view on the right side, right-click the watcher node agent,Click Properties and from the Properties dialog click the Security tab.
  5. Check “Allow this agent to act as proxy and discover managed objects on other computers” check box. See Fig. 1

Fig. 1 Agent properties window

How to Import the Lync Server 2010 Management Pack

For instructions about importing a management pack, see How to Import a Management Pack in Operations Manager 2007 (

After the Lync Server 2010 Management Pack is imported, it is recommended to create a new management pack in which you store overrides and other customizations.

Import Management Pack

  • On the server that is running System Center Operations Manager 2007 R2.
  • Install the MSI Lync Server 2010 Management Pack.msi.
  • The file Microsoft.LS.2010.Monitoring.mp is installed to %Program Files%\System Center Management Packs\Lync Server 2010 Management Pack ,
    or %Program Files%\System Center Management Packs\Lync Server 2010 Management Pack.
  • Use the Operations Manager “Import Management Packs” to import the MP.
  • Or from command line run “MPimport.exe <path to ManagementPack>”.

How Central Discovery works

The first step after administrator imports the Lync Server 2010 management pack is for Central Discovery to execute automatically and discover the entire Lync Server deployment. This effectively jump starts all component monitoring.

Central discovery script is a Windows PowerShell script that discovers all Microsoft Lync Server 2010 Roles/Components/Services to be monitored. This script runs on the Operations Manager agent that is discovered as the Central Discovery Watcher Node. The default algorithm to automatically pick a machine as central discovery watcher node is as follows:

  1. Find the Microsoft Lync Server 2010 Front End pool where Central Management Store is installed (usually the Front End Pool which has been installed first).
  2. Select all the member machines in the Front End Server role (this could be an Enterprise Edition Consolidated pool with multiple Front End Servers using a HLB or DNS load balancing).
  3. Identify the computer which is currently the active master.
  4. This machine will be used to discover all roles and components in the topology. For more details, see Fig. 2.

The entire topology information is available in Central Management Database and a single script in the management pack retrieves the topology data and populatesOperations Manager 2007 classes and relationships for all server roles & components. (This is different from the Office Communications Server 2007 R2 management pack behavior where each server role was individually discovered using registry keys on the local machine).

Important Note: Edge pool discovery is turned off by default.

To turn it on override value of DiscoverEdgeServerRole property of LS Central Topology Discovery object to ‘True’.

Verify Component monitoring works

Verify whether component monitoring is working by looking at the built-in views in the management pack.

Look at state view “5.Pools” (under Lync Server2010 folder). Please see Fig 2.

If this view shows one or more instances, then central discovery/component monitoring has started working.

Note: It may take several minutes for this view to refresh depending on the size of the topology.

Fig. 2 Discovered pools State view

Troubleshooting Central Discovery failures

Here are common steps to troubleshoot central discovery failures:

Problem: No instances seen in state view “5.Pools”.

Steps:

  1. Open view in folder “Topology Discovery\Discovery State view”. See Figure #3 for details.
  2. Ensure that all Front End Servers in the Front End pool that contains Central Management Server are listed in this view.
  3. If a Front End Server is missing from this view, then please check:
  4. Did the Operations Manager agent get pushed to that server?
  5. Is this front-end server heart-beating in Operations Manager correctly? Is this server down for maintenance or otherwise unavailable?
  6. Ensure that the server (FQDN) that will be picked by algorithm in section “How central discovery works” shows “Lync Server Discovery Script” column as “Healthy” see Figure #3:

Fig. 3 Topology Discovery State View

If “Lync Server Discovery Script” column does not show “Healthy,” then ensure that this machine is heart-beating without errors in System Center Operations Manager and is not down for maintenance.