HP HA s Software guide

Installing and Using EMS

The Role of EMS HA Monitors in a High Availability Environment

The Role of EMS HA Monitors in a

High Availability Environment

The weakest link in a high availability system is the single point of failure. EMS HA Monitors can be used to report information that helps you detect loss of redundant resources, thus exposing single points of failure, a threat to data and application availability.

Because EMS is a monitoring system, and does not do anything to modify the system, it is best used with additional software that can take action based on the events sent by EMS. Some examples are:

•EMS HA Monitors and MC/ServiceGuard

MC/ServiceGuard uses the EMS monitors to determine the health of resources, such as disks, and may fail over packages based on that information. Conﬁguration of EMS monitoring requests for use with MC/ServiceGuard packages is done from the Cluster area for Package Conﬁguration in SAM, or by editing the ASCII package conﬁguration ﬁle.

However, if you also want to be alerted to what caused a package to fail over, or you want to monitor events that affect high availability you need to create requests from the SAM interface in the Resource Management area as described in “Using EMS HA Monitors” on page 17, and in subsequent chapters.

MC/ServiceGuard may already be conﬁgured to monitor the health of nodes, services, and subnets, and to make failover decisions based on resources status. Conﬁguring EMS monitors provides additional MC/ServiceGuard failover criteria for certain network links and other resources.

•EMS HA Monitors with IT/Operations or Network Node Manager

EM HA Monitors S can be conﬁgured to send events to IT/Operations and Network Node Manager.

•EMS HA Monitors with your choice of system management software

Because EMS can send events in a number of protocols, it can be used with any system management software that supports either SNMP traps, or TCP, or UDP messages.

14

Chapter 1

Image 14

HP HA s Software manual Role of EMS HA Monitors in a High Availability Environment

Contents

Using EMS HA Monitors B5735-90001 August Legal Notices Contents Monitoring Cluster Resources Monitoring Network Interfaces Glossary TroubleshootingMonitoring System Resources Contents B5735-90001 Printing HistoryPrinting Date Part Number Edition Page Preface PublicationsPage Installing and Using EMS Event Monitoring Services High Availability Monitors What are EMS HA Monitors? Chapter Role of EMS HA Monitors in a High Availability Environment Installing and Removing EMS HA Monitors Installing EMS HA Monitors Removing EMS HA Monitors Status AvailMB Using EMS HA MonitorsEvent Monitoring Service Resource Class Hierarchy Pvpvlink Event Monitoring Service Screen Selecting a Resource to Monitor Screen Using Wildcards Creating a Monitoring Request Monitoring Request Parameters How Do I Tell EMS When to Send Events? What is a Polling Interval? Which Protocols Can I Use to Send Events? What is a Notiﬁcation Comment? Removing Monitoring Requests Copying Monitoring RequestsModifying Monitoring Requests Conﬁguring MC/ServiceGuard Package Dependencies Service EMS Package Conﬁguration Screen Package Resource Dependencies Screen Resource Parameters Screen Using EMS HA Monitors Monitoring Disk Resources Monitoring Disk Resources Disk Monitor Reference Disk Monitor Resource Class Hierarchy Physical Volume Summary Interpreting Physical Volume Summary Disk Monitor Reference Physical Volume and Physical Volume Link Status Interpreting Physical Volume and Physical Volume Link Status Logical Volume Summary Interpreting Logical Volume Summary Logical volume is inactive Logical Volume StatusInterpreting Logical Volume Status Logical volume is DOWN, a complete copy of the data is not Logical Volume Number of Copies Interpreting Logical Volume Copies Rules for Using the EMS Disk Monitor with MC/ServiceGuard Rules for Using the EMS Disk Monitor with MC/ServiceGuard Rules for RAID Arrays Pvsummary Calculations Case Conclusion State Creating Volume Groups on Disk Arrays Using PV Links Adding PVGs to Existing Volume Groups # mkdir /dev/vgdatabase Creating Logical Volumes Rules for Mirrored Individual Disks Creating Disk Monitoring Requests Disk Monitoring Request Suggestions Parameters When Disks Fail Resources to Monitor for RAID Arrays RAID Array Example Pvgup Return Resources to Monitor for Mirrored Disks Mirrored Disks Example Monitoring Parameters Resource Notify Condition Option Number Pvsummary Valid Meaning Value Devices Resources to Monitor for Lock DisksExample for Interpreting the pvsummary for Mirrored Disks Resource Monitoring Parameters Notify Condition Option Resources to Monitor for Root Volumes Monitoring Cluster Resources Cluster Cluster Monitor ReferenceCluster Monitor Resource Class Hierarchy Cluster Status Interpreting Custer Status Node Status Interpreting Node Status Package Status Interpreting Package Status Creating Cluster Monitoring Requests Monitoring Network Interfaces Interpreting LAN Interface Status Network Monitor ReferenceNetwork Monitor Resource Class Hierarchy Chapter 30 sec Net/interfaces/lan/status/LANname When value isConﬁguring Network Monitoring Requests Hour Monitoring System Resources System Monitor Reference System Resource Monitor Class Hierarchy Resource Name Value Range Interpretation Number of UsersInterpreting Number of Users Resource Name Value Interpretation Range Job QueuesInterpreting Job Queues Resource Name Filesystem Available SpaceFilesystem Available Space Creating System Resource Monitoring Requests 6Troubleshooting EMS Directories and Files Etc/opt/resmon/log Logging and tracing EMS Logging EMS Tracing Performance Considerations System Performance IssuesNetwork Performance Issues Testing Cluster Monitor Requests Testing Monitor RequestsTesting Disk Monitor Requests Testing Network Monitor Requests Making Sure Monitors are Running Glossary Notiﬁcation See alert Glossary Glossary Index Index Index