HP UX 11i Workload Management (gWLM/WLM) Software How the automatic restart works, Related events

Page 42

2.(Optional) Edit the property com.hp.gwlm.node.HA.minimumTimeout

in the file /etc/opt/gwlm/conf/gwlmagent.properties to set the minimum number of seconds that must pass before a managed node considers itself separated from its SRD. Set this property to ensure that minor network problems do not cause a managed node to prematurely consider itself separated.

gWLM uses this value only if it is larger than 10 multiplied by gWLM’s allocation interval. For example, with an allocation interval of 15 seconds, a node can go 2.5 minutes without communicating with its SRD before the node’s gWLM agent attempts to re-connect with the SRD.

This feature works best when one managed node is lost at a time or all managed nodes are lost.

NOTE: If a vpar is borrowing cores from other vPars when it loses contact with its SRD, those borrowed cores might be separated from the SRD. If the vpar might be down for an extended time, check that the SRD has reformed without that vpar and that it has enough cores to meet its commitments. If not, try using vparmodify to reclaim some of the cores. (With the vpar down, you will not be able to modify it locally, and only some versions of HP-UX Virtual Partitions allow you to easily modify a remote vpar.)

Similarly, if an npar has several active cores (due to Instant Capacity) when it loses contact with its SRD, you might have to manually size the npar to reclaim those cores for nPars still in the SRD. For more information, see the Instant Capacity documentation.

How the automatic restart works

When a managed node boots, the gWLM agent (gwlmagent) starts automatically if

GWLM_AGENT_START is set to 1 in the file /etc/rc.config.d/gwlmCtl. The agent then checks the file /etc/opt/gwlm/deployed.config to determine its CMS. Next, it attempts to contact the CMS to have the CMS re-deploy its view of the SRD. If the CMS cannot be contacted, the SRD in the deployed.config file is deployed as long as all nodes agree.

In general, when an SRD is disrupted by a node’s going down, by a CMS's going down, or by network communications issues, gWLM attempts to reform the SRD. gWLM maintains the concept of a cluster for the nodes in an SRD. In a cluster, one node is a master and the other nodes are nonmasters. If the master node loses contact with the rest of the SRD, the rest of the SRD can continue without it, as a partial cluster, by unanimously agreeing on a new master. If a nonmaster loses communication with the rest of the SRD, the resulting partial cluster continues operation without the lost node. The master simply omits the missing node until it becomes available again.

You can use the gwlmstatus command to monitor availability. It can tell you whether any hosts are unable to rejoin a node's SRD as well as whether hosts in the SRD are nonresponsive. For more information, see gwlmstatus(1M).

NOTE: Attempts to reform SRDs might time out, leaving no SRD deployed and consequently no management of resource allocations. If this occurs, see the HP Matrix Operating Environment Release Notes and follow the actions suggested in the section titled “Data Missing in Real-time Monitoring.”

Related events

You can configure the following System Insight Manager events regarding this automatic restart feature:

•Node Failed to Rejoin SRD on Start-up

•SRD Reformed with Partial Set of Nodes

•SRD Communication Issue

42 Additional configuration and administration tasks

Page 41 Page 43
Image 42
Page 41 Page 43
Contents HP Global Workload Manager 7.4 User Guide Acknowledgements Contents Additional configuration and administration tasks Support and other resourcesIndex Overview Benefits of using gWLMComparison of PRM, WLM, and gWLM features GWLM OverviewConcepts and terms for using gWLM Scheduler fss group. gWLM manages a workload byWould make for a workload-without actually affecting GWLM management model SRDHow gWLM allocates CPU resources Available interfaces AssumptionsWhere to find additional information Finding more gWLM informationLearn about other gWLM commands Configuring gWLM to manage workloads Policy typesPolicy types Choosing a policy type Choosing a policy typeGetting started with gWLM Using the wizardSeeing how gWLM will perform without affecting the system Combining the different policy typesSeeing gWLM in action # /opt/gwlm/bin/gwlmagent# perl -e ’print $$\nwhile 1 ’ & 1 Common uses for gWLM Fixing the amount of CPU resources a workload getsCommon configuration tasks Setting up gWLM initial setup stepsChanging from advisory mode to managed mode Creating a new policyEditing a policy Changing which policy is associated with a workloadAdding a new compartment or GiCAP group member to an SRD Stop managing a workloadStop managing an SRD Quick Link Option Monitoring workloads Viewing gWLM reports in monitor-Only modeMonitoring workloads and gWLM High-Level viewMonitoring gWLM from the command line Message logsHP-UX /var/opt/gwlm/gwlmcommand.log.0 Viewing HP Systems Insight Manager events Monitoring gWLM with GlancePlusSecuring gWLM communications SecurityGeneral security topics Securing database communicationsSecurity Additional configuration and administration tasks Manually adjusting CPU resourcesManually adjusting memory resources Setting aside space for historical dataCreating a database maintenance plan Setting cache size for historical configuration dataSetting database recovery model Tips for backup and restoreSetting gWLM properties CMS propertiesSetting gWLM properties Agent properties # Severe # Warning # Info # Config # Fine # Finer # Finest Controlling gWLM’s startup behavior Communications ports# /opt/gwlm/bin/gwlmagent --enablestartonboot How the automatic restart works Related eventsManually clearing an SRD Node Failed to Rejoin SRD on Start-up event# gwlm undeploy --srd=SRD--force Nesting partitions Nested partitionsChanging the gWLM resource allocation interval Using gWLM with Hyper-ThreadingChanging the interval on the command line Changing the interval in HP System Insight ManagerUsing gWLM with hosts on multiple LANs Shows a possible scenarioCreating Golden Images Multiple network interface cardsIncorrectly configured host name or IP address Get the IP address configured for the host using nslookupUnable to create new native thread Enabling or disabling dumping core for gWLMCore dump file is available in root directory Support and other resources How to contact HPSubscription service Information to collect before contacting HPHP authorized resellers Related informationTypographic conventions User input Documentation feedback Compatibility with agents Global Workload Manager known issues LimitationsUnable to build a single shared resource domain Compatibility with PRM and WLMMaking a configuration change to a large SRD is slow Rare incompatibility with virtual partitionsHost name aliases are not supported Deleting workloads takes a long time Integrity VM prevents discovery of psets and fss groupsInformation error during shutdown Process placement using psrset is ignoredCustom metrics lost on redeploy Major issues Documentation or minor issues Remove old configuration data from the gWLM databaseCell-local processors and iCAP environment CMS is slow to respondError during discovery of compartments Combining psets and virtual partitionsModifying Java while gWLM is running Missing or unexpected historical data system clocks differ Sample missing at start or end of gwlmreport outputOnly one SRD is allowed to be deployed Application hangs in fss group Processes moved to default pset or default fss groupSRD deployment times out and displays a blank screen Scripts not placed in correct workloadsConfiguration of agent and CMS not synchronized Changes in sizes in DiscoveryUnable to remove workload from nested partitions SRD Missing historical data gWLM CMS daemon/service restarted Negative current size for NonvmUnmanaging a VM or vPar that is on leaves SRD undeployed Index SRDIndex

UX 11i Workload Management (gWLM/WLM) Software specifications

HP-UX 11i Workload Management (gWLM/WLM) software is an integral component of HP's premier UNIX operating system, designed to enhance system performance and resource management across diverse workloads. This advanced tool allows system administrators to monitor, control, and allocate resources effectively to achieve optimal performance, reliability, and service levels in enterprise environments.

One of the main features of gWLM/WLM is its ability to classify workloads and manage them according to specific policies set by the administrator. By using service level objectives (SLOs), administrators can define the performance criteria for various applications and workloads. gWLM continuously tracks these workloads, ensuring that they adhere to the defined SLOs, thus maintaining a high level of application performance.

The software employs resource pools, which segment resources such as CPU, memory, and I/O bandwidth among different workloads based on predefined priorities. This capability ensures that critical applications receive the resources they require, even during peak usage periods, thereby preventing resource starvation that could lead to system slowdowns or crashes.

Another significant characteristic of gWLM/WLM is its real-time monitoring and reporting capabilities. The software provides detailed insights into resource utilization, workload performance, and system health. Administrators can access this information through a user-friendly interface, allowing for informed decision-making and proactive management.

Integration with HP Serviceguard adds another layer of functionality, enabling high availability for critical applications. gWLM can orchestrate workload migration to ensure that service levels are maintained, even in the event of hardware failures or resource contention.

The technology behind gWLM/WLM is built on advanced algorithms that leverage historical data and predictive modeling to optimize resource allocation dynamically. This means that as workloads change, the system can automatically adjust resource distribution to meet performance targets without the need for constant manual intervention.

gWLM also supports integration with various enterprise management tools, enabling administrators to implement comprehensive monitoring and management strategies across the IT infrastructure. The scalability of gWLM allows organizations of all sizes to benefit from its robust workload management features, ensuring that they can adapt to changing demands in their operational environments.

In summary, HP-UX 11i Workload Management software offers a sophisticated solution for optimizing resource utilization, managing workloads effectively, and maintaining high performance in complex enterprise environments. Its comprehensive features and technologies make it an essential tool for any organization seeking to enhance their IT operations.
Top Page Image Contents