OMEGAMON XE for Messaging V730 Long client presentation

All AboutOMEGAMON XE for Messaging for z/OS Version 7.3

Brief Overview

3

OMEGAMON XE for Messaging - Brief OverviewProvides complete solution for MQ and Broker monitoring

• Efficiency and cost saving through integration, Messaging offers an enterprise-wide single point of control with other OMEGAMONs and distributed platform ITCAM Agents for WebSphere Messaging within both the Tivoli Enterprise Portal and the Enhanced 3270 User Interface

• Superior problem determination capability with real-time status and statistical monitoring about availability and performance, along with historical data collection for reporting, performance analysis, trend prediction and enterprise-wide business impact analysis

• Reduced time-to-resolution of problems with automated problem situation detection and corrective actions, in conjunction with a wealth of workspaces for root cause analysis and correlation with related data about other monitored subsystems

OMEGAMON XE for Messaging includes capability to gain improved visibility and management of messaging subsystems

IBM MQ for z/OS (WebSphere MQ) IBM Integration Bus for z/OS (WebSphere Message Broker)

4

OMEGAMON XE for Messaging - Brief OverviewIntegrated health monitoring solution

• Monitor the health of all queue managers in your enterprise using either the Enhanced 3270 UI or the Tivoli Enterprise Portal

• Navigate directly to other OMEGAMONs

OMEGAMON XE for Messaging - Brief OverviewIBM MQ (WebSphere MQ) Monitoring

5

• Queue manager availability, health and performance• Queue status, usage and statistics• Channel status and performance• MQ event monitoring and archival • Application connections and topology• Supports ITM features with historical reporting,

situations, event forwarding, take-action

• Buffer pool, page set, message manager, log manager and topic manager statistics

• Pub/sub topics and subscriptions• Dead letter queue and message manipulation• Application (MQI monitoring) statistics• Queue sharing group status• MQ cluster monitoring

OMEGAMON XE for Messaging - Brief OverviewIBM Integrated Bus (WebSphere Message Broker) Monitoring

6

• Broker availability and performance• Topology with status for broker, execution groups and

message flows• Message flow node topology• Broker status events for broker deployment• Supports ITM features with historical reporting,

situations, event forwarding, take-action

• Accounting Message Flow Statistics include message flow CPU and elapsed timings, message rate and error counts

• Resource Statistics provide Execution Group level statistics for JVM, Parsers, Files, ODBC, JDBC and SOAP

OMEGAMON XE for Messaging - Brief OverviewIBM MQ (WebSphere MQ) Configuration

7

• Enterprise-wide MQ configuration database• Discovery of existing MQ resources• Prototypes (templates) and variables for configuring

multiple MQ nodes• Validation prior to deployment• Scheduled and immediate deployment of changes

• Synchronize actual configuration to defined (database), or defined to actual

• Identify discrepancies• Audit trail• Backup/restore of database• Recovery of MQ environments

OMEGAMON XE for Messaging - Brief OverviewAutomated problem situation detection and action

8

• Proactively monitor your Messaging environment• Wide variety of product-provided sample situations• Take action can be defined to be any system command,

and for MQ, any MQSC command• Both TEP and Enhanced 3270UI have event consoles

OMEGAMON XE for Messaging - Brief OverviewHistorical data for reporting, trending, and analysis

9

• Near term historical workspaces in Enhanced 3270UI and TEP allow easy access to past status and statistical data for problem determination

• Messaging data can go to the Tivoli Data Warehouse, and Tivoli Common Reporting with Cognos can be used to create and automate performance and availability reports

OMEGAMON XE for Messaging - Brief OverviewSame monitoring solution across platforms

10

• Even in the z/OS Enhanced 3270UI, you can monitor your distributed platform queue managers• Just zoom on any white field for more data on any item impacting the health of your queue manager• Common attributes are available for all platforms for monitoring queue managers, queues, channels,

application connections, MQ events, messages, MQ clusters, brokers, message flows, and more…• Messaging agents on distributed platforms are available in ITCAM for Applications bundle product • Remotely monitor MQ Managed File Transfer environment on z/OS from agent in that bundle

What’s New in v7.3

12

OMEGAMON XE for Messaging v7.3 - What’s NewIncreased problem determination capability

• Understand information related to when the problem began with Enhanced 3270 User Interface Near-Term-History

• Easily see when a queue depth began to rise, and select for details

• Effortlessly explore data around when a channel goes into retry status

13

OMEGAMON XE for Messaging v7.3 - What’s NewIncreased problem determination capability

• Configure Near-Term-History collection for MQ attribute groups directly in the Enhanced 3270 User Interface

• Summary and detail Enhanced 3270UI workspaces support historical problem determination and trending analysis for the following objects: • Queue Managers• Queues• Channels• Applications, if enable Application Statistics, which also includes:

• Application Transaction/Programs• Application Queues

• Buffer Manager (Buffer Pools)• Log Manager• Message Manager• Topic Manager• Page Sets

14

OMEGAMON XE for Messaging v7.3 - What’s NewReduced time-to-resolution of problems

• Quickly view CICS or z/OS Enhanced 3270 User Interface embedded data for applications or queues, with automatic detection of application type

• MQ workspace includes relative CICS or z/OS monitoring data

• Click on any white field name to zoom to workspaces in the other OMEGAMON

• Easily navigate forward and backward among the workspaces

15


• Directly access queue manager z/OS address space monitoring data in Enhanced 3270 User Interface workspaces for both MSTR and CHIN

• Zoom from Current Queue Manager Status to this MQ workspace with z/OS CPU monitoring data

• Zoom on Job Name for direct access to the following z/OS options:

16


• Immediately display any MQ data available with MQSC display or ping commands from most MQ workspaces in the Enhanced 3270 User Interface

17

OMEGAMON XE for Messaging v7.3 - New Content• Enhanced 3270 UI for WebSphere MQ Monitoring Enhancements:

• Support Near-Term-History data from the PDS with historical workspaces to be able to identify changes to metrics over time

• Provide embedded data capability with other OMEGAMONs to enhance cross-product problem determination and resolution

• Add Topic Manager data workspaces• Improve detail workspaces for queue managers, queues, and channels including tabbed workspace organization and

parameter definitional data• Add Find command capability to find queue managers • Add MQSC display command support to issue any MQSC display or ping command • Add Namelist workspace• Reduce enterprise-wide query usage in workspaces for improved response time

• Other Enhancements:• Include new monitoring attributes from WebSphere MQ v7.x releases for queue managers, queues and channels • Provide support for IBM MQ v8 (WebSphere MQ renamed)• Provide support for IBM Integration Bus v9+ (WebSphere Message Broker renamed)• Add userid attribute to MQ Event data to increase usability of situations • Improve performance of z/OS agent queue manager data requests• Add dynamic tracing information to the product troubleshooting guide

• Distributed-platform-only Enhancements: • Windows: provide 64-bit native support and show queue manager start date and time • Update Error Log format to allow user figure out MQ installation information• Add basic MQTT channel data summary

• Use PARMGEN configuration including First Time User support

• Pre-req common level of ITM among OMEGAMON family: ITM 6.3 Fixpack 2

IBM MQ Monitoring

IBM MQ Monitoring Summary

19

• Queue manager availability, health and performance• Queue status, usage and statistics• Channel status and performance• MQ event monitoring and archival • Application connections and topology• Buffer pool, page set, message manager, log manager and topic manager statistics• Pub/sub topics and subscriptions and topology• Application (MQI monitoring) statistics• Dead letter queue and message manipulation• Queue sharing group status• MQ cluster monitoring• Supports ITM features with historical reporting, situations, event forwarding, and take-

action

Are all the MQ resources in your enterprise performing as expected for your messaging applications?

20

Queue Managers• General queue manager health assessment is based several factors such as:

• Availability of queue manager, channel initiator, command server

• Queue health related to high queue depths, transmission queues with messages, dead letter queue messages, and put or get inhibited queues

• Channel health related to current channels not running, percent active or current channels too close to maximums, in-doubt channels, and high numbers of client connections

• Parameters can be set to pass overrides to defaults used by the agent for health assessment so that the agent can assess as desired by user or site

• Historical recording allows assessing general health and status changes over time

21

Queue Managers continued• Select a queue manager on the health overview to see its health indicators more clearly

in Current Queue Manager Status, then zoom from any white field to get to more information supporting that field; note that the health overview also has a full list of options to get to all the various data

22

Queue Managers continued• Zoom from QMgr Status to go to the workspace allowing direct z/OS address space

monitoring for the queue manager master address space

23

Queue Managers continued• Zoom from Channel Initiator Status to the workspace about the channel initiator, then

additional zoom field Chan Init Active goes to the same z/OS address space monitoring for the channel initiator address space

Embedded Data from z/OS !

24

Queue Managers continued• Many queue managers in the enterprise can make the health overview list quite long, so

use the find command to narrow the list• Select any row in the resulting list to get to the Current Queue Manager Status

workspace, which leads to all data available for the queue manager• The find command for an exact queue manager name leads directly to that workspace

instead of the following list• Find commands can be used from most MQ workspaces, and if you do not remember

syntax, just enter “find”

25

Queue Managers continued• Both TEP and the Enhanced 3270UI can access the same sets of monitoring data• The user interfaces are quite different in style and focus

26

Queue Managers continued• Definitional detail parameter information confirms how your queue manager is configured

27

Queues• Queues with high depth are determined by default based on the high depth threshold set

for the queue, but queues with any depth may also be of interest for further exploration

28

Queues continued• Select a queue to see real-time queue status, including oldest message age on the queue

and several other interesting attributes (some attributes must be enabled in queue manager with MONQ setting)

• Applications with the queue open are also shown, this one has a batch application putting messages to it and a CICS application getting messages from it

29

Queues continued• Select the CICS application using the queue for direct access to CICS monitoring of it;

zoom on any white field for more details in the CICS product workspaces

EmbeddedData

from CICS !

30

Queues continued• Select the batch application using the queue for direct access to z/OS monitoring of it;

zoom on Job Name field for various options for z/OS address space workspaces

EmbeddedData

from z/OS !

31

Queues continued• Besides looking at queues with a depth, there are several ways to list queues by zooming

from Current Queue Manager Status, but also you can find any queue in the enterprise easily...

32

Queues continued• The H option for History available from any list of queues allows going back in time to

determine issues, like when a queue depth began to rise

• If needed, the history timespan can be changed from the 2 hour default using a View menu option

33

Queues continued• Select a row from the history summary for a convenient detail view that also includes

queue status history if enabled and messages were on the queue at the time • Easy one-click navigation at bottom to previous or next historical interval details

34

Queues continued• Transmission queues usually should not have messages remaining on them• Applications with queues opened for output may be the ones impacted most by a channel

issue causing a transmission queue to have depth

35

Queues continued• Select a transmission queue for detailed status, and immediately see the associated

channel and its status, which is selectable for further details

36

Queues continued• Regularly sampled queue statistics are available in both TEP and the Enhanced 3270UI• Recent interval samples also are available for determining very recent trends• Historical statistics are based on the sampled data, and where applicable, the attributes

are summarized over the historical interval

37

Queues continued• Queue definitional detail parameters allow understanding how your queues are configured

38

Channels• Channels that are not in running status, especially current channels that have been started,

may imply application communication is not occurring as expected• The Channel Not Running Summary quickly identifies any channels not in running status

39

Channels continued• Select the channel for channel status details, including the status of the remote queue

manager and channel status on that side, if applicable for the type of channel

40

Channels continued• The channel issue may also have been caught due to the rising depth of the transmission

queue associated with the sending side

41

Channels continued• Selecting the transmission queue for details, the channel status of running for the sender

channel does not explain the rising queue depth problem, but….

42

Channels continued• Selecting the channel for more details shows that the receiving side queue manager

seems healthy, but the receiver channel is paused so messages remain on the xmit queue

43

Channels continued• Looking for health issues on the receiving queue manager, there is a queue with high

depth; select it to see that the channel initiator is trying to put messages on it, but it is full or near full, causing the channel to pause each time it is full – the receiving application is too far behind in processing

44

Channels continued• Channel Initiator and Summary Statistics includes data about the channel initiator and also

has the channel summary data that is most useful when there are a lot of clients connected to a server connection channel, since it summarizes data for all instances

45

Channels continued• The H option for History available from any list of channels allows going back in time to

determine issues, like when a channel went into retry status

• Again, if necessary, adjust the timespan for the history display to show a longer period of time or a selected period of time

46

Channels continued• Select a row from the history summary for a convenient detail view• Easy one-click navigation at bottom to previous or next historical interval details

47

Channels continued• Regularly sampled channel statistics are available in both TEP and the Enhanced 3270UI• Recent interval samples also are available for determining very recent trends• Historical statistics are based on the sampled data, and where applicable, the attributes

summarized over the historical interval

48

Channels continued• Channel definitional detail parameters show how your channels are configured

49

MQ Events• MQ Events as shown in Enhanced 3270UI; select one for the most current data on resource

• The agent automatically monitors event queues, and also has options for sharing with other applications if needed

50

MQ Events continued• If enabled in the queue manager, MQ Events can be the best performing way for situations

to detect problems with queues and channels• The Event Archive feature captures events, including detail attributes, when configured for

historical collection, in order to facilitate event auditing, such as for configuration changes

51

MQ Events continued• The MQ Event Search menu item for the Event Archive workspace allows specifying

conditions for finding matching events• When the queue manager produces many kinds of events, this can be a big help for finding

particular events of interest

• The XML Event Details column contains all data included with the MQ event in XML format

• When there are many attributes, more than one row may occur for the same event to contain all the data

52

Application Connections• Find out which applications are connected to the queue manager, how they are

interconnected, and which queues they have open with the Application Connections workspace, which is an alternative workspace in the Queue Manager Status navigator item

• Topology notes queue in critical status because it is full

53

Application Connections continued• Link from node in topology to different forms of the topology, and zoom is available too• Here link is to Application Topology for Selected Queue to see applications with queue

open

54

Buffer Pools• Monitoring buffer pool statistics is recommended since buffer pool I/O can affect system

performance; these statistics come from MQ SMF type 115 data

55

Page Sets• Monitoring page set statistics is important to avoid a full page set preventing messages

being put to assigned queues, and to help check for buffer pool efficiency

56

Message Manager• Message Manager statistics indicate MQI call usage counted at the queue manager level

to understand how active the queue manager is processing requests

57

Log Manager• Examine Log Manager statistics to insure there are not problems with the writing to the

active log and to check for that log manager related parameters are sufficient

• Usage tip, click on white value to popup complete length

58

Topic Manager• Topic Manager statistics indicate publication and subscription activity counted at the

queue manager level to understand how busy the queue manager is with pub/sub

59

Publish Subscribe• The TEP Publish Subscribe navigator item contains several workspaces showing

definitions of topics and subscriptions and their status, including topic status with respect to subscribers and publishers

60

Publish Subscribe continued• MQ Search is a menu item available by right clicking on a row in a table in several of the

pub/sub workspaces, and the exact search performed is case sensitive to the workspace it is invoked from

• Search helps narrow down the set of topics and subscriptions of interest• There is also an enterprise-wide search capability for subscriptions

61

Publish Subscribe continued• Subscribing topology is shown on the Topic Status – Subscribers workspace• It includes nodes for topic, subscriptions, applications and destinations• Subscription node can link to Subscriptions Definitions and Subscription Status• Destination node can link to Queue Status and Application Topology for Selected Queue• Application node can link

to Application Topology for Selected Application

• Status of the destination node indicates status of the destination queue with respect to its current depth compared to its high depth threshold

• Remember that the same destination queue can be shared by multiple subscriptions

62

Application Statistics• The optional MQI monitoring feature can be configured for the monitoring agent to provide

three levels of application statistics for z/OS messaging applications• Application Summary shows current application connections on top subpanel (which does

not require MQI monitoring enablement), and application statistics on bottom subpanel

• Select a row for an application connection to get to embedded data from CICS or z/OS based on type (same data as available for open queues)

• Select a row for application statistics to see transaction/program and queue levels of the statistics

63

Application Statistics continued• Each level reports similar statistics, relative to the application, transaction/program within

application, or queue in use by application

64

Application Statistics continued• Use the “R” option for recent statistics samples (available at each level)

• Use the “H” option for historical at each level, for summary over historical collection interval

65

Application Statistics continued• All levels of data, within current, recent and historical workspaces, also are available in

TEP

66

Dead Letter Queue• Option D for a queue manager lists the messages on the dead letter queue

• For each message, options are to do a message manipulation action or view the contents

67

Dead Letter Queue continued• Deleting a message requires a confirmation

• Message forwarding for a DLQ message pre-fills the queue and queue manager name from the header information, so that by default, if these values are not changed, the message is “retried” to the original destination queue

• Actions and viewing contents require sufficient authority according to agent parameters and queue manager security settings

68

Message Manipulation• One of the options for queues is to display messages on the queue

• The same options as for DLQ messages are available

• Select a row to view the contents of the message

69

Message Manipulation continued• Two message manipulation actions are available on the options menu available for queues

• Clear Queue uses the MQSC CLEAR QLOCAL command• Purge Queue uses destructive MQGET MQI calls to clear the queue

• Message manipulation requires proper authorization via agent parameter settings and queue manager security settings; authorization can be at the queue level

• Note that TEP has the same capabilities with respect to the dead letter queue and message manipulation

70

Queue-Sharing Groups• Queue-sharing group nodes in the enterprise are listed on the health overview workspace• Only one agent monitors a queue-sharing group, but should that agent go offline, another

agent monitoring a queue manager in the group will take over monitoring it

• Select the node to see status related to queue managers in the queue-sharing group

71

Queue-Sharing Groups continued• Select a row to see all the Couple Facility Structure information from MQ side

72

Queue-Sharing Groups continued• Select a structure in the top subpanel to see details including z/OS CF information

• Options from zoom on CF Name

Embedded Data from z/OS !

73

Queue-Sharing Groups continued• From queue-sharing group queue manager status, use option “C” to see channels for the

queue-sharing group and directly get to data from queue managers about them• Here the “Q” option shows queues associated with the queue-sharing group

• Select a queue to see queue status and statistics from each queue manager with the queue (here only one queue manager shows, but there typically would be multiple)

74

MQ Clusters• The “G” option for a queue manager displays a workspace indicating cluster(s) associated

with the queue manager and other queue managers in the cluster(s)

75

MQ Clusters continued• Select a row for channels associated with the cluster related to the given queue manager

• Select channel for convenient display of status and statistics from each queue manager in the cluster (more options available for data from particular queue manager, like historical)

76

MQ Clusters continued• Use “Q” option for queues associated with the cluster related to the given queue manager

• Select a queue for convenient display of status and statistics from each queue manager hosting it in the cluster (more options available for data from particular queue manager)

77

MQSC Commands• Any MQSC command that makes a configuration change can be issued via take-action • Several actions are provided with the product, but any MQSC command can be used in

either the TEP or Enhanced 3270UI• In TEP, create a take-action command of type MQSeries

• In Enhanced 3270UI, the “Z” option on action menus allows entering any MQSC command

• Actions are logged by the agent and available in the MQ Action Log for auditing

78

MQSC Commands continued• Any MQSC display or ping command can be issued in the Enhanced 3270UI using the

“mqsc” command, which will invoke a workspace with the data to display • This same facility is used automatically in the “Parameters” tabs in the Enhanced 3270UI

to display parameters for queue managers, queues and channels

79

Product-Provided Take-Action Commands• TEP:

• Enhanced 3270UI:

80

Product-Provided Sample Situations

MQSeries_Alias_Queue_ProblemMQSeries_Appl_MQ_RespTime_HighMQSeries_Appl_MQI_Failures_HighMQSeries_Appl_Msgs_Put_HighMQSeries_Authority_ProblemMQSeries_Automation_ChlStartMQSeries_Automation_Fix_XmitQMQSeries_Bufpool_Buffer_Shrt_CMQSeries_Bufpool_Buffer_Shrt_WMQSeries_Bufpool_High_GetPag_IOMQSeries_Channel_Active_HighMQSeries_Channel_Autodef_ErrorMQSeries_Channel_Current_HighMQSeries_Channel_Initiator_ProbMQSeries_Channel_Instance_HighMQSeries_Channel_Out_Of_SyncMQSeries_Channel_Remain_IndoubtMQSeries_Channel_SSL_ErrorMQSeries_Channels_IndoubtMQSeries_ChInst_per_Client_HighMQSeries_CICS_Channel_StoppedMQSeries_Cluster_QMgr_SuspendedMQSeries_Conversion_ErrorMQSeries_Dead_LetterMQSeries_Delayed_Message_GroupMQSeries_High_Delayed_MessagesMQSeries_Inhibit_Problem

MQSeries_Listener_Not_StartedMQSeries_Local_DestQ_Depth_HighMQSeries_Local_DestQ_FullMQSeries_Local_Object_UnknownMQSeries_Logging_High_RBA_CritMQSeries_Logging_High_RBA_WarnMQSeries_Logging_SuspendedMQSeries_LogMgr_Buffer_WaitsMQSeries_LogMgr_High_Archv_ReadMQSeries_LogMgr_Reads_DelayedMQSeries_Manager_InactiveMQSeries_Manager_Inactive_EventMQSeries_MQ_Channel_StoppedMQSeries_No_Queue_Messages_ReadMQSeries_Old_Message_On_QueueMQSeries_PageSet_UnavailableMQSeries_PageSet_Usage_HighMQSeries_PubSub_ProblemMQSeries_QMgr_Instance_StandbyMQSeries_Queue_Depth_HighMQSeries_Queue_FullMQSeries_Queue_Full_UntilMQSeries_Queue_Manager_ProblemMQSeries_Queue_Manager_QuiesceMQSeries_Queue_Not_Being_ReadMQSeries_Queue_Service_Int_High

MQSeries_QSG_CFBackup_DelayMQSeries_QSG_CFStruct_FailedMQSeries_QSG_CFStruct_HighEntsMQSeries_QSG_CFStruct_HighMemMQSeries_QSG_CFStruct_InhibitMQSeries_QSG_DB2_FailedMQSeries_QSG_QMgr_FailedMQSeries_QSG_QMgr_NotStartedMQSeries_QSG_Queue_HighUseMQSeries_QSG_StrConn_FailedMQSeries_Remote_Queue_ErrorMQSeries_Remote_Queue_Problem

81

Historical Collection• Attribute groups recommended to be considered for enabling for historical collection for z/OS:

• Queue Long Term History* • Queue Status (only if you enable MONQ in queue manager) • Channel Long-Term History* • Channel Summary (only if you use client connections to SVRCONN channels) • Current Queue Manager Status• Buffer Manager Long-Term History* • Log Manager Long-Term History* • Message Manager Long-Term History* • Topic Manager Long-Term History* • Page Set Long-Term History* • Application Long-Term History* (only if you enable and use Application Statistics feature) • Application Queue Long-Term History* (only if you enable and use Application Statistics feature) • Application Transaction/Program Long-Term History* (only if you enable and use Application Statistics feature)

• Attribute groups recommended to be considered for enabling for historical collection for distributed platforms:• Queue Long Term History* • Queue Status (only if you enable MONQ in queue manager) • Channel Long-Term History* • Channel Summary (only if you use client connections to SVRCONN channels) • Current Queue Manager Status • MQI Statistics (only if you enable STATMQI in queue manager) • MQ Queue Statistics (only if you enable STATQ in queue manager, and if so, you may not want Queue Status historical in addition) • MQ Channel Statistics (only if you enable STATCHL in queue manager but do not enable MONCHL for Channel Long-Term History) • Error Log (but it is highly recommended to configure filters to reduce amount collected)

• Attribute groups marked with * above require HISTORY(YES) in agent parameters to populate sampled historical data.

IBM Integration Bus Monitoring

IBM Integrated Bus Monitoring Summary

83

• Broker availability and performance• Topology with status for broker, execution groups and message flows• Message flow node topology• Broker status events for broker deployment• Accounting Message Flow Statistics include message flow CPU and elapsed timings,

message rate and error counts • Resource Statistics provide Execution Group level statistics for JVM, Parsers, Files,

ODBC, JDBC and SOAP• The same attributes are available for all supported platforms• Supports ITM features with historical reporting, situations, event forwarding, and

take-action

Are your messages flowing through your brokers as expected?

84

Improved support for v7 and later brokers• New subnode, navigator, workspaces and tables

for the latest broker releases

• TEP navigator simplified into the following functional areas:

• Broker Status• Broker Status Events• Accounting Message Flow Statistics• Monitor Node Broker Statistics• Resource Statistics

• Several alternative workspaces in each navigator item to explore all the data in the functional area

• No changes for V6 brokers, which are now no longer in support

• WebSphere Message Broker subnode applies to all IBM Integration Bus releases and WebSphere Message Broker v7 and v8

85

Navigator item – Broker Status• Default Broker Status navigator workspace contains status and properties in the following views:

• Broker Topology• Broker Status• Execution Group Status• Message Flow Status

86

Broker Status continued• Application and Library categorization attributes are included in topology and for each

message flow listed

87

Broker Status continued• Link to the Execution Group Status workspace for a selected integration server (execution

group) to see its part of broker topology, status, and properties

88

• Link to the Message Flow Status workspace for a selected message flow to see its relative position in broker topology, status, and properties (including statistics enablement)

Broker Status continued

89

• Link to the Message Flow Node Topology workspace for any message flow in a topology or table view to see the message flow structure similar to what is displayed in the broker toolkit (defaultCollectNodeData agent parameter must be YES for this workspace)

Broker Status continued

90

• Broker Status Events workspace reports events published by the broker, including:

• Configuration changes for broker components• Status changes for broker components• Includes Configuration/status changes for applications or libraries

Navigator item – Broker Status Events

91

• These runtime statistics are captured automatically by the agent when the broker has been enabled to produce them with the mqsichangeflowstats command, at the interval defined via mqsichangebroker command parameter statisticsMajorInterval

• Default navigator workspace Archive Accounting Message Flow Statistics covers message flow performance with archive statistics (the recommended ongoing production monitoring)

• There are many alternative workspaces available by right-clicking this navigator item, or by linking from rows, and all levels archive/snapshot accounting statistics are available

Navigator item – Accounting Message Flow Statistics

92

• Link to the Archive Accounting Thread Statistics workspace for a selected message flow to see thread-level detail (one row per thread)

• Determine if you need to define additional instances for your message flow to increase throughput with more parallel processing via more threads

Accounting Message Flow Statistics continued

93

• Link to the Archive Accounting Node Statistics workspace for a selected message flow to see node-level detail (one row per node in message flow)

• Find out which nodes are using more CPU time or causing a delay in elapsed time for the message flow


94

• Link to the Archive Accounting Terminal Statistics workspace for a selected message flow or processing node to see terminal-level detail (one row per node terminal)

• See if the failure terminals in your message flow have been invoked or not


95

• Alternative workspace Snapshot Accounting Message Flow Statistics covers the same performance statistic attributes as archive, but the interval is a fixed 20 seconds, meaning it should be enabled for specific problem determination of a message flows, but not for ongoing monitoring


96

• Link to the Snapshot Accounting Thread Statistics workspace for a selected message flow to see thread-level detail (one row per thread)

• Determine how the threads are performing in 20 second intervals


97

• Link to the Snapshot Accounting Node Statistics workspace for a selected message flow to see node-level detail (one row per node in message flow)

• See how nodes are performing in 20 second intervals


98

• Link to the Snapshot Accounting Terminal Statistics workspace for a selected message flow or processing node to see terminal-level detail (one row per node terminal)

• See if failure terminals are being invoked in 20 second intervals


99

• These statistics are captured automatically by the agent when the broker has been deployed with one or more CandleMonitor nodes in message flow(s)

• Default navigator workspace Monitor Node Broker Statistics has CandleMonitor node statistics summarized at broker level, but only covering message flows containing the node

• There are many alternative workspaces available by right-clicking this navigator item, or by linking from rows, for various summarization levels

Navigator item – Monitor Node Broker Statistics

100

• Link to the Monitor Node Execution Group Statistics workspace to see the statistics summarized at the integration server (execution group) level

Monitor Node Broker Statistics continued

101

• Link to the Monitor Node Message Flow Statistics workspace to see the statistics summarized at the message flow level


102

• Link to the Monitor Node Sub-Flow Statistics workspace to see the statistics summarized at a sub-flow level as defined during the configuration of CandleMonitor nodes (requires configuration and is not automatic with sub-flow definition in the toolkit)


103

• Link to the Monitor Node Base Statistics workspace to see the statistics at the basic level reported by the configured CandleMonitor nodes


104

• Alternative workspace Monitor Node Events displays events generated by CandleMonitor nodes if configured to generate event messages by setting the eventMessage attribute

• Intended for use in failure, low-frequency paths of message flows in order to allow events to be generate in case of problems processing messages

• Events are automatically generated for exceptions detected by CandleMonitor nodes


105

• These runtime statistics are captured automatically by the agent when the broker has been enabled to produce them, with the mqsichangeresourcestats command

• Default navigator workspace JVM Resource Statistics reports on JVM memory and garbage collection for each integration server (execution group)

• There are alternative workspaces available by right-clicking this navigator item for other types of resource statistics, all of which are generated at 20 second intervals

Navigator item – Resource Statistics

106

• Alternative workspace Parser Resource Statistics reports on parser memory, read/write activity and failures pertaining to message trees and bit streams for each parser per message flow combination

Resource Statistics continued

107

• Alternative workspace File Resource Statistics reports on file read/write activity in the local file system as enacted by any type of file processing node


108

• Alternative workspace ODBC Resource Statistics reports on each ODBC DSN that has been accessed since the integration server (execution group) started


109

• Alternative workspace SOAP Input Resource Statistics reports on inbound messages the SOAPInput node is receiving and on replies the SOAPReply node is sending, indicating how many are successful or result in SOAP faults


110

• Alternative workspace JDBC Connection Pools Resource Statistics reports on each JDBC provider configurable service that uses connection pools


111

• WMB Start Broker • To start a broker. If the queue manager is down, it will also start the queue manager.

• WMB Stop Broker• To stop a broker.

• WMB Stop Broker and its Queue Manager• To stop a broker and the queue manager used by this broker; it stops only the broker on z/OS systems.

• WMB Change Trace Broker• To change the trace property for the broker’s object.

• WMB Change Broker• To change the broker’s configuration parameters.

• WMB Change Flow Stats• To change the Archive/Snapshot configuration (including enable/disable).

• WMB Change Properties• To change the properties for the broker or resources.

• WMB Start Message Flow(s)• To start execution group or message flows.

• WMB Stop Message Flow(s)• To stop execution group or message flows.

• WMB Reset Statistics• To reset all the statistics that are collected by the CandleMonitor node to 0s.

• WMB Create User Statistics• To create CandleMonitor user statistics for the Tivoli Enterprise Portal logon ID that issues the command.

• WMB Sample User Statistics• To make the agent take a sample of the CandleMonitor node statistics for the Tivoli Enterprise Portal logon ID that issues the command.

• WMB Destroy User Statistics• To delete all user statistics associated with the Tivoli Enterprise Portal logon ID that issues the command and stop collecting them.

• WMB Refresh Broker Data• To make the agent re-collect all broker configuration data from the broker.

Product-Provided Take-Action Commands

112

• WMB_Average_Flow_Time_High• The Message Flow Statistics.Average Flow Seconds value is found to exceed a

threshold.• WMB_Broker_Not_Started

• The Broker Status.Broker Status value is found to be Not Started.• WMB_Broker_Qmgr_Not_Connected

• The Broker Status.Queue Manager Status value is found to be Not Connected.• WMB_Monitor_Node_Events

• A Monitor Node Event occurs.• WMB_MsgFlow_Elapsed_Time_High

• The maximum time that any message takes to go through a given message flow during an archive statistics interval exceeds 100 milliseconds. You can create a new situation based on this situation using a threshold value that is appropriate for your environment. Use the predefined situation also as an example for what workspace links you can define with the Link Wizard for your custom situation event.

• WMB_Exception_Terminal_Invoked• Messages have gone down a failure path in a message flow. You can create a new

situation based on this situation using parameters that are appropriate for your environment. Use the predefined situation also as an example for what workspace links you can define with the Link Wizard for your custom situation event.

Product-Provided Sample Situations

113

• The following historical data workspaces can be linked to from workspaces discussed previously:• Historical Broker Status*• Historical Execution Group Status*• Historical Message Flow Status* • Historical Broker Status Events*• Historical Monitor Node Broker Statistics• Historical Monitor Node Execution Group Statistics• Historical Monitor Node Message Flow Statistics• Historical Monitor Node Sub-Flow Statistics• Historical Monitor Node Base Statistics• Historical Monitor Node Events • Historical Archive Accounting Message Flow Statistics*• Historical Archive Accounting Thread Statistics• Historical Archive Accounting Node Statistics• Historical Archive Accounting Terminal Statistics• Historical JVM Resource Statistics• Historical Parsers Resource Statistics• Historical File Resource Statistics• Historical ODBC Resource Statistics• Historical SOAPInput Resource Statistics• Historical JDBC Connection Pools Resource Statistics

• Blue* in above list signifies the most recommended historical collection to enable

Other Product-Provided Workspaces

• For archive accounting workspaces, the following recent archive accounting workspaces can be linked to: • Recent Archive Accounting Message Flow Statistics• Recent Archive Accounting Thread Statistics• Recent Archive Accounting Node Statistics• Recent Archive Accounting Terminal Statistics• The number of rows of data that are displayed in above workspaces depend on the value of the

defaultRetainRecentArchiveSamples parameter, which determines the minimum number of recent archive records to retain.

• For archive accounting workspaces, the following recent snapshot accounting workspaces can be linked to:• Recent Snapshot Accounting Message Flow Statistics• Recent Snapshot Accounting Thread Statistics• Recent Snapshot Accounting Node Statistics• Recent Snapshot Accounting Terminal Statistics• The number of rows of data that are displayed in above workspaces depend on the value of the

defaultRetainRecentSnapshotSamples parameter, which determines the minimum number of recent snapshot records to retain.

Other Product-Provided Workspaces

THANK YOU!

OMEGAMON XE for Messaging V730 Long client presentation

Software

Transcript of OMEGAMON XE for Messaging V730 Long client presentation