The Oracle TSAM monitoring console defines the Tuxedo components you want to monitor as well as track events and alerts. The Oracle TSAM console is composed of two sections:
Monitoring Center: Oracle TSAM administration, monitored data access, and alerts definition.
Recent Events: Tracks event or alert results defined in the monitoring center.
User Management allows you to control and set up who can access and manage the Oracle TSAM console. All authorized users are listed on the User Management page. You can add or delete users, as well as change passwords and descriptions.
Table 1-1 lists the User Management page properties.
Table 1-1 User Management Page
Name
Description
User Name
User name
Description
Describes user
Note:
A super-user admin is created (by default) without a password. When you log in for the first time, you are requested to enter the super-user admin password.
Add New User
Click the Add User button to add a new user. The Add New User page appears.
When you have finished entering the required information, click the Add User button to create a new user. If the information is entered correctly, you are returned to the User Management page and a new user is added to the user list.
If the information is not entered correctly, you must validate the information and re-enter.
The Back button discards new user information and returns you to the User Management page.
Change Password
To change a password, do the following steps:
Click the check box next to the name of the user whose password you want to change.
Click the Change Password button.
The Change User Password page appears.
Note:
Users can only change their respective password.
Note:
Only the super-user admin can change all user passwords.
Note:
The user name and description cannot be changed.
Table 1-3 lists the Change User Password page properties.
Table 1-3 Change Password Page
Name
Description
User Name
User name[4..15]
Old Password
Enter old password[4..15]
New Password
Enter password[4..15]
Confirm New Password
Re-enter pasword[4..15]
User Description
Describes user[0..50]
When you have finished entering the required information, click the Modify button to change the user password. If the information is entered correctly, you are returned to the User Management page.
If the information is not entered correctly, an error message appears.
Change Description
To change a user description, do the following steps:
Click the check box next to the name of the user whose description you want to change.
Click the Change Description button.
The Change User Description page appears.
Note:
Users can only change their respective description.
Note:
Only the super-user admin can change all user descriptions.
Table 1-4 lists the Change User Description page properties.
Table 1-4 Change User Description Page
Name
Description
User Name
User name[4..15]
User Description
Enter new user description[0..50]
When you have finished entering the required information, click the Modify button to change the user description. If the information is entered correctly, you are returned to the User Management page.
If the information is not entered correctly, an error message appears.
Delete User
Click the Delete User button at the bottom of User Management page is to delete users.
To delete a user, do the following steps:
Click the check box next to the user you want to delete.
Click the Delete User button to delete users.
Notes:
Only the super-user admin can delete users.
Note:
The super-user admin cannot be deleted.
Policy Management
Oracle TSAM provides comprehensive monitoring behavior control based on the Tuxedo infrastructure. Policy Management allows you to do the following:
Organize monitoring requirements into significant and useful monitoring policy set solutions
Manage monitoring policy sets (including import, export and usage tracking)
Define monitoring policy set entries for Tuxedo components
Selectively enable and disable real-time communication with monitored Tuxedo back end components
Monitoring Policy Sets Management
The Policy Management page allows you to Create, Edit, Clone, Import, Export, and Delete monitoring policies. Table 1-5 lists the Monitoring Policy Sets Management page properties.
Table 1-5 Monitoring Policy Sets Management Page
Name
Description
Latest Used Policy Set
Oracle TSAM tracks the usage of a defined monitoring policy set. The “latest used” lists the last policy set applied Tuxedo.Editing a policy does not impact this value.
Available Policy Sets
Oracle TSAM stores user defined policy sets on the server. All current available policy sets are listed. The directory which holds the policy set definition can be customized using the Oracle TSAM Web application file tsam.config.schemesdir in the web.xml.
Name
Name of the policy set.
Last Modified Time
Displays last time the policy set was modified.
Create
Creates a new monitoring policy set.
Edit
Edit an existing monitoring policy set. Only one monitoring policy can be selected at a time.
Clone
Creates a new monitoring policy set by cloning the contents of an existing one. Only one monitoring policy can be selected at a time. The policy set Name attributes are not copied. The latest policy action in the new policy set is disabled by default.
Import
Imports a policy set definition file from desktop and uploads to server.
Export
Exports policy set information into a policy set definition file.
Note:
Export can also be performed by clicking Policy Set Name
Delete
Deletes one or multiple policy sets
Table 1-6 lists the Monitoring Policy Set File Format properties.
Table 1-6 Monitoring Policy Set File Format
Element Name
Description
TSAM_POLICY
Root tag name of the policy set definition file
NAME
Name of policy set
DESCRIPTION
Policy Set description string
POLICY
A policy in a policy set. You can define multiple policies.
TUXOBJECT
Sub-element of POLICY. It represents which Tuxedo back end policy is impacted by the policy. Its format is:
Domain Machine Group Server
“Any” is a built-in key word which means any sub-components under the current component. For example, “TUXDOM1 Any APPGRP Any” indicates all servers in the “APPGRP” group of domain “TUXDOM1” are applied to the monitoring policy/
DEFINITION
The monitoring definition string. Has the same format as TMMONITOR.
For more information, see Oracle TSAM Agent, Monitoring Properties, in the Tuxedo System and Application Monitor (TSAM) Administration Guide.
STATUS
Indicates policy status. 1 indicates the policy is applied to a Tuxedo application. 0 indicate the policy is not applied.
COMMENTS
Comment string
Notes: Typically, you do not need to edit the policy set definition file. It is only referenced for special usage (for example, using a policy set definition to be generated with third party software).
Define a Monitoring Policy Set
Clicking Create, Edit, or Clone displays the Policy Set Definition page. Table 1-7 lists the Policy Set Definition page properties.
Monitoring Policy Set Entry Management
Table 1-7 Monitoring Policy Set Entry Page
Name
Description
Monitoring Policy Set
Displays current policy set information
Name
Name of the policy set. Name is required for “Create” and “Clone”. “Edit” does not allow name change.
Description
General policy set description
Overwrite
By default, “Create” and “Clone” fail if a policy set exists with the same name. Selecting this check box allows an existing file to be overwritten.
Save
Saves current policy set changes. When policy set contents change, the save button is highlighted. I f a policy set is closed before saving changes, the changes are lost.
Back
Returns to the Policy Management page
New
Creates a new policy
Edit
Edit a selected policy. Only one policy can be selected for editing at a time.
Enable
Initiates policy setting for Tuxedo components. If a policy is already “Enabled”, re-apply the policy.
Disable
Disables monitoring of a Tuxedo component. Cancels all monitoring settings even if the same component is covered by other Policy Management entries.
Delete
Deletes the selected policy set.
No.
The sequence of a policy in current policy set. It is auto-increased when a new policy is added.
Latest Action
Displays the policy set status. “Enabled” indicates the policy has been successfully applied to a Tuxedo application. “Disabled” indicates the policy is not applied a Tuxedo application.
Tuxedo Component
Specifies the Tuxedo component that the policy applies to.
For example, “Domain[TUXDOM1]/Machine[windows]/Group[APPGRP]/Server[simpserv:1]” indicates the following:
Tuxedo domain “TUXDOM1”
Machine “windows”
Group “APPGRP”
application server “simpserv”
server ID “1”
“Any” indicates all sub-elements of the component are selected. It is generated by the Tuxedo Object Selection panel
Specification
The TMMONITOR specification string for this policy. It is generated by the Policy Management Definition panel.
For more information, see Oracle TSAM Agent, Monitoring Properties, in the Tuxedo System and Application Monitor (TSAM) Administration Guide.
Define a Policy
To define a policy, do the following steps:
Select the Tuxedo component to be monitored.
Select the monitoring policy for the selected Tuxedo component.
The Policy Definition page consists of the Tuxedo Component section and the Policy Management Definition section.
Lists the current available Tuxedo domains. Only one domain can be selected per policy.
Machine
Lists the current available machines in the selected Tuxedo domain. “Any” indicates that all the machines in the domain are candidates for monitoring control.
Group
Lists the current available groups for a selected machine. “Any” indicates all available groups on the node are eligible for monitoring control.
Server
Lists the current available servers in the selected group. A specific server only can be selected when a particular group is selected. The server format is “server name:server id”. “Any” indicates all available servers in the group are eligible for monitoring control.
The Policy Management Definition page allows you to select the Tuxedo application type to want to monitor. There are four Tuxedo application types:
Table 1-9 lists the Policy Management Definition properties.
Table 1-9 Policy Management Definition
Name
Description
Enable
Check box
Click any or all of the four application check boxes to enable a specific application for monitoring.
Call Path
Ratio
The TMMONITOR.“appratio” policy. It accepts a value range of 1-65535. It indicates that monitoring is started among how many requests. The default value is “1” indicating each request is monitored.
Call Path
Interval
The TMMONITOR “appinterval” policy. It accepts a value range of 0-65535. It indicates a period of time (in seconds) that monitoring can be started. It is exclusive of ratio settings. The default value is “0” indicating that a time interval does not take effect.
Call Path
No Log
Click this check box to enable “appnolog”. For the selected component, the TMMONITOR plug-in is not invoked, that is the monitor data will not be collected.
This policy reduces the overhead caused by maximum level monitoring. tpgetcallinfo() can still be used to retrieve monitoring information.
Call Path Bridge Decode
Click this check box to enable the BRIDGE process involved in the call path monitoring. By default, BRIDGE will not decode the application message. If this policy is specified, BRIDGE decodes the application message and log the performance metrics.
Note:
If the network load is heavy, BRIDGE performance is significantly affected.
Service
Ratio
The TMMONITOR “svcratio” policy. It accepts a value of 1-65535. The default value is “1” indicating each service invocation is monitored.
Service
Interval
The TMMONITOR “svcinterval”. It accepts a value of 0 - 65535. It indicates a period of time (in seconds) that monitoring can be started. It is exclusive of ratio settings. The default value is “0” indicating that a time interval does not take effect.
System Server
Interval
The system server monitoring data collection interval (in seconds). It accepts a value of 30-65535. The default value is “300” indicating that for every 300 seconds, data collection occurs in the monitored system servers.
Transaction
Ratio
The TMMONITOR “tranratio” policy. It accepts a value of 1-65535. The default value is “1” indicating that every Transaction routine is monitored.
Note:
To guarantee a complete transaction trace, we do not recommend changing the default value
For more information, see Oracle TSAM Agent in the Tuxedo System and Application Monitor (TSAM) Administration Guide.
When you have completed your Policy Definition page selection, click Add/Update for your selections to take effect.
Table 1-10 lists the Save Entry and Comments properties.
Table 1-10 Save Entry and Comments
Name
Description
Add/Update
Adds or updates the selected policy to the current policy set. The new policy is disabled by default
Comments
Adds a comment to the policy
Import a Policy Set from Local Desktop
Click the Import button. The Import Policy page is displayed.
Table 1-11 lists the Import Policy page properties.
Table 1-11 Import Policy Page
Name
Description
File Name
The file name on local desktop
Browser...
Browse for a file on the local desktop
Overwrite
Click this check box to overwrite the policy set with imported file with the same name. By default, the uploaded file is not accepted if a policy set with the same name already exists.
Upload
Uploads the file from the desktop.
Close
Closes the import panel.
Note:
If an invalid policy set file is uploaded, an error message is displayed.
Monitoring Policy Set Limitations
Oracle TSAM does not check for Tuxedo component entries that overlap in the same or different policy set. If multiple policies are applied at the same time, the policies are applied to Tuxedo components in the order they are created.
Disabling a Tuxedo component policy also disables all other policy settings for that Tuxedo component.
Since TMMONITOR is also a mechanism that can be used to control monitoring behavior, Oracle TSAM does not guarantee consistency between Oracle TSAM monitoring policy settings and TMMONITOR settings.
For more information, see Oracle TSAM Agent in the Tuxedo System and Application Monitor (TSAM) Administration Guide.
Tuxedo Configuration
Tuxedo Configuration allows you to view and refresh Tuxedo domain configuration information. If Oracle TSAM is running, Tuxedo configuration information is received from the LMS at boot time.
Note:
If Oracle TSAM is not running at boot time, or if there are Tuxedo-side configuration changes, you must manually retrieve the configuration information.
Table 1-12 lists the Tuxedo Configuration page properties.
Table 1-12 Tuxedo Configuration Page
Name
Description
Domains
Lists Tuxedo domains in the following format: <DOMAINID>:<master machine name>:<IPCKEY>.
<DOMAINID> and <IPCKEY> are configured in the UBBCONFIG *RESOURCE section; <master machine name> is configured in the UBBCONFIG *MACHINE section.
View
Displays a visible representation of the domain.
Refresh
Manually retrieves the selected domain configuration information
Expand Tree
Displays the following detailed Tuxedo domain information: Domain ID (DOMAINID), Logical Machine ID (LMID) and Tuxedo version, Group name and group number (GRPNO), Server name, server ID (SRVID)
Show service
Lists the services provided by a given server.
View Tuxedo Domain Configuration
To view domain configuration information, do the following steps:
Select a domain from the Domains list.
Note:
Press and hold the CTRL key to select multiple domains
Note:
Some servers (including TMS, BRIDGE) that are not configured in UBBCONFIG may be displayed.
Click the View button.
Click the Expand Tree to display detailed domain information.
Click the Show Service check box to display service information.
Refresh Tuxedo Domain Configuration
To refresh a domain configuration, do the following steps:
Select a domain from the Domains list.
Note:
Press and hold the CTRL key to select multiple domains.
Note:
Some servers (including TMS, BRIDGE) that are not configured in UBBCONFIG may be displayed.
Click the Refresh button.
Any of the following possible results are displayed:
Successful: The configuration information is retrieved successfully, and the configuration tree is updated.
Failed: The “refresh” failed. Configuration information was not retrieved.
Timed out: Configuration information was not retrieved within a given time frame.
The default timeout is 30 seconds and is configured using the “tsam.config.timeoutwithtuxedo”parameter in the Oracle TSAM web.xml file.
Data Management
Data Management purges Oracle TSAM monitoring database.Data purge options are as follows:
Purge Tuxedo Component/Server/Service Data purges data based on the Tuxedo Component. Table 1-13 lists the Purge Tuxedo Component/Server/Service Data properties.
Table 1-13 Purge Tuxedo Component/Server/Service Data
Name
Description
From
Start time [DD MM YYYY HH MM]
To
End time [DD MM YYYY HH MM]
Scope
Purge Scope:
Tuxedo Component: Purges Tuxedo component and all relative data
Server Monitor: Purges Server Monitor data
Service Monitor: Purges Service Monitor Data
Domain
Domain name
Machine
Machine name
Group
Group name
Server
Server
Server Available: Lists all available servers
Server Selected: Lists server(s) selected for purge
Fast Search (Left/Right): Input partial server name, the cursor highlights the first matched server name
Service
Service
Service Available: Lists all available services
Service Selected: Lists service(s) selected for purge
Fast Search (Left/Right): Input partial service name, the cursor highlights the first matched service name
Purge Tuxedo Component/Server/Service Data
To Purge Tuxedo Component/Server/Service Data, do the following steps:
Select From and To to enter purge time period.
Click the Scope check box. If a scope is not selected, data cannot be purged.
Tuxedo Component: Purge the Tuxedo Component data and its database data
Note:
When Tuxedo Component is selected, From and To are ignored.
Server Monitor Data: Purge Server Monitor data from database, but not the Tuxedo Component
Service Monitor Data: Purge Service Monitor data from database, but not the Tuxedo Component
Select any of the following and click Purge.
Domain: Select a domain and click Purge. If Scope (Tuxedo Component) is checked, Purge deletes the selected domain from the database and all the Machine, Group, Server, Service and their monitoring data under this domain.
If Scope (Tuxedo Component) is unchecked, Scope (Server Monitor Data and Service Monitor data) is automatically checked. If Server Monitor Data is checked, Purge deletes all the Server Monitor Data under the selected domain. If Service Monitor data is checked, Purge deletes all the Service Monitor data under the selected domain.
Machine: same procedure as Domain.
Group: same procedure as Domain.
Server: same procedure as Domain.
Service: same procedure as Domain.
Notes:
If a Domain is selected, all the servers and services are populated to the Server Available list and Service Available list respectively. Use the > (single selection) or >> (all) button to add to the selected list. The Purge button only applies to the servers or services in the selected list.
Note:
Fast Search (Left/Right) helps to quickly locate a server or service.
Purge Call Path/Transaction Data
Purge Call Path/ Transaction Data purgesCall Path/ Transaction monitoring data. Table 1-14 lists the Purge Call Path/ Transaction Data properties.
Table 1-14 Purge Call Path/ Transaction Data
Name
Description
From
Start time [DD MM YYYY HH MM] of time period
To
End time [DD MM YYYY HH MM] of time period
Call Path/ Transaction
Call Path/ Transaction purges the following monitoring data:
Call Path
Transaction
Purge Call Path/Transaction Data
Call Path and Transaction monitoring data is independent of the Tuxedo Component.
To purge Call Path/ Transaction data, do the following steps:
Select From and To to enter purge time period.
Click the Call Path or Transaction check box.
Click the Purge button to delete the monitoring data for Call Path/Transaction from the database.
Purge All Data
Purge All Data purges all monitoring data in Purge Tuxedo Component/Server/Service Data and Purge Call Path/ Transaction Data.
Purge All Data
Click Purge All to clean all the monitoring data from database, include Tuxedo Component and Call Path/ Transaction.
Notes:
Only the super-user admin can purge monitoring data.
Note:
When you click the Purge button, the monitoring data is purged from the database. All other active sessions will be reminded to logout and re-login again to avoid data inconsistencies.
The Call Path Tree Page displays Tuxedo call path application request information when Application Monitoring is turned on.
When Application Monitoring is turned on, Tuxedo associates each call path request with a unique ID name (Correlation ID). The Correlation ID denotes where the request originated and remains unchanged when the request travels across Tuxedo.
A Correlation ID consists of the following field values separated by a space:
Tuxedo domain ID: <DOMAINID>:<master machine name>:<IPCKEY> in theUBBCONFIG *RESOURCE section. For more information, see Table 1-12.
Logical Machine ID: LMID in the UBBCONFIG *MACHINES section
Process Name:
“client”: used for native clients
“JSH”: used for Jolt clients
“WSH”: used for /WS clients
<server name>: used for the server name
Process ID (pid)
Thread ID
A counter in range of 1..99999999 (starting from 1)
Correlation ID Example
TUXEDO:lcsol18:200401 SITE1 client 18505 1 1.
Table 1-15 lists the Call Path Tree page properties.
Table 1-15 Call Path Tree
Name
Description
Query By Correlation ID
Searches call path information for given complete Correlation ID.
Filter
Selects specific search criteria.
Enable Paging
Lists found Correlation IDs.
Rows per page
Enter number of rows to display per page.
Initiator
Performs search based on domain, node, server, or client name.
Display Correlation ID Call Path Information
To display correlation ID call path information, do the following steps:
Fill in the Correlation ID under Query By Correlation ID.
Click the Query button to the right of the Correlation ID field.
The Correlation ID table appears.
Display Calls from an Initiator with Conditions
To display calls from an initiator with conditions, do the following steps:
Clear Correlation ID if it is not empty.
(Optional) Fill in the Initiator information (Domain,Machine, Server or Client).
Fill in Filter information if you want to limit the scope of the result. You can:
Specify Root Service to display calls for a particular service
Specify Min Elapsed Time to display calls started for the time the service is active
Specify the search scope by using the following return codes:
any: search all call paths regardless od return code
failed: search failed call paths
succeeded: search all successful call paths
Specify the search scope by using:
Active Call Path: search for unfinished call paths
Call Path for latest: search for call paths started no earlier than <n> minutes. A value(0) indicates search all call paths.
Call Path during: search for call path started during a time period.
Click the Query button to the left of Enable Paging field.
The Correlation ID table appears.
Table 1-16 lists the Correlation ID Table properties
Table 1-16 Correlation ID Table
Name
Description
No.
Sequence number. Starts at 1.
Correlation ID
The Correlation ID associated with the request.
Service
The service name invoked by the initiator.
Start Time
Request start time. The format is ‘YY/MM/DD hh:mm:ss:SSS’ (SSS = millisecond).
Execution Time
Time elapsed (in seconds) since the request is sent out and the reply is returned.
Status
Indicates finished request.
Indicates pending request.
Indicates failed request.
Note:
A call path is displayed as “timed out” if it remains active for more than 10 minutes. The timeout threshold can be changed using the <web.xml> tsam.callpath.timeout parameter.
Return Code
Displays a string representation and the tperrno value for the finished request. The display content is “TPOK/0” when there is no error.
GTRID
Displays the Global Transaction ID associated with the request, if any.
Clicking on the table column header will sort the table according to the selected column.
Clicking a Correlation ID displays more detailed request processing information. This detailed information is displayed in three sections:
Tuxedo request and reply are sent by the IPC message queue, or the network when the source and target are on different machine. The Call Path Tree lists the request originator (process name and PID), the Correlation ID (as its root), and the invoked service as its only child.
When you click on a Call Path Tree child, the Transport Detail table appears. Table 1-17 lists the Transport Detail table properties.
You can also display the call info, the server name, the physical message queue ID and logical queue name, and/or the server location information by clicking the corresponding check box (show call info, show server name, show queue info, show location).
Note:
A Call Path Tree child may also contain children if it makes calls when handling the request. A sequence number is displayed at the beginning of the service node of the Call Path Tree.
Transport Detail Table
Table 1-17 lists the Transport Detail table properties.
Table 1-17 Transport Detail Table
Name
Description
No.
Sequence number. Starts at 1.
From
The requesting application.
It consists of following fields separated by a space: DOMAINID:<hostname>:<IPCKEY>, LMID, Process Name, Server ID (SRVID in UBBCONFIG) if it is a server, and PID
To
The sending application.
It has the same format as “From” column.
Start Time
Transport start time. The format is ‘YY/MM/DD hh:mm:ss:SSS’ (SSS = millisecond).
Elapsed Time
Seconds spent on this transport.
Description
Description about the transport. It denotes the type of this transport (IPC queue or network) and the type of message (request or reply).
Service Summary Table
Table 1-18 lists the Service Summary table properties.
Table 1-18 Service Summary Table
Name
Description
No.
Sequence number. The number is also shown in the beginning of invoked service in the Call Path Tree.
From
The requesting application.
It consists of following fields separated by a space: <DOMAINID>:<master machine name>:<IPCKEY>, LMID, Process Name, Server ID (SRVID in UBBCONFIG) if it is a server, and PID.
Action
Call information.
To
The sending application.
It has the same format as “From” column.
Status
Indicates finished request.
Indicates pending request.
Indicates failed request.
Wait Time
Time elapsed (in seconds) from the time the request is sent until the time when it starts to be processed.
Exec Time
Time elapsed (in seconds) from the time the request is sent until the time the reply is returned.
GTRID
Global Transaction ID associated when the request is processed, if called in transaction mode or if AUTOTRAN is enabled.
Clicking the GTRID field in displays the Transaction Query page.
Configure Call Information Cache
All call paths started after Oracle TSAM boots are cached using 100000 (tsam.config.maxappsize) as the default cache size (number of call paths). A call paths is removed from cache if it is not finished within 600 seconds (tsam.config.maxappactive). If it finishes before 600 seconds, it remains cached for another 300 seconds (tsam.config.maxappdone) after it is finished.
The parameters are configurable in the Oracle TSAM web.xml file.
Table 1-19 lists the Call Information Cache configuration parameters.
Table 1-19 Call Information Cache Configuration Parameters
Name
Description
tsam.config.maxappsize
Max number in cache for a monitored application request.
Default value: 10000.
tsam.config.maxappactive
Max cache time (in seconds) for monitored application request in active status.
Default value: 600.
tsam.config.maxappdone
Max cache time (in seconds) for monitored application request in done status.
Each Tuxedo call triggers a set of service invocations. The back-end service operation is transparent to the caller. It may be a local service, remote host, and even a remote domain. A challenge for application monitoring is figuring out the exact call path tree (focusing on services). One type of service invocation tree is called a pattern. Oracle TSAM provides the capability to summarize a call pattern using collected performance data.
Define the Query Conditions
The call pattern statistics is a computing and database intensive task. Oracle TSAM issues as many query conditions as possible to limit the time used on pattern computing.
The service name of the first service invoked by the monitored request. It can only be selected when Domain is selected in the Initiator panel.
Minimum Request Number
The minimum number of requests that form a pattern. Usually a call pattern contains a large number of calls. This threshold prevents listing a pattern with a small number of calls. The default value is 100.
Request Only
The pattern is decided by the request message. The reply messages can be dropped to reduce the time used. Reply statistics are lost.
Time Window
The From and To enable the query window to second level. The default setting is the last day.
Click Submit.
Note:
This is a time intensive process. While you wait for results, you can continue with other monitoring tasks.
A simple string indicates the pattern. By clicking it can see the pattern tree.
Number of Requests
How many requests contained in this call pattern
Pattern Tree
The pattern tree is service focus and the nodes are the services name involved in this pattern. The root node is always “INITIATOR”. Click the Expand Tree button to expand the tree node.
Service on Pattern Tree
Click each service on pattern tree the transportation statistics to this service will be displayed
Average Transportation Time
The time (millisecond) is computed with all the requests contained in this pattern
Transportation Description
The transportation to a service might have following type:
IPC Queue (Request)
IPC Queue(Reply)
Domain (Request)
Domain (Reply)
BRIDGE (Request)
BRIDGE (Reply)
Get Pattern with Selected Conditions
Input Filter conditions, such as root service, time window etc.
Limit Initiator scope. Select Domain, Machine, Server or Client.
Click the Submit button to get pattern information.
When the pattern tree is generated click Initiator to expand all tree nodes.
Click each service to see service transportation statistics.
Service Monitoring provides a visual representation of the total number of successful and failed executions for selected services on a particular Tuxedo domain. The monitoring results are displayed in a chart which can be refreshed in specific time intervals (seconds).
Table 1-23 lists the Service Monitoring page properties.
Table 1-23 Service Monitoring Page
Name
Description
Domain
Domain name
Machine
Machine that the service run on
Group
Group name
Server
Server name
Most Active
Lists most active services
Time Window
Time span to aggregate the active services’ success and failure number
Service Available
Lists available services
Service Selected
lists selected services
Fast Search (Left)
Input any characters, then any matched service in Service Available will be highlighted.
Fast Search (Right)
Input any characters, then any matched service in Service Selected will be highlighted.
Initiate Service Monitoring
To initiate service monitoring, do the following steps:
Select a domain from the Domain drop-down list. When a domain is selected, the service(s) for the domain appear in the Service Available drop-down list.
For more specific Domain information, you can also selectinformation from the Machine, Group, Server drop-down lists. The default for each list is ANY, which indicates that all services are included.
There are two service selection options:
Select service(s) from Service Available to Service Selected. Use“>” to select individual services, or “>>” button for the entire list of services.
Select a number of top services from the Most Active drop-down list.
Note:
The max number of service that can be monitored at one time is 20.
Select Time Interval.
Click Submit. The live monitoring chart appears.
Click the refresh time interval check box above the chart to periodically refresh the chart.
The chart displays the total number of successful and failed service executions in a given time frame. The X-axis displays the service name.The Y-axis displays the number of requests handled during the most recent time period (in <n> minutes), where <n> is specified in the Time Window.
Service Historical Data Query provides a visual representation of the statistical information for an individual service. The monitoring results are displayed in a chart.
Table 1-24 lists the Service Historical Data Query page properties.
Table 1-24 Service Historical Data Query Page
Name
Description
Domain
Domain
Machine
Machine
Group
Group
Server
Server
Time Interval
Time span to aggregate the services’ success and failure number
Service
Service
From
Start time (DD MM YYYY HH MM) of time period
To
End time (DD MM YYYY HH MM) of time period
Fast Search
Input any characters, then any matched service is highlighted.
Initiate Service Historical Data Query Monitoring
To initiate service historical data query monitoring, do the following steps:
Select a domain from the Domain drop-down list. When a domain is selected, the service(s) for the domain appear in the Service Available drop-down list.
For more specific Domain information, you can also selectinformation from the Machine, Group, Server drop-down lists. The default for each list is ANY, which indicates that all services are included.
Select a service from the Service list box
Select Time Interval.
Select time period by From and To.
Click Submit. The live monitoring chart appears.
The chart displays the execution summary for a given service during the specified From and To time range. The X-axis displays the time interval. The Y-axis displays number of executions during the time interval.
Average Execution Time and Average Wait Time are also shown in the table.
Server Monitoring displays the Sever Live Monitoring information for the Tuxedo server, which includes the server snap and throughput for the given time period.
Note:
Server monitoring data includes both user monitoring data and Tuxedo system data.
Table 1-25 lists the Server Monitoring page properties.
Table 1-25 Server Monitoring Page
Name
Description
Domain
Domain name
Machine
Machine name
Group
Group name
Server
Server name.
Note:
Presently, only GWTDOMAIN and BRIDGE system servers are supported.
Available Remote Domains or Hosts
If the selected server is GWTDOMAIN, then it lists all Available Remote Domains under this selected server. If the selected server is BRIDGE, then it lists all Available Remote Hosts under this selected server.
Time Window
Time Window
Initiate Server Monitoring
To initiate server monitoring, do the following steps:
Make a selection from each of the following drop-down lists: Domain, Machine, Group, Server: Available remote domains and hosts are shown in the Available Remote Domains or Hosts drop-down list.
Make a selection from the Available Remote Domains or Hosts drop-down list.
By default, all remote domains or hosts are selected.
Select Time Window.
Click Submit. The server statistics chart appears.
Note:
The Submit button is enabled only when the Available Remote Domains or Hosts drop-down list is populated.
Note:
To periodically refresh the chart, click the refresh time interval check box.
The charts display server information for the past Time Window. The Snapshot chart displays real-time sampling statistical data. The Throughput chart displays accumulated statistical data during a given time period.
The X-axis displays the time. Table 1-26 lists the Y-axis display.
Table 1-26 Snapshot and Throughput Y-axis Display
System Server
Snapshot
Throughput
BRIDGE
Number of messages queued for the remote host at the time the data is sampled. A large value means a busy network.
Number of messages sent to the remote host during the last sampling period (specified using sysinterval).
GWTDOMAIN
Number of active outbound requests at the time the data is sampled.
Number of messages that have been sent to remote domain during the last sampling period (specified using Monitoring Policy Definition sysinterval).
Server Historical Data Query page displays Tuxedo server statistical information.
Note:
Server monitoring data includes both user monitoring data and Tuxedo system data.
Table 1-27 list the Server Historical Data Query page properties.
Table 1-27 Server Historical Data Query Page
Name
Description
Domain
Domain
Machine
Machine
Group
Group
Server
Server
Available Remote Domains or Hosts
If the selected server is GWTDOMAIN, then it lists all Available Remote Domains under this selected server. If the selected server is BRIDGE, then it lists all Available Remote Hosts under this selected server.
Accumulated number of messages sent by GWTDOMAIN or BRIDGE in the sampling period.
Message Size (bytes)
Accumulated number of the size of messages sent by GWTDOMAIN or BRIDGE in the sampling period.
Initiate Server Historical Data Query
To initiate a server historical data query, do the following steps:
Make a selection from each of the following drop-down lists: Domain, Machine, Group, Server: Available remote domains and hosts are shown in the Available Remote Domains or Hosts drop-down list.
Make a selection from the Available Remote Domains or Hosts drop-down list.
By default, all remote domains or hosts are selected.
Select Time Interval.
Select time period by From and To.
Click Submit. The server statistics chart appears.
Note:
Submit button is enabled only when the Available Remote Domains or Hosts drop-down list is populated.
The Server Historical Data Query chart displays throughput data for a given server during the specified From and To time range. The X-axis displays the time interval. The Y-axis displays the accumulated number of messages sent to a remote domain or a remote host during the specified Time Interval.
The Alert Management page lists current Oracle TSAM alerts. It allows you to “Add”, “Edit” and “Delete” Oracle TSAM alerts. Table 1-31 lists the Alert Management page properties.
Table 1-31 Alert Management Page
Name
Description
Name
Alert name. Must be given when this alert is created.
Type
Alert type. Oracle TSAM supports four alert types: “Call Path”, “Service”, “System” and “Transaction” which correspond to the four performance monitoring categories.
Description
Alert description.
Add
Add a new alert
Edit
Edit a selected alert
Delete
Delete selected alert(s)
Alert Definition
The Alert Definition page appears when you click the “Add” or “Edit” button.
Note:
“Edit” displays the settings of the selected alert. The name and type cannot be changed.
Table 1-32 lists the Alert Definition page properties.
Table 1-32 Alert Definition Page
Name
Description
Alert Name
Name of the alert. Must be given in order to create an alert
Alert Type
The type of alert to be created. Four types are supported:
Call Path: For call path monitoring related alerts
Service: For service monitoring related alerts
System: For system server monitoring related alerts
Transaction: For transaction monitoring related alerts
Alert Definition
Defines the items that generate an event. They are alert-type specific.
Call Path Type:
Elapse time.
The time elapsed since this request was monitored. An event is generated once the time threshold is reached. The unit is in milliseconds.
IPC Queue Length.
The request IPC queue length on the call path tree of the monitored request. It is an integer threshold that generates an event if any request queue length exceeds the defined value.
Execution Error.
Boolean threshold that generates events when Tuxedo service execution encounters error.
Service Type
Execution time.
The elapsed execution time for a service. An event is generated once the time threshold is reached. The unit is in milliseconds.
IPC Queue Length.
The monitored application server request IPC queue length. It is an integer threshold that generates events if the queue length exceeds the defined value.
Execution Error.
Boolean threshold that generates events when Tuxedo service execution encounters error.
Waiting Time in Queue.
Numeric threshold that generates events when the waiting time in the IPC queue goes beyond the defined value. The unit is in milliseconds.
System Type
Network Link Lost.
An event is generated when a network link is lost. If the link is not connected, the event is generated periodically by the monitoring interval.
Transaction Type
Transaction Call Failure.
The XA call failure results in event generation based on Transaction Call Failure criteria. For example, if an xa_commit() return code is less than 0, a Transaction Call Failure event is generated."
Severity
Four security levels can be selected:
Information
Warn
Critical
Fatal
Action
The action defines which action need to be performed when the evaluation is true. Current only Tuxedo Event is supported. When Tuxedo Event is selected, Tuxedo Event Name and Tuxedo Event Message must be supplied.
Description
A description string for this alert
See Also
Event Query
Events
The Events page displays current Oracle TSAM-generated events. The Show All Events button lists all the events.
Click the ShowFilter button; the Filter Conditions panel appears. After the Show Filter button is clicked, it changes to the Hide Filter button. Click the Hide Filter button; the Filter Conditions panel is hidden.
Click the Submit button in the Filter Conditions panel to list events based on the filter criteria.
The bottom-left panel of the Oracle TSAM console lists a summary of generated events. Clicking the event Severity in the panel lists the events on the Recent Events page. The event details page layout is similar to the Alerts/Events page.
Note:
Recent Events only include Events that were generated during the last Event Cache Time.
Clean the Event in Cache
Clicking the Clean buttonclears all the events stored in cache. Events can also be queried on the Alerts/Events page.
Customize the Event Cache Time
You can specify how long an event remains stored in cache. Enter the required time by changing the tsam.config.windowevt value in the Oracle TSAM web.xml file. The default value is 60 minutes.
