In this section: |
You can monitor every server type and change its operation parameters through the Web Console. Those changes can affect the behavior and performance of the server.
This topic describes how to monitor the following:
It also describes how to perform certain administrative tasks on a server.
You can view the system and system memory status from the Status page.
The Status page opens, as shown in the following image.
You can also access the last agent trace by selecting Status from the Monitor section of the ribbon.
In this section: |
From the Web Console, you can:
As you monitor Data Service activity, you can also perform the following administrative tasks from the Web Console:
This topic describes how Server Administrators can display the Data Service Agents page for all Data Services or for an individual Data Service. Server Administrators access the page from the Workspace on the Web Console menu bar, using the Agents menu option for Data Services.
If you are a user other than a Server Administrator, you can access the same page and the same statistics from My Console on the Web Console menu bar. With the Manage My Agents option, you can monitor and manage your own agents, that is, the agents that match your user ID. If granted the applicable privileges, you can also monitor, or monitor and manage, the agents of the other users in your group.
The Data Services Agents page opens on the right.
The table in Data Service Agents Performance Parameters describes all the statistics that you can display on the page.
The pane that opens lists the available statistics, grouped by category, such as General, Client Information, and FOCUS Information. A description of each statistic is provided in the right column of the pane, to help you tailor the page for your needs.
Select the check box for each statistic that you want to display on the page, and de-select the check box for each statistic that you do not want to display.
Optionally, you can:
Click OK when you have made your selections.
The following image shows the General and Client Information sections.
The following image shows the FOCUS Information, Adapter Information, and Change Data Capture Info sections.
The following image shows the File Listener Info and System sections.
The pane that opens lists the values for the State statistic. Select the check box for each value that you want to reflect on the page, and deselect the check box for each value that you do not want to reflect. For example, if you select the check box for in use and the check box for crashed, only the agents whose State is in use or crashed will be displayed on the page.
Optionally, you can:
The following image shows the pane from which you can filter the values for State on a Data Service Agents page. In this example, the agents whose State is in use or crashed will be displayed on the page.
You are returned to the Data Service Agents page. The following image shows a sample customized report with selected statistics. In this example, the agents identified as Tscomid 1 through 6 were killed, as described in How to Kill One or More Data Service Agents.
In the following image, the cursor is pointing to the Refresh Now button.
Since agents are associated with services, you may wish to display current service statistics.
The Data Service Agents page associated with the selected service is displayed on the right. The table in Data Service Agents Performance Parameters describes all the statistics that are provided on the page.
The following statistics can appear on a Data Service Agents page. Each agent is represented as a separate row on the report.
General
Identification number associated with the Data Service Agent.
Name of the data service for the agent.
Current state of the agent. Possible values are in use, DBMS call, idle, aborted, crashed, stopping, or starting.
A session connects and is in use until the moment that it disconnects. This includes all time spent between a session suspension and resumption when the process is not using the CPU but still has resources allocated for at least one session.
DBMS call indicates that the agent is waiting for a database to return data.
Idle means that the agent is not connected. It has no sessions at all, whether active or suspended.
Aborted and crashed are abnormal states resulting from a fatal software error detected by the program (aborted) or by the operating system (crashed). The agent process is no longer running, and these states are provided for diagnostic purposes. The Server Administrator can clear (using the kill option) such agents after the problem is investigated.
Stopping and starting are normal transitory states, which are self-explanatory.
Client Information
User ID associated with the server connection.
Group ID associated with the server connection.
Session ID uniquely identifying the client session.
For some protocols, the network address of the connected client.
Character code page used by the client connected to the session.
Unique identifier assigned to the submitted deferred request.
Unique identifier assigned to the last submitted scheduled or CMASAP request.
FOCUS Information
Last time that a request to the agent was made. This value is used to calculate the time that an agent is idle, in cases in which an idle agent limit was set.
First eight characters of the last instruction executed by the data access agent.
Displays the procedure currently being executed or completed by each agent. For example, when a procedure, which is a part of a job, executes another procedure, the name of this procedure is shown in a Procedure column.
Displays calling procedures or calling flows executed or submitted by agents.
Note: This column is not shown by default on the Data Services Agents page. To display it, click Choose Columns, select its check box on the Select the columns that will appear on the Agents page, and click OK.
Name of the last Master File that was processed by the agent. The displayed name can be up to 64 characters.
Last Java class executed. This value includes explicitly called Java classes (such as CALLJAVA), and implicitly called classes (such as GRAPH FILE).
Time from the moment that the server receives the request (regardless of the requesting tool, that is, Web Console, DataMigrator, or WebFOCUS) until the request completes and the response is sent to the requester.
Response time minus DBMS processing time.
Last &RETCODE that is set upon procedure (FOCEXEC) completion. Zero (0) is normal. Any other value indicates an abnormal completion.
Number of FOCUS I/O operations performed by the agent.
Unique identifier assigned to the last request that used the foccache directory.
Number of transactions or HLI commands performed by the agent.
Adapter Information
Number of external I/O operations performed by the agent.
Amount of DBMS processing time in seconds.
Change Data Capture Information
Suffix of the listening adapter.
Data source that is listened on.
Current status of the adapter.
Number of processed LUWs (files).
Interval at which polling takes place.
Amount of time during which there is no activity, and after which a timeout occurs.
Maximum number of LUWs (files) that are processed by the adapter in one request.
File Listener Information
Suffix of the listening adapter.
Data source that is listened on.
Extension of processed files.
Current status of the adapter.
Number of processed LUWs (files).
Number of bytes processed by the file listener adapter.
Number of selected files waiting to be processed by the file listener adapter.
Rounded number of seconds that the file listener adapter spent in processing files.
Rounded number of seconds that the file listener adapter spent in waiting for files.
Average number of files per second that the file listener adapter processed.
Average number of bytes per second that the file listener adapter processed.
Average time in seconds that the file listener adapter spent in waiting for files.
Interval at which polling takes place.
Amount of time during which there is no activity, and after which a timeout occurs.
Maximum number of LUWs (files) that are processed by the adapter in one request.
System
Total CPU time used by the process.
Amount of memory, in kilobytes, used by the process.
Disk space, in kilobytes, used by the process.
Operating system specific process identification number that is associated with the agent.
The Agent Statistics pane opens. On this pane, you will see the statistics described in Data Service Agents Performance Parameters, followed by system-extended statistics for the selected agent process.
The following table describes the system-extended statistics for an individual Data Service Agent.
Statistic |
Description |
---|---|
PageFaultCount PeakWorkingSetSize (KB) WorkingSetSize (KB) QuotaPeakPagedPoolUsage QuotaPagedPoolUsage QuotaPeakNonPagedPoolUsage QuotaNonPagedPoolUsage PagefileUsage PeakPagefileUsage
|
Standard Windows memory management statistics for the monitored process. |
Total Execution Time |
Cumulative execution time of transactions performed by the agent. |
You can display and customize a status report for all Data Services, and display aggregated statistics for all Data Services or for a selected service.
The Data Services Status page is displayed on the right, as shown in the following image.
The table in Aggregated Statistics for Data Services describes all the statistics that you can display on the report.
The Data Service Statistics pane opens on the right. The following image shows a sample Data Service Statistics pane.
The table in Aggregated Statistics for Data Services describes the statistics that are provided.
The aggregated statistics for the selected service are displayed on the right. The table in Aggregated Statistics for Data Services describes the statistics that are provided.
The following aggregated statistics are provided for Data Services:
General
If you are displaying aggregated statistics for all Data Services, this value is All Services.
If you are displaying aggregated statistics for a selected service, this value is the name of the service configured for Workspace Manager in the edaserve.cfg file.
Performance
Number of agents that are currently running.
Maximum number of agents that were running at any given time.
Number of attempted connections to servers since they were started, or from the time that statistics were reset.
This value is different from the total number of sessions, since a persistent session can connect multiple times for different requests.
Number of successfully completed connections to servers since they were started, or from the time that statistics were reset.
Average number of seconds for server response. This value does not include data transfer time. See Calculating and Displaying Averages, for more information.
Response time, averaged over the time interval defined by the edaserve.cfg parameter stats_runningavg. The default value is 600 seconds.
Average DBMS processing time in seconds.
Average DBMS processing access time in seconds.
Average response time minus DBMS processing time in seconds.
Average response time minus DBMS processing access time in seconds.
Number of currently queued connections. Not applicable when queuing is off.
Currently running jobs are included in the connection total, but not in connections completed or in resource failures.
Number of connections that went into the queue and timed out (not available when queuing is off).
Maximum number of connections that are queued at any given time (not available when queuing is off).
Average number of seconds spent in queue by all connections (not available when queuing is off).
Errors
Number of connections rejected for lack of available agents. This value includes timed-out queued connections.
Number of connections that failed due to invalid credentials. For details about errors and failures, see the edaprint.log file, which is saved where the server is installed.
Number of completed connections with WebFOCUS errors during request execution. For example, the adapter was not properly configured, or the data description or procedure was not found.
System
Total CPU time used by the server.
Amount of memory, in kilobytes, used by the server.
Total disk space, in kilobytes, used by the server.
Variables that are averages (avg) are fractional numbers of seconds rounded to the nearest millisecond for display purposes, but actually computed in higher precision, depending on the operating system:
The accuracy of the three corresponding averages is limited only by the precision of the operating system. In the rare case in which a machine is faster than the precision of its time-measuring, accuracy side effects may occur:
The Connection Queue displays the statistics for the current list of connections, enabling administrators to monitor, and if necessary, cancel individual connections.
A connection refers to a physical connection between Client and Server. There are two types of connections:
A queued connection waiting for an agent becomes active as soon as an agent is available. If the maximum time to wait in the queue is reached, the connection is automatically cancelled by the Workspace Manager.
The Connection Queue report is displayed on the right.
The table in Connection Statistics describes the statistics on the report.
Each connection is represented as a separate row on the report.
The following statistics appear on the Connection Queue report.
Name of the data service for the connection.
Connection type, either connect or resume.
User ID that connected to the session.
Type of authentication used to connect, either trusted, explicit, or IWA.
Unique identifier for the network connection created by a listener.
Time at which the connection was activated or queued. This value determines when a queued connection times out.
Network address of the connecting client.
Unique identifier assigned to the last request that used the foccache directory.
To cancel a connection, click the row for the connection on the Connection Queue report, and from the menu, click Kill Connection, as shown in the following image.
You can perform certain administrative tasks as you monitor Data Service activity.
You can clear aggregated Workspace Manager statistics for Data Services and start collecting them from the beginning.
You can save the last FOCUS command executed by an agent as a new procedure and run it separately from the Web Console. This provides a great tool for debugging server issues.
The Create New Procedure as Text page opens, as shown in the following image.
The Save Procedure page opens, as shown in the following image.
There are a number of ways to kill Data Service Agents:
The running agent is terminated, invalidating any current connection to that agent. If a request is then issued from such a connection, an error message is returned. Stopping an agent is therefore an emergency administrative measure, as it disrupts the application. After an agent is terminated, the Data Services Agents pane refreshes automatically. The corresponding row of the terminated agent remains, and the State is listed as stopping until the row eventually disappears. You can also click Refresh Now to update the pane and remove the row.
Important: Use this option only when Kill This Agent fails.
In the Web Console, look for a row for the agent in the Data Services Agents pane that is currently connected to the desired user ID, then right-click that row and select Kill Agents With The Same User ID, as shown in the following image.
To kill agents from the command line, the syntax is:
edastart -killuser <userid>
In addition, for the z/OS platform, a z/OS operator MODIFY command can also be used, as follows:
/F <server_jobname/started task>,-killuser <userid>
Use this method to start additional agents, even when there are agents already running.
On the Data Services Agents pane, you will see that the new agents are listed, with the State of starting.
If a procedure generates an error, you can display the error message on the active agent.
The Data Service Agents pane opens.
The latest error is displayed in a separate window.
The following error was generated on an agent when a procedure was run against a file that could not be found. Clicking the number of the agent that ran the procedure produced the following message:
(FOC205) THE DESCRIPTION CANNOT BE FOUND FOR FILE NAMED: XYZ BYPASSING TO END OF COMMAND
Tip: The error message is also stored in the Session Log, which you can access by clicking the Show Session Log Window icon on the Web Console menu bar. In the Session Log, you can click the message to see expanded information, including troubleshooting suggestions. For example:
(FOC205) THE DESCRIPTION CANNOT BE FOUND FOR FILE NAMED: No data description can be found for the file requested. Check the Master File Descriptions allocated or available, or the spelling of the file name.
For more information on the Session Log, see Showing the Session Log.
In this section: |
From the Web Console, you can:
As you monitor Java Service activity, you can also perform the following administrative tasks from the Web Console.
This topic describes how Server Administrators can display a Performance Report for the agents for all Java Services or statistics for an individual Java Service Agent.
The Java Services Agents pane opens on the right.
The following image shows a sample report.
The table in Performance Report for Java Service Agents describes the statistics displayed on the report.
In the following image, the cursor is pointing to the Refresh Now button.
The following statistics appear on the Performance Report for Java Service Agents. Each agent is represented as a separate row on the report.
Unique identification number assigned to each Java Service Agent. You can start and kill JSCOM3 connections. Each started connection is issued a unique number. If you kill a connection, its number is retired. A new connection is assigned an entirely new Jscomid.
Port number for each Java Service Agent. Numbering starts with a base number and increases by one for each active agent.
Current status of the instance. Possible values are starting, active, stopping, or stopped.
Operating system specific process identification number associated with each Java Service Agent.
Total number of clients connected to JSCOM3, both active and inactive.
Active client connections to JSCOM3.
Name of the user who executed the class.
Last executed Java class. This includes explicitly called Java classes (such as CALLJAVA), and implicitly called classes (such as GRAPH FILE).
Amount of memory, in kilobytes, available to the Java Virtual Machine (JVM).
The Statistics pane opens, as shown in the following image. On this pane, you will see the statistics described in Statistics for an Individual Java Service Agent.
The following statistics are for an individual Java Service Agent.
General
Name of the Listener that is serving the Java Service.
Additional property of the Service or Listener, where available.
Current state of the connection. Possible values are starting, active, stopping, or stopped.
Operating system specific process identification number associated with the Java Service.
Port number for the Java Service. Numbering starts with a base number and increases by one.
System
Standard Windows memory management statistics for the monitored process.
Cumulative execution time of transactions performed by the agent.
Connections
Unique identification number assigned to the Java Service Agent. You can start and kill JSCOM3 connections. Each started connection is issued a unique number. If you kill a connection, its number is retired. A new connection is assigned an entirely new Jscomid.
Total number of clients connected to JSCOM3, both active and inactive.
Total number of active client connections to JSCOM3.
Name of the user who executed the class.
Last executed Java class. This includes explicitly called Java classes (such as CALLJAVA), and implicitly called classes (such as GRAPH FILE).
You can display aggregated statistics for all Java Services.
The Java Services Statistics pane opens on the right, as shown in the following image.
The table in Aggregated Statistics for Java Services describes the statistics that are provided.
You can perform certain administrative tasks as you monitor Java Service activity.
You can clear aggregated Workspace Manager statistics for Java Services and start collecting them from the beginning.
On the Performance Report, you will see the status stopping. If you click Refresh Now, you will see the status stopped.
Tip: You can restart the agent by right-clicking the applicable row on the Performance Report and clicking Start from the menu.
Tip: You can restart the agents by right-clicking the service in the navigation pane, and clicking Start All from the menu.
On the Performance Report for all Java Service Agents, you will see that the number of agents that you specified is removed from the list.
Use this method to start one or more additional agents, even when there are agents already running.
On the Performance Report for all Java Service Agents, you will see that the new agents are listed, with the Status of starting, as shown in the following image.
In this section: How to:
Reference: |
From the Web Console, you can:
As you monitor Special Service and Listener activity, you can also perform the following administrative tasks from the Web Console.
The Performance Report for all Special Services and Listeners opens on the right, as shown in the following image.
The table in Performance Report for All Special Services and Listeners describes the statistics displayed on the report.
The following statistics appear on the Performance Report for all Special Services and Listeners.
Type of Special Service (for example, FDS, DFM, or SCHEDULER), or type of Listener (for example, HTTP or TCP).
Status of the Special Service or Listener. Possible values include active, not active, or stopped.
Operating system specific process identification number associated with each Special Service or Listener running in a Workspace Manager address space.
Port number or directory associated with the Special Service if applicable (for example, FDS port or DFM directory), or the port number associated with the Listener.
Total number of requests processed by the Special Service or Listener.
Total CPU time used by the process, displayed as hours:minutes:seconds.milliseconds (for example, 0:00:05.625).
Amount of memory, in kilobytes, used by the process.
The Statistics pane opens. On this pane, you will see the statistics described in Statistics for an Individual Special Service or Statistics for an Individual Listener.
Tip: You can also display the Statistics pane for FDS, HTTP, TCP or SOAP by expanding the Special Services and Listeners folder, right-clicking one of these Special Services or Listeners, and clicking Statistics from the menu.
The following statistics appear for an individual Special Service.
General
Type of Special Service (for example, FDS, DFM, or SCHEDULER).
Additional property of the Special Service, where available.
Current state of the Special Service connection. Possible values are starting, active, stopping, or stopped.
Operating system specific process identification number associated with the Special Service.
Applies to the Special Service FDS.
Port number for the FDS. Numbering starts with a base number and increases by one.
System (Applies to the Special Services FDS and SCHEDULER)
Standard Windows memory management statistics for the monitored process.
Cumulative execution time of transactions performed by the agent.
Connections for FDS
Total number of active connections.
Hard-coded limit for the number of connections (512).
Total number of opened File Control Blocks (FCB) by all users.
Hard-coded limit for the number of opened FCBs at the sink side (4096).
User ID of the last connection to the FDS Service.
Last command executed by the FDS Service.
Connections for SCHEDULER
Current status of the Scheduler. For example, scanning means that the Scheduler is in the process of scanning the application path for scheduled jobs, and ready means that the Scheduler is ready to process a job.
Timestamp for the last full scan of the application path. A full scan is performed on the Scheduler Start and Force Rescan options, which are available on the Scheduler menu.
Timestamp for the last incremental scan. An incremental scan looks for files that have changed since the last scan. The last scan can be full or incremental, depending on which one happened last.
Timestamp for the next incremental scan. An incremental scan is based on the value specified by the Scheduler property sched_interval, which is in seconds.
Number of jobs currently scheduled to run.
Number of all runs since the startup of the Scheduler.
Number of all jobs submitted through CMASAP since the startup of the Scheduler.
DFM for SCHEDULER
Disk space, in kilobytes, in the directory dfm_dir that is available for the DFM Service write results of deferred requests.
Number of requests processed by the DFM Service since server startup.
Number of responses in the directory dfm_dir that are ready for a user to view (get).
The following statistics appear for an individual Listener.
General
Type of Listener, based on its network protocol (for example, HTTP or TCP).
Additional property of the Listener, where available. For example, an HTTP Listener can be CLM-enabled.
Current state of the Listener. Possible values include active, not active, and stopped.
Operating system specific process identification number associated with the Listener.
Port number for the Listener. Numbering starts with a base number and increases by one.
Total number of requests processed by the Listener.
System
Standard Windows memory management statistics for the monitored process.
Cumulative execution time of transactions performed by the agent.
The Deferred List displays the current status of all deferred requests. You can stop the queuing or execution of deferred requests, and remove requests from the queue.
The Deferred List is displayed on the right, as shown in the following image. In the image, the Get and Delete menu options for an individual row are shown.
Each row displays statistics for a procedure that has been submitted for deferred execution. For details about these statistics, see Deferred List Statistics.
Optionally, you can:
The following statistics appear on the Deferred List.
Unique identifier that the server assigns to each deferred request submitted.
ID of the user who requested deferred execution of the named procedure.
Can be one of the following: Ready, Queuing, Executing, Agent crashed, Connect failed, Exceed limit, or Unknown.
Name of the application that contains the procedure being run in deferred mode.
Name of the procedure being run in deferred mode.
Time at which the request for deferred execution was submitted.
Time at which the deferred request is scheduled to run.
Number of days that the generated report will be kept on the server.
The output is displayed on the right.
You can display a Performance Report for all web sessions and statistics for a single web session.
or
In the navigation pane, expand the Special Services and Listeners folder, right-click HTTP, and select Web Sessions, as shown in the following image.
The Performance Report for all web sessions opens on the right.
The table in Performance Report for All Web Sessions describes the statistics displayed on the report.
The following statistics appear on a Performance Report for all web sessions.
Unique number assigned to a web session at logon to the server until the end of the session. The session ends by logging off the server or by closing the browser.
A unique number that the server assigns to a specific user for the duration of the user session. Every time you connect to the Web Console, a new session number is assigned.
ID of the connected user.
IP address or machine on which the web browser running the Web Console resides.
Time at which the web session originated.
Last time at which the user was active in this web session.
Mechanism that performs user authentication at the time the web session is created.
Number of messages that a user received in this session.
The Web Session Statistics pane opens.
The table in Statistics for an Individual Web Session describes the statistics displayed on the pane.
The following statistics appear for an individual web session.
Unique number assigned to a web session at logon to the server until the end of the session. The session ends by logging off the server or by closing the browser.
ID of the connected user.
Number of HTTP requests processed during the current web session.
Time at which the web session originated.
Last time at which the user was active in this web session.
Type of user logon, for example, explicit or cookie.
Standard HTTP client that is used for connection to the HTTP Listener.
Uniform Resource Identifier (URI) of the last Web Console page that was processed by the current web session.
IP address or machine on which the web browser running the Web Console resides.
A unique number that the server assigns to a specific user for the duration of the user session. Every time that you connect to the Web Console, a new session number is assigned.
You can perform several administrative tasks for Special Services and Listeners.
In the following image, the Quiesce menu option is selected for the TCP listener.
The Performance Report or statistics pane reflects the changed status:
For more information, see Showing the Session Log.
How to: |
Custom Pages let you create pages that monitor the various services available from the Workspace folder, as well as run procedures. These pages are represented by icons on the navigation pane of the Applications page.
Custom Pages allow you to include multiple monitoring pages on a single page, creating a combination of Data Agents, Java Services, Listener and Special Services, Cluster Manager, and Resource Management pages. They also allow you to include procedures that generate reports. These pages and reports can be arranged horizontally, vertically, or in a cascade. Refresh and autorefresh will update all sections of a Custom Page. Custom Pages can be saved in a server application for future use.
Note: Custom Pages are only supported when the appearance of the Web Console is set to an RIA option.
Custom Pages are created in Application directory folders.
The Custom Page workspace opens, as shown in the following image.
The Select Page(s) to Add dialog box opens, showing the Monitor Pages folders.
The monitor page is added to the Selected Page(s) field, as shown in the following image.
The monitor page is added to the workspace.
The Select Page(s) to Add dialog box opens, showing the application folders.
The procedure is added to the Selected Page(s) field, as shown in the following image.
The procedure is added to the workspace.
The pages in the following image are arranged in a cascade.
The Save Custom Page opens.
Optionally, you can replace an existing page by selecting the Overwrite existing file check box.
You can add other pages at any time using the Add new windows (+) button next to the Arrange Windows drop-down menu.
The Custom Page is added to the Application folder. Right-click and select Open to open the page.
iWay Software |