Job Management Partner 1/Performance Management User's Guide
You can view a summary of results for operating status, stopped status, and Normal and Abnormal status counts to check the operating status of servers and agents, as well as the alarm status of agents. You can also view alarm and agent events with Abnormal status and Warning status. A view showing summarized results together with Abnormal and Warning status events is called a summary display.
- Organization of this subsection
- (1) Prerequisite conditions
- (2) Agent types for which summary displays are supported
- (3) Procedure for displaying a summary
- (4) Operational status classifications
- (5) Counting unit and counting range for a summary display
- (6) Printing a summary display
(1) Prerequisite conditions
The following prerequisites must be met prior to using the summary display.
- Requirements for server operating status monitoring
- To be able to monitor server operating status, the version of the PFM - Manager for the connection destination must be 08-11 or later, the version of PFM - Web Console must be 09-00 or later, and the health check function must be enabled.
- Requirements for agent operating status monitoring
- To be able to monitor agent operating status, the version of the PFM - Manager for the connection destination must be 08-11 or later, the version of PFM - Web Console must be 09-00 or later, and the health check function must be enabled. In addition, you must set the following health check agent properties in the Service Properties window of the PFM - Web Console Services tree window:
- Monitoring Level in Health Check Configurations: Service
For details on the health check function, see 13.2 Using the health check function to check the operating status of monitoring agents and their hosts.
(2) Agent types for which summary displays are supported
The following table lists the types of agents for which summary displays are supported, according to window type.
Table 3-5 Types of agents for which summary displays are supported
Agent type Window type Server Operational Status window Agent Operational Status window Agent Alarm Status window PFM - Agent Yes Yes Yes Remote agent Yes Yes Yes Group agent No No Yes Remote Monitor Collector service Yes Yes No
- Legend:
- Yes: Count supported
- No: Count not supported
Summary displays are not supported for the following agents:
- Any agent that has been deleted from the Products tree in the Agents tree window using the jpctool service delete command, but remains in the User Agents tree.
To update the agent trees to reflect the results of the jpctool service delete command, restart PFM - Manager and PFM - Web Console after executing the command.
(3) Procedure for displaying a summary
To check the operational statuses by displaying a summary:
- From the monitoring console browser, log on to PFM - Web Console.
The main window appears.
- In the navigation frame of the main window, select the Agents tree tab.
The Agents tree window is displayed.
- From the Display format pull-down menu in the navigation frame, choose the display format for the Agents tree.
The Agents tree appears in the selected display format.
- When User Agents is selected:
The Agents tree that has User Agents (logged-on-user-name) as the root appears.
- When Products is selected:
The Agents tree that has Products as the root appears.
- In the navigation frame, select the folders to be counted for the summary display.
Depending on whether you select the root of the Agents tree, or you select a desired folder other than the root, the counting unit and counting range for summary display differ. For details on the counting unit and counting range for summary displays, see 3.3.5(5) Counting unit and counting range for a summary display.
- From the method frame, select the View Summary button.
The System Operational Status Summary window appears in the information frame. The following figure shows an example of the System Operational Status Summary window.
Figure 3-9 Example of the System Operational Status Summary window
The items displayed in the Server Operational Status, Agent Operational Status and Agent Alarm Monitoring Status views are described below.
: Indicates that an Operational Status or Monitoring Status view is currently displayed.
: Indicates that an Operational Status or Monitoring Status view is not currently displayed.
The following table describes the items displayed in the Server Operational Status or Agent Operational Status views.
Table 3-6 Items displayed in the Server Operational Status and Agent Operating Status views
Item Meaning Pie chart A pie chart showing the operating status of the servers or agents in the folder selected in the navigation frame. The meanings of the colors are as follows:
For details on operational statuses, see (4) Operating status classifications.
- Green: Percentage of operating servers or agents.
- Red: Percentage of stopped servers or agents.
- Blue: Percentage of servers or agents with Unconfirmed status#1.
Table A table listing the operating status of the servers or agents in each folder selected in the navigation frame. The meanings of the items in a table are as follows:
The table shows the servers or agents with the five largest stopped counts, in descending order. The remaining servers are shown as Others.
- Number operating: Number of servers or agents that are operating normally.
- Number stopped#2: Number of servers or agents that are currently stopped (if the number stopped is one or greater, the cell is colored red).
- Total: Total number of servers or agents (including those with Unconfirmed status).
- Operating rate: Percentage of operating servers or agents (green indicates the percentage operating, red indicates the percentage stopped, and blue indicates the percentage with Unconfirmed status#1).
You can specify in the initialization file (config.xml) the number of servers or agents to be displayed. For details on the config.xml, see the chapter that describes installation and setup in the Job Management Partner 1/Performance Management Planning and Configuration Guide.[User tree view] check box This check box appears in the Agent Operational Status view when you select User Agents in the Agents tree. Selecting this check box displays the operating status at the folder levelthat the logged-on user created in the Agents tree. [Product type view] check box This check box appears in an Agent Operational Status view when you select User Agents in the Agents tree. Selecting this check box displays the operating status at the folder level where the agents are grouped according to the PFM - Agent product or the PFM - RM product.
- #1:
- For details on how to take action when operational statuses become unknown, see 14.2.5(2) The operating status of a server or agent is Unconfirmed or Not Supported.
- #2:
- In the Server Operational Status window, when the number of Stopped is one or more, clicking the number displays the Stopped Host List window. If you do this, you can check the host names of the stopped servers. The following figure shows an example of the Stopped Host List window.
Figure 3-10 Stopped Host List window
The following table describes the items displayed in the Agent Alarm Monitoring Status view.
Table 3-7 Items displayed in the Agent Alarm Monitoring Status view
Item Description Pie chart A pie chart showing the alarm monitoring status of agents in each folder selected in the navigation frame. Each color in the chart indicates an alarm status#. The meanings of the colors are as follows:
The numbers in the pie chart indicate the number of agents for each alarm status. The numbers do not indicate the number of bound alarms.
- Green: Percentage of agents with Normal alarm status.
- Yellow: Percentage of agents with Warning alarm status.
- Red: Percentage of agents with Abnormal alarm status.
Table A table showing the alarm monitoring status of agents in the folder selected in the navigation frame. The meanings of the items in a table are as follows:
The table shows agents with the five greatest Abnormal status counts in the Alarm Monitoring Status view, in descending order. The remaining agents are shown as Others.
- Normal count: Number of agents with Normal alarm status.
- Warning count: Number of agents with Warning alarm status.
- Abnormal count: Number of agents with Abnormal alarm status (if the number of Abnormal alarm status agents is one or greater, the cell is colored red).
- Status ratio: Alarm event status ratio (green indicates the percentage with Normal status, yellow indicates the percentage with Warning status, and red indicates the percentage with Abnormal status).
You can specify the number of alarms to be displayed by setting the maxDisplayAlarm parameter in the initialization file (config.xml). For details on the config.xml, see the chapter that describes installation and setup in the Job Management Partner 1/Performance Management Planning and Configuration Guide.
- #
- The alarm status is determined as follows:
- Normal
- All bound alarms show a Normal status or there are no bound alarms.
- Warning
No bound alarm shows an Abnormal status, but at least one bound alarm shows a Warning status.
- Abnormal
At least one bound alarm shows an Abnormal status.
- When the Always notify check box is selected for an alarm, the alarm always shows a Normal (green) status, because the alarm is not evaluated. However, remote and group agents are evaluated. The Remote Monitor Collector service is not evaluated because an alarm cannot be bound to it and it is not counted as a parameter in the pie chart.
- If more than one alarm is bound to a single agent, these alarms are evaluated in order of priority. The priority starting from the highest is abnormal, warning, and normal.
- Reference note:
- The number of agents is counted as follows:
- (Example 1) An agent has six alarms bound to it, where two alarms each show Normal, Warning, and Abnormal statuses.
- The agent is classified with an Abnormal status of the highest priority. Thus, the agent is counted as an Abnormal agent.
- (Example 2) An alarm for which the Always notify check box is selected is bound and the alarm status is Abnormal.
- An alarm status for which the Always notify check box is selected is classified as Normal. Thus, the agent is counted as a Normal agent.
You can check for Abnormal and Warning status alarm events and agent events in the Events view. Unlike the Event Monitor window that displays all alarm and agent events that have been issued, the Events view displays only those alarms and agent events that have been issued with an Abnormal or Warning status and are currently pending. Any alarm event that is returning to Normal status is not shown. For details on displayed contents of each event, see 7. Displaying Events or see the description of the Event Monitor window in the manual Job Management Partner 1/Performance Management Reference.
The target alarms of the Agent Alarm Status window are those for which the Always notify check box is not selected. For details, see 6.9.3 Notes on evaluating alarms.
(4) Operational status classifications
The meaning of the operating statuses shown in the Server Operational Status and Agent Operational Status views are described below.
(a) Operating statuses in the Server Operational Status view
The operating statuses displayed in the Server Operating Status view are determined based on the health check results from an agent monitoring the server. The health check results for the server are classified into three operating statuses: Operating, Stopped, and Unconfirmed.
In the Server Operational Status view, a single operating status is shown for each server. If more than one agent monitors a single server, the health check results from the agents might differ. In such a case, the health check results determined to have the highest priority in the range based on (5) Counting unit and counting range for summary display is shown as the operating status of the server in the Server Operational Status view.
The following table lists the agent health check results, resulting operating statuses, and their priority levels.
Table 3-8 Agent health check results, resulting operated statuses and priority levels
Agent type Health check result Judgment of operational status Judgment priority
- PFM - Agent
- Remote Monitor Collector service
Host Not Available Stop 1 Not Supported Running 2 Running 3 Incomplete 4 Stopped 5 Unconfirmed 6 Remote agent Host Not Available Stop 7 Running Running 8 Incomplete 9 Unconfirmed Unknown 10 Not Supported 11 The following is an example of where the health check results from a PFM - Agent differ from those of a remote agent, while both are monitoring the same server.
In this example, the following health check results are assumed:
- The health check result from the PFM - Agent is Host Not Available.
This health check result is determined to have priority level 1.
- The health check result from the remote agent is Not Supported.
This health check result is determined to have priority level 11.
The operating status for the server is determined as follows:
- If PFM - Agent and the remote agent are in the same folder
For both the pie chart and the table, Host Not Available with the higher priority is selected. Therefore, the operating status of the server is Stopped.
- If PFM - Agent and the remote agent are in different folders
For the pie chart, Host Not Available with the higher priority is selected. Therefore, the operating status of the server is Stopped.
For the table, the result depends on the folder where the agent is located. A Host Not Available status is selected in the folder where PFM - Agent is located. Therefore, the operating status of the server is Stopped. A Not Supported status is selected in the folder where the remote agent is located. Therefore, the operating status of the server is Unconfirmed.
(b) Operating statuses in the Agent Operational Status view
The operating statuses displayed in the Agent Operational Status view are determined based on the health check result from an agent. The health check results are classified into three operating statuses: Operating, Stopped, and Unconfirmed.
The following table lists the agent health check results and the resulting operating statuses.
Table 3-9 Agent health check results and resulting operating statuses
Agent type Health check result Judgment of operational status
- PFM - Agent
- Remote agent
- Remote Monitor Collector service
Running Running Incomplete Stop Stopped Host Not Available Not Supported Unknown Unconfirmed
(5) Counting unit and counting range for a summary display
The following table lists the counting units and counting range for a summary display.
Table 3-10 Counting units and counting range for a summary display
Selected folder in the Agents tree Pie graph /table Counting unit Counting range Desired folder Pie graph Selected folder All of the agents under the selected folder Table Folders directly under the selected folder
(The selected folder will be counted as Other.)All of the agents under the folders directly under the selected folder
(The agents directory under the selected folder will be counted as Other.)Figure 3-11 and 3-12 show the summarized units and ranges for summary display.
Figure 3-11 Summarized units and ranges for summary display (when a route is selected)
Figure 3-12 Summarized units and ranges for summary display (when a folder is selected)
(6) Printing a summary display
To print a summary display, click the Stop button on the System Operation Status Summary Monitoring window and click the Print button. An example of the System Operation Status Summary Monitoring print window is shown below.
Figure 3-13 System Operation Status Summary Monitoring print window
All Rights Reserved. Copyright (C) 2009, Hitachi, Ltd.