지금 지원 담당자와 채팅
지원 담당자와 채팅

Foglight for Oracle (Cartridge) 5.9.4.10 - User Guide

Installing and Configuring Agents Using Foglight for Oracle
Viewing the Databases Dashboard Assigning Instances to Users Selecting an Instance to Monitor Foglight for Oracle Overview Dashboard Overview view Advisories view SQL Performance Investigator (SQL PI) Oracle Activity Drilldown Pluggable Databases Drilldown Storage Drilldown Reviewing Configuration Settings Reviewing the Alert Log Reviewing Monitored Data Guard Environments Reviewing ASM Instances Reviewing Exadata-related Information
Administering Foglight for Oracle Reporting Reference Glossary

Viewing the Foglight for Oracle Real-time Summary Page

Viewing the Foglight for Oracle Real-time Summary Page

The following image displays the main elements of the Foglight for Oracle real-time summary page. The table that follows the image identifies each element, and provides a link to display more information, corresponding to the logical dataflow within Foglight for Oracle Instance Homepage.

The main groups (panes) of gathered icons and gauges, which represent the main activity areas in the Oracle connection process, are as follows:

Components representing instance data flow — the main activity area in the Foglight for Oracle Instance Homepage includes several panes and flows that represent the data flow in the Oracle operation. The dataflow is represented in a top-down design, that is, from the session to the physical disk storage.

Use the Instance identification indicators to identify the instance, its type and its operation period.

These indicators are:

The indicators included in the Instance pane provide various details about the instance. All of the indicators display real-time behavior, color-coded for severity.

The Sessions pane monitors all session types, that is: system, user, and Oracle sessions. Use this pane to view the response time compared with the number of sessions and the instance’s level of activity. A high response time value may result from a long queue, that is: an overly high percentage of active users.

Long queues can indicate one of the following issues:

The Sessions pane displays the total number of sessions, distributed according to the following distinctions:

The parameters in this pane lead to the Activity > Sessions panel. For details, see Viewing Detailed Sessions Data .

The Client applications represented graphically in the Sessions pane communicate with Oracle by sending and receiving network packets and by submitting SQL statements for execution by Oracle. The flows, described in the following table, help indicate performance issues if their values are too low.

The rate per second of calls (both user and recursive) that executed SQL statements.

The rate at which network packets are being received by Oracle from Client applications.

When this icon’s color is yellow, clicking it displays text that describes the deviations that triggered this display, along with a representation of the rate at which Oracle is encountering network packet errors.

The rate at which network packets are being sent from Oracle to Client applications.

When this icon’s color is yellow, clicking it displays text that describes the deviations that triggered this display, along with a representation of the rate at which Oracle is encountering network packet errors.

Use the Data Guard section for monitoring Oracle Data Guard on primary databases. This section displays the gap between the last archive log on the primary database and last applied archive log on the standby database.

Oracle Data Guard can be configured as either physical or logical standby. For details about the preparations required for monitoring each of these setups, see Monitoring the Data Guard physical standby and Working with logical standby (SQL Apply) .

As Oracle Data Guard supports physical standby databases that use Redo Apply technology, use this section to monitor the rate of applying redo log files on standby database instance.

10
Click Verify Connectivity

The Workload chart displays the workload of the average active sessions, plotted over time.

The workload can be plotted as either:

To display the wait events breakdown, click the arrow to the right of the Breakdown link.

The Process Activity pane enables the monitoring of the processes ran by the Oracle instance, using performance indicators such as the total number of blocked processes and the rate per second of parses that are carried out during the most recent sample.

The Process Activity pane displays the following parameters.

The number of statements executed during the most recent sample.

The average number of blocked processes that were collected during the specified time range.

Blocked processes are processes that are waiting for another process to release a resource that the process is currently locking. Such processes can sometimes result in bottlenecks.

The Blocked Process indicator changes its color when one or more processes become blocked.

Total number of parse calls. This figure is the total of both hard parse calls (calls that require loading the SQL source code into RAM for parsing) and soft parse calls.

The number of currently running transactions.

For further details, see Reviewing Instance Usability .

To communicate with the Oracle memory, the Oracle processes use logical reads and soft and hard parses, represented as flows in the homepage.

The Memory Activity session provides a view on the Oracle instance’s memory utilization, as described in the following table.

Total (MB)

The actual size currently allocated to the System Global Area (SGA) memory.

Buffer Cache (MB)

 

The amount of memory currently allocated to the buffer cache. The buffer cache, which is used for caching frequently accessed data blocks, is normally the largest memory cache used by Oracle.

Shared Pool (MB)

The size of the shared pool. The shared pool caches SQL statements, PL/SQL programs, object definitions, and session memory for MTS sessions. Performance can improve when the shared pool is sized properly, thereby avoiding re-parsing and reloading.

Buffer Cache Hit (%)

The percentage of block requests, which found the block in the buffer cache, out of the entire block requests (including requests that required disk reads).

 

The effectiveness of the library cache, computed by comparing the number of queries submitted to Oracle server for execution, which found the parsed execution plan in the library cache (soft parsing), with the entire number of queries submitted during the specified time range (including queries that required the parsing of a new statement, also known as “hard parsing”).

Total (MB)

The actual size currently allocated to the Program Global Area (PGA) memory.

PGA Hit (%)

The percentage PGA cache hit ratio.

When the value of this metric is 100%, it indicates that all work areas executed by the system processed the entire input data using exclusively the PGA memory, thereby making optimal use of this memory. Any lower value indicates that at least one work area required one or more extra passes over the input data.

The Storage pane contains the components listed below, which display storage status. The information shown in this pane is aggregated to the RAC level.

If archiving is configured, the following indicators display storage-related information about the most critical archive destination, that is: the archive destination with shortest time to failure.

This section contains the following indicators:

If storage is managed using Automatic Storage Management (ASM), and the ASM instance is installed and configured, the following indicators display storage-related information about ASM:

The Disk Activity section displays the following indicators:

The real-time summary page contains flows and graphic images that are used for tracking physical I/O activity, as listed below.

All of the indicators display real-time behavior, color-coded for severity. Clicking each indicator displays a popup, which shows the metric’s value, plotted over time.

Reviewing Instance Usability

The Instance Usability panel displays aggregated information regarding the instance’s usability and availability.

This panel supports the following tasks:

The Instance pane contains the following indicators:

The Alarms pane displays the usability-related outstanding alarms. These alarms alert the following issues per instance: response time, instance availability, instance connection availability, OS connection availability, instance connection time, and Listener status.

The Instance Availability section displays the instance availability in percentages, plotted over the time range. If the instance was unavailable once or more during this time, the relevant message describing the last occurrence of unavailability appears in the Error Message box.

The Listener Availability section displays the availability of all listeners configured for the instance, in percentages, plotted over the time range. If one or more of the listeners were unavailable during this time, a message describing this issue appears under the Listener Errors section.

The Connection Time section displays the instance’s connection time, in seconds, plotted over the time range.

The Response Time section displays the instance’s response time, in milliseconds, plotted over the time range.

Reviewing Cluster Usability

Use the Cluster Usability panel to compare between the various RAC nodes, regarding the following indicators:

To investigate a single instance, select the requested instance from the list at the top of the panel. The display switches to instance usability. For details, see Reviewing Instance Usability .

Reviewing Locks Activity at the Cluster Level

The Activity Locks panel for a RAC comprises the following panes:

The Locks Activity pane displays a breakdown of all lock types and their general properties.

Lock Name

The name of the lock.

Gets

The total number of enqueue requests or enqueue conversions for this type of enqueue.

Gets Rate

The rate of lock gets.

Gets (percent)

The percentage of lock gets for this type of lock within the total number of locks.

Enqueue Waits

The total number of times an enqueue request or conversion resulted in a wait event.

Waits Rate

The number of wait events per second.

Wait (percent)

The percentage of wait events for this type of lock within the total number of locks.

Lock Wait

The total amount of time spent waiting for the enqueue or enqueue conversion.

Average Lock Duration

The average length of time for a lock, in milliseconds.

Lock Efficiency

The percentage of lock requests that did not escalate to a blocking lock.

Alarms

The number of warning, critical, and fatal alarms that were invoked for this Oracle database instance.

Holding the cursor over each alarm type displays a dwell that shows the most recent alarms invoked for this database instance, sorted by severity.

Click each alarm type to drill down and manage the alarms. See the Foglight Online Help, Monitoring System-Wide Alarms for details on the alarm information.

The Lock Tree (Current) pane displays current data, which is available only in real-time mode.

Clicking any locking or locked session table row displays a popup that is used for viewing further details. For information see Session Details Popup .

PDB

In multitenant architectures, the name of the pluggable database where the lock took place.

Session

The identifier for a session that holds or acquires the lock.

Serial

The session’s serial number. The combination of SID and serial number provides a unique identification of the session.

DB User

The database user name.

Object Wanted

The object requested for lock.

Lock Type

The type of user or system lock.

The locks on the user types are obtained by user applications. Any process that is blocking others is likely to be holding one of these locks. The user type locks are:

Lock Mode

The lock mode in which the session holds the lock. This row can have one of the following values:

Lock Request Mode

The lock mode in which the process requests the lock. This row can have one of the values listed below.

Lock Duration

The duration of the lock in seconds.

Cross Instance

Defines whether the lock is between different cluster nodes:

Instance

The name of the instance to which the blocking or blocked session belongs.

Kill Session

Used for terminating the selected session, by means of the Kill Session dialog box that is displayed upon clicking this column.

 

SID

The session identifier

Serial

The session’s serial number. The combination of SID and serial number provides a unique identification of the session.

Process

The process number

DB User

The database user name

OS

The operating system name

Program

The name of the program that is processing the SQL statements

Client

The Client computer identifier

Host

The host computer identifier

Current SQL

The SQL text of the SQL statement that is currently being processed in this session

The Locks Historical Tree pane displays the overall locks activity that took place within the selected time range, as well as locks that exceeded a predefined threshold.

This pane comprises the following sections:

관련 문서