MySQL Enterprise Monitor Glossary

MySQL Enterprise Monitor Access Control enables you to manage asset visibility, application administration, data access, roles, and users.

For more information, see Chapter 23, Access Control.

See Also role.

Advisors filter and evaluate the information collected by the Monitoring Agents and present it to the Events page when defined thresholds are breached. Advisors can be in one or more Categories.

The MySQL Enterprise Monitor product comes with a number of default advisors. Based on your organization's particular needs, you may create or adapt custom advisors.

See Also custom, notification, schedule, standard.

A set of Advisors that are designed to enforce MySQL best practices for that specific category. For example, there are predefined advisor categories such as Administration, Availability, Cluster, Replication and Security.

See Also advisor.

The MySQL Enterprise Monitor component that actively collects data from a host and its MySQL server instances. The data is transmitted to the Service Manager.

See Also instance, Service Manager.

The web server that runs the Tomcat servlet container that powers the UI. The MySQL Enterprise Monitor Service Manager installation includes the Apache web server.

See Also Service Manager, Tomcat.

A component that the MySQL Enterprise Monitor product monitors, such as a MySQL server instance, or a CPU or file system within a server machine. Contrast with metric, which is a property within the component that is measured.

See Also instance, metric.

The Groups shown in the Asset Selector are the ones you create and manage (on the MySQL Instances Dashboard, or during Agent installation) and those that are discovered automatically from monitoring replication topologies. It is on the left side of many UI pages that controls content in the main-body of the page.

See Also asset, graph.

The mechanism that detects the set of master, slave, and master/slave servers in a replication configuration, and displays the topology in the Server tree.

See Also Asset Selector, master, replication, slave, topology.

The agent backlog is a caching mechanism which stores monitoring data in the event the agent cannot communicate with the MySQL Enterprise Service Manager. The backlog can store 10MB of monitored data in active RAM.

See Also Agent.

A function of Event Notification, it is a time period where events do not trigger notifications via the Event Handler, in that information is not reported from a MySQL Server. Typically, this is during a maintenance period when the database might go through an unusual workload that does not require raising any notifications.

See Also Agent, instance, notification.

An aspect or component of a system whose capacity imposes a limit on performance. In MySQL Enterprise Monitor, you identify bottlenecks in areas such as I/O or memory usage using graphs, and use advisors to automatically raise events when problems occur.

See Also advisor, event, graph.

The “bundled MySQL server” refers to the MySQL server delivered with the MySQL Enterprise Monitor product, to serve as the repository for the collected data.

See Also repository.

The C programming language.

A synonym for normalized query.

See Also normalized query.

In replication, a kind of topology where each server acts as a source/replica server, connected to at most two other servers.

See Also circular replication, master/slave, replication, topology.

In a replication configuration, a topology where every server is both a source and a replica, connected to two other servers, forming a ring structure. Forms a ring structure.

See Also replication, ring replication, topology.

The MySQL Enterprise Monitor product interfaces with many different databases and other kinds of servers. Each of these components can have its own login and security credentials. MySQL Enterprise Monitor pulls performance data from the MySQL servers that you monitor, stores the resulting data in a repository that is also a MySQL server, and sends alerts by communicating through other kinds of servers such as SMTP for e-mail alerts and NMS for SNMP traps. It pulls support-related data from the My Oracle Support site. You view the results in the UI, which is protected by its own login and optionally by LDAP authentication.

See Also LDAP, NMS, repository, SMTP, SNMP.

A high severity level for alarms. Within a rule, you can designate the threshold values that specify whether an alarm triggers a Critical, Warning, or notice alert.

See Also advisor, event, Notice, threshold, Warning.

User-created rules and graphs. Contrast with standard.

See Also graph, standard.

Data values derived from server status variables, operating system status information, and MySQL table information. You can reference these items using mnemonic names in expressions when you create or edit rules.

See Also expression.

The diagnostic report includes detailed information about your server (or multiple servers if you selected a server group), including configuration, hardware, MySQL options/variables and historical graphs.

If you intend to communicate with MySQL Enterprise Monitor support, it is strongly recommended you provide this report with your support request.

See Also My Oracle Support, service request, support issue.

The highest severity level for an advisor. Within an Advisor, you can designate the threshold values that specify whether an event triggers an Emergency, Critical, Warning, or notice event.

See Also advisor, Critical, event, Notice, Warning.

Events are displayed if an Advisor Threshold is crossed, and are used to inform you of errors or potential problems with your implementation. Monitor events using the Events page, and define how events are handled by configuring Event Handling.

See Also Critical, Events, Notice, notification, Warning.

Event handlers are conditions associated with notifications. If the condition is met, the notification is sent.

See Also Critical, event, Events, Notice, notification, Warning.

A page within the UI, for monitoring events. See Section 26.1, “Events”.

See Also event.

A textual report showing the internal mechanisms used by a query, and estimates for the “cost”, such as amount of data to process, involved in each step. Performance monitoring involves checking whether queries that are slow or frequently run could be sped up or made less resource-intensive, by examining the EXPLAIN plan to check if the MySQL optimizer has chosen the most efficient ways to process indexes, order join clauses, and so on. The MySQL Enterprise Monitor product includes a number of features for visualizing EXPLAIN plans.

See Also query.

The part of an Advisor that tests a metric against a threshold value.

See Also advisor, metric, threshold.

Additional considerations apply when the server you are monitoring is separated from the Service Manager by a firewall.

See Also remote, Service Manager, SSL.

An aspect of a SELECT query that often indicates a performance or scalability issue. The query scans every row in a table, rather than using an index to look up a subset of rows. It can be a non-issue for small tables that are cached in memory. It can be unavoidable when querying large tables to prepare reports. Performance issues are most likely when the table being scanned is involved in a join operation, when the query is run frequently, or when the result set only references a small fraction of the rows in the table.

To diagnose possible issues due to full table scans, choose a time period of heavy SQL activity from one of the graphs, use the Query Analyzer to locate queries that process large numbers of rows, and examine the explain plan for the queries. The notation in the explain plan that indicates a full table scan is Type=ALL.

See Also EXPLAIN plan, graph, query, Query Analyzer.

A visual representation of server activity, resources, or other metric over time. You view graphs using the Metrics view.

See Also metric.

An implementation detail from MySQL Enterprise Monitor 2.x that is no longer present. A heartbeat signal that is sent from an Agent to the Service Manager was defined as the heartbeat, and was present in the connecting URL.

See Also Agent, Service Manager.

A computer running one or more instances of MySQL server. Typically each host is monitored by a MySQL Enterprise Monitor Agent.

See Also instance, MySQL server.

A mysqld daemon running on a MySQL server. There might be multiple instances running on the same server machine. An instance is one of the asset MySQL Enterprise Monitor can monitor.

The Instances view manages the instances, and the MySQL Process Discovery advisor discovers local instances on the host that the Agent is installed on.

See Also asset.

A persisted instance of a run-time metric evaluation. These may store the raw metric data, or the result of an expression or function against a metric. Instruments are generally stored for things that are displayed in the GUI, such as graph data.

See Also graph, metric.

Core information about a host or MySQL server instance. This data is collected by the Agent. The inventory includes details such as the MySQL server version number, supported storage engines and replication configuration. The data in the inventory helps to determine what other kinds of data can be collected from the hosts and MySQL servers.

See Also Agent, instance, replication.

The MySQL Enterprise Monitor product is partly built on Java technology, using the Tomcat servlet container for the web-based GUI. It uses a JRE on the machine that runs the Service Manager, its performance is affected by the Tomcat configuration parameters, the Agent is written in Java, and during troubleshooting you might use diagnostic information from the Java environment.

See Also JDBC, Query Analyzer, Tomcat.

The underlying database API used by many Java applications.

See Also Java.

Oracle JavaScript Extension Toolkit (JET) is a collection of open source JavaScript libraries along with a set of Oracle contributed JavaScript libraries used to build applications that consume and interact with Oracle products and services.

See Also UI.

An authentication mechanism that can control access to the UI. On Linux, Unix, and macOS systems, you might have a separate LDAP server where each user has their own credentials. On Windows systems, the LDAP protocol is used to connect to an Active Directory server for the same authentication purposes. Typically, you map LDAP roles to corresponding MySQL Enterprise Monitor roles, to enable groups of users to have basic or administrative access to the Dashboard without configuring each user individually, or giving them a new user ID and password, or requiring extra work to revoke access when they leave the organization.

See Also credentials, role.

MySQL Enterprise Monitor manages log files for the Tomcat, repository, agentand Service Manager components, as well as a configuration report pertaining to the initial installation.

See Also repository, Service Manager, Tomcat.

A programming language that is used for parts of the Service Manager. Although you do not need to know this language to operate MySQL Enterprise Monitor, you may need to specify options containing file paths related to Lua scripts or libraries.

See Also Agent, Query Analyzer, Service Manager.

In a replication configuration, a database server that sends updates to a set of slave servers. It typically dedicates most of its resources to write operations, leaving user queries to the slaves. In complex topologies, a server can be both a master and a slave, known as a master/slave.

See Also master/slave, replica, replication, slave, source, topology.

In replication, a server that acts as a slave to receive updates from another server, and also acts as a master to propagate changes to another set of slave servers. Keeps the top-level master from having to service too many slaves, and allows certain kinds of changes to be propagated to a subset of slaves. This topology is represented as a tree, with all the intermediate nodes being master/slave servers.

See Also master, replication, slave, topology, tree.

Any property that is measured using a numeric value. Within MySQL Enterprise Monitor, such measurements can be displayed over time as a graph, or an alert can be sent when a threshold value is reached. Each metric is collected from an asset. For example, how much time is taken by a database query, or how full is the file system on a server machine.

See Also asset, graph, threshold.

In a replication configuration, a kind of topology consisting of different combinations of master, slave, and dual-purpose master/slave servers.

See Also chain replication, circular replication, master, master/slave, replication, slave, topology, tree.

To view information about the state, health, activity, and history of a resource such as a MySQL server instance. Monitoring can help to diagnose problems, spot worrisome trends before they turn into problems, reassure when systems are operating normally, and notify when an operator needs to take corrective action. With MySQL Enterprise Monitor, the component you interact with during monitoring is the UI.

See Also instance.

The website for filing bugs and service requests with Oracle Support. (Commercial customers such as MySQL Enterprise Monitor users now use the official Oracle support channel rather than the MySQL bug database.) In MySQL Enterprise Monitor, you interact with the support site through the What's New tab.

See Also service request, What's New tab.

The MySQL database instance running on a host computer. More than one MySQL instance can run on the same computer, either as separate mysqld processes or through virtualization.

See Also instance.

A programming framework used by many Windows applications.

Acronym for Network Management System, a type of monitoring system that is separate from the MySQL Enterprise Monitor product. If your network has this kind of monitoring capability, MySQL Enterprise Monitor can notify the NMS of network issues by translating events into SNMP traps.

See Also event, SNMP.

A condensed form of the query text used to treat similar queries as if they were identical, for monitoring performance. When MySQL Enterprise Monitor normalizes queries, it disregards differences in keyword capitalization, whitespace, and most comments. It replaces literal values with placeholders and transforms multi-row insert statements and IN clauses, to group similar statements with different parameters when measuring how much time is consumed by a particular type of query.

See Also canonical query, query.

A low severity level for events. Within an advisor, you can designate the threshold values that specify the severity level that is triggered.

See Also advisor, Critical, event, threshold, Warning.

The e-mail or other form of communication triggered by an alarm. Notifications are defined using Event Handlers.

See Also event.

A collection of users who receive a notification when advisor alerts occur. These users might have login credentials for the UI, but this is not a requirement.

See Also event, event handler, notification.

A scripting language typically embedded within web pages.

QRTi stands for "Query Response Time index". It is a "quality of service" measurement for each query, and it uses the Apdex formula for that calculation. For more information about Apdex, see: Apdex on Wikipedia.

See Also Query Analyzer.

An abbreviation for Query Analyzer used in some keyword names.

See Also Query Analyzer.

In the MySQL Enterprise Monitor context, any SQL statement whose performance might be monitored. Includes not just SQL SELECT queries, but also DML statements such as INSERT, UPDATE, and DELETE, and DDL statements such as CREATE TABLE and TRUNCATE TABLE.

See Also normalized query, slow query log.

The MySQL Enterprise Monitor component that tracks data about MySQL queries and summarizes that data using graphs and tables. You interact with it (for example, filtering the displayed queries or selecting a specific time period) using the Query Analyzer tab in the UI.

See Also graph, query, Query Analyzer view.

A page within the UI that displays output and controls the options for the Query Analyzer component. See Chapter 32, Query Analyzer View for more information.

See Also Query Analyzer.

Additional considerations apply when the server you are monitoring is separated from the Service Manager, for example by a wide-area network or by a firewall.

See Also firewall, remote monitoring, SSL.

An Agent on one host that monitors different hosts, and MySQL instances on different hosts.

Remote monitoring has limitations, versus local monitoring. Remote monitoring does not provide Operating System level data, such as CPU, file, and network utilization information.

See Also Agent, remote.

Synonym for slave in replication topologies.

In a replication configuration, a database server that receives updates from a source server. Typically used to service user queries, to minimize the query load on the master. In complex topologies, a server can be both a source and a replica, known as a source/replica.

See Also master, master/slave, replication, topology.

A set of database features that mirrors the same data across a set of servers. Used for reliability in case of server failure, and to speed up queries by dividing the work across servers. Because replication involves so many aspects of reliability and performance, it is an important aspect to monitor and has a dedicated Replication view in the UI.

See Also topology.

A view within the UI that monitors aspects of replication. See Chapter 28, Replication Dashboard.

See Also replication.

The database that stores the monitoring data collected by the MySQL Enterprise Monitor product. It can be a separate database instance that is part of the MySQL Enterprise Monitor installation (the bundled MySQL server), or you can use an existing database of your own. The bundled MySQL server is a level of MySQL database that is fully tested with the MySQL Enterprise Monitor product, and can be kept separate from your other databases to avoid any extra load on them. You might use an existing server that has spare capacity, fast storage devices, tuned configuration parameters, a backup system, or other conveniences that can benefit the MySQL Enterprise Monitor data storage as well.

See Also bundled, instance.

A level of access privilege for the UI. One of manager (highest privilege), dba, read-only, or agent (specialized privilege for sending data). Each user account registered with the Dashboard must have one of these roles. Components such as the Agent and the use the agent role. To simplify the process of granting credentials for large numbers of users, you can map LDAP roles to these MySQL Enterprise Monitor roles.

See Also Agent, credentials, LDAP.

Determines how often each advisor is evaluated.

See Also advisor.

A dynamic filter against an asset class.

See Also asset.

The core MySQL Enterprise Monitor component that receives the monitoring data from the Agent and Query Analyzer components. It displays this information through the GUI web-interface, and manages the Advisors, Events, and Event Handlers, as well as access to Reporting, Graphing, and Query Analysis.

See Also advisor, Agent, event, Query Analyzer.

Generic term for a bug that you have filed or a support request that you have opened, on the My Oracle Support site. Formerly, we referred to these as support issues.

See Also My Oracle Support, support issue.

A page in the UI for viewing or editing configuration settings of the Service Manager.

See Also Service Manager.

A measure of the seriousness of an Event. Event Handling policies can take into account the severity level in determining the appropriate course of Notification action.

See Also event handler.

In a replication configuration, a database server that receives updates from a master server. Typically used to service user queries, to minimize the query load on the master. In complex topologies, a server can be both a master and a slave, known as a master/slave.

See Also master, master/slave, replica, replication, source, topology.

A MySQL Server facility for tracking queries that consume considerable time and resources. MySQL Enterprise Monitor provides more information about query performance than the slow query log, and does not currently use the slow query log itself.

See Also query.

The e-mail protocol for sending alerts as e-mail messages. To send alerts this way, you configure e-mail settings similar to those in an e-mail client.

A protocol for sending event notifications (“SNMP traps”) to an NMS. The MySQL Enterprise Monitor product can turn notifications for selected rules into SNMP traps. In contrast to typical alerts that are only raised when some issue occurs, SNMP traps are broadcast for all state changes, so that corrective action can be cancelled when an issue is cleared.

See Also NMS, notification.

Synonym for master in replication topologies.

In a replication configuration, a database server that sends updates to a set of replica servers. It typically dedicates most of its resources to write operations, leaving user queries to the replicas. In complex topologies, a server can be both a source and a replica, known as a source/replica.

See Also master/slave, replica, replication, slave, topology.

In replication, a server that acts as a replica to receive updates from another server, and also acts as a source to propagate changes to another set of replica servers. Keeps the top-level source from having to service too many replicas, and allows certain kinds of changes to be propagated to a subset of replicas. This topology is represented as a tree, with all the intermediate nodes being source/replica servers.

See Also replica, replication, source, topology, tree.

For MySQL Enterprise Monitor, the default port used for SSL connections is 18443. If you are connecting using SSL, the built-in MySQL Enterprise Monitor certificate is self-signed and may be highlighted as “unsafe” within the browser on initial connection. To prevent problems accessing the site, add an exception for the certificate to your browser for this server.

The predefined advisors and graphs supplied as part of MySQL Enterprise Monitor. Contrast with custom.

See Also custom, graph.

The Status Summary widget is displayed on top of every page, updates dynamically, and shows current status counters for Hosts monitored, MySQL Instances monitored, MySQL Instances with invalid connection configurations, Unmonitored MySQL Instances, and Emergency Events. The counters are live links for displaying details or resolving the issues they are reporting.

Former term for bugs and service requests. In the My Oracle Support system, now known as a service request, which is the preferred term.

See Also service request.

Thresholds are the predefined limits for Advisors. If the monitored value breaches the defined threshold, an event is generated and displayed on the Events page for the asset. Advisor thresholds use a variety of different value types, depending on the monitored value. Some use percentages, such as percentage of maximum number of connections. Others use timed durations, such as the average statement execution time. It is also possible to check if specific configuration elements are present or correct.

See Also Critical, event, expression, Notice, Warning.

Apache Tomcat is a component of the Service Manager. It is the servlet engine that powers the UI.

See Also Apache, Java, Service Manager.

In a replication configuration, the way in which the different source, replica, and dual-purpose source/replica servers are connected. In MySQL Enterprise Monitor, the configurations are classified in the Replication view as one of TREE, RING, or MIXED.

See Also master, master/slave, replication, Replication view, slave.

A graphical representation of the replication topology displayed in a dedicated Topology view.

See Also replica, replication, Replication view, source.

A data structure often used to represent relationships between MySQL servers. In the UI, servers are displayed in the Server Tree sidebar. In a replication configuration, setting up some machines as dual-purpose master/slave servers produces a nesting relationship that is represented in the Server Tree.

See Also Asset Selector, master/slave, replication.

Predicting when a problem may occur. For example, predicting that a disk is expected to run out of disk space 30 days in the future.

The UI is a web-based interface to the MySQL Enterprise Service Manager. The back end is a Java application powered by the Tomcat server.

A unique identifier used to distinguish each MySQL instance, host machine, and agent. Because there is so much flexibility in spreading components across multiple systems or running multiple instances and agents on the same system, the combination of these different UUIDs identifies where information came from and the source of any issues. Always generate a new UUID for any one of these components, rather than copying or reusing an existing UUID value.

In MySQL 5.5, or earlier, the UUID for the server is stored in the table mysql.inventory. When a component such as the Agent connects to that MySQL server, the applicable MySQL user must have privileges to read this table.

See Also Agent, instance.

A medium severity level for alarms. Within a rule, you can designate the threshold values that specify for each severity level.

See Also advisor, Critical, event, Notice, notification, threshold.

A page within the UI that provides updates and news related to MySQL Enterprise Monitor and your My Oracle Support account.

See Also My Oracle Support.