MySQL Enterprise Monitor is available as part of the MySQL Enterprise subscription, learn more at http://www.mysql.com/products/.
Once an advisor is enabled, it runs at set intervals. If it finds nothing of interest, it does not create events, which also means that notifications are not generated.
Events are defined using a number of different levels according to the severity of the alert. The eight levels are:
Failure: Indicates a failure for the given rule or event. Usually, this indicates that the rule on which the event is based has failed, probably due to a typo or other error. Occasionally, it can indicate a failure to retrieve the required information for a given rule.
Emergency: The event is an emergency and requires immediate attention. This is a panic condition, and means the system is unusable.
Critical: The event is critical and requires immediate attention. Critical events normally indicate that a serious problem has occurred or is about to occur. Examine the event, determine the cause, and fix the issue as soon as possible.
Warning: The event is something to be aware of, but does not affect the operation of your server.
Notice: The event is a notice for informational purposes. Notice events call attention to issues that do not affect the operation of your server, such as a minor configuration issue.
This was formerly known as the Informational level.
Success: The rule executed successfully with no issues. The result of the rule is OK. It also signals when an event that was previously in a Critical or Failure state returns to normal.
Unknown: The current status of the event/rule cannot be determined.
Closed: The issue has been corrected and marked closed.
For convenience, the event states are also represented by icons, as defined in the following table. The table also shows the relative level of the alert from the highest (Emergency) to the lowest (Unknown). The order represented here is used when identifying when an event has transitioned between levels (for example, from Success to Critical), hence triggering a notification, and when sorting events based on their current state.
|A red and orange flames icon indicates an emergency alert.|
|An round red icon indicates a critical alert.|
|A triangular yellow icon indicates a warning.|
|An blue octagon with the letter "i" indicates an informational alert.|
|A green check beside an event indicates that the Advisor has run successfully and no alert created.|
|A skull icon indicates that the status of the Advisor is unknown.|
When events are triggered, they appear on the Events screen. Emergency and Critical events also appear on the Overview screen. The notification group or groups associated with a specific advisor receive a notification when an alert is triggered. For more information about creating notification groups, see Chapter 18, Event Handling.
To view open events, click the Events tab. The tree-view on the left determines which server or server group these events belong to. Open events are shown in tabular format.
The event table has the following columns:
Current: An icon indicating the current severity of the alert.
Worst: An icon indicating the worst severity recorded for this alert.
Subject: The name of the asset the alert applies to, which is typically a MySQL instance.
Topic: A short description of the rule that is violated. Click the + on the left of the event entry to view additional information.
Time: The approximate time the event occurred.
Actions: Possible actions, where clicking
X will open the
By default, all events are shown but the list of events can be filtered using the form displayed above the event list. The options include filtering by:
Time Range - this time is filtered by the time the event was first seen
Choose the options and click the Filter button to refresh the display. To limit the number of items that appear on a page, choose a different value from the Show entries drop-down list box.
The drop-down list box showing severity status has the options:
(any) shows all events, and also
those advisors that have run successfully.
A 'successful' Event is one that currently evaluates past the level of the "Success" threshold, and is indicated by a green tick icon.
All Alerts shows only those advisors that
have been violated.
Some columns can be sorted by clicking the individual column headings. The alerts shown in Figure 17.1, “MySQL Enterprise Monitor User Interface: Events Screen with Search enabled”.
By default, the columns are sorted by Current and Time. If the default sorting is changed, the changes are stored in the browser's cache. If the browser's cache is cleared, the default sort is restored.
The server shown in Figure 17.1, “MySQL Enterprise Monitor User Interface: Events Screen with Search enabled”, is filtered
Any. Typically, when filtering by severity
All Alerts and, if you see an
Notice alert, use
All Alerts filter to see when the rule last
ran successfully. Narrowing down the time frame can help determine
the cause of the alert.
Besides filtering for severity, you can filter for a specific time
period using the
Time Range dropdown. You can
also filter by specific rules or categories of rules. The
State dropdown list box lets you choose
Closed events. To
avoid excessive scrolling, you can also limit the number of events
shown on a specific page.
For more information about an alert, click anywhere on the row with the rule name. A slider window appears, showing a description of the alert and the exact time of occurrence. This slider provides links to useful resources and advice for resolution.
For expression-based advisors, where possible, the results of the expression are listed in the Evaluated Expression section of the Details frame.