MarkLogic 9 Product Documentation
Ops Director Guide — Chapter 8

SUPPORT view

The SUPPORT view shows alerts for application servers in your enterprise, allows you to view and filter server logs, and displays tasks that run on your managed clusters.

This chapter covers the following topics:

System Alerts
Event Logs
Task Console

System Alerts

Use the SYSTEM ALERTS tab to review alerts from all monitored servers throughout the enterprise.

Alerts are associated with all types of resources. For example, if you have a resource group of databases containing db1 and db2, then only alerts associated with db1 and db2 and their forests are displayed.

The columns displayed for SYSTEM ALERTS are described in the following table.

Column	Description
Resource Name	The name of the resource related to the alert.
Severity	Color-coded alert severity level. For details, see Alert Levels.
Code	The alert code.
Start Time	Alert timespan starting date and time
Stop Time	Alert timespan ending date and time
Cluster	Name of the cluster that generated the alert
Alert Details	Detailed description of alert
Status	Alert status (Open/Closed/Acknowledged)
Action	Select Acknowledge to remove the alert from the list or Analyze to navigate to the ANALYZE view of the resource. For details, see the ANALYZE View.

You may export data from the System Alerts tab as a CSV (Comma Separated Values) file by clicking the Export icon in the upper right corner. The following rules apply:

The resulting CSV file will have the same columns as the System Alerts table in the UI.
All available columns are exported, regardless of whether they are visible or hidden in the UI.
All records that satisfy the set filter parameters are exported.
Data types of the fields in the CSV file correspond to those in the UI. In case of boolean values, Yes/No in the UI corresponds to TRUE/FALSE in the CSV file.

You may then import the CSV file into other applications (such as Excel) for further processing or analysis.

Once you have reviewed an alert, you can mark specific or all alerts as acknowledged to remove them from the list. Once acknowledged, the alert is suppressed from further notifications. The acknowledge action toggles depending on the state of an alert. If the alert is suppressed then the option changes to Unacknowledge, which unsuppresses the alert.

Select Acknowledge All to acknowledge all alerts, whether or not they are currently displayed. Acknowledge Selected is enabled if one or more alerts is selected via checkboxes. Select Acknowledge Selected to acknowledge those selected alerts.

Acknowledged alerts are indicated by an icon to the left of each acknowledged alert. The Acknowledged state is a property of the alert that is shared for all users. You can include acknowledged alerts in the table by selecting the Show Acknowledged checkbox. Uncheck to exclude acknowledged alerts.

You can narrow the results by specifying a time period, resource name, and alert severity.

Time Span -- You can select from the list of predefined time spans or select a custom date and time range by choosing a start date and time and end date and time, as described in Date and Time Filters.

Alert Level -- You can select one or more alert levels to narrow the list of alerts that have the specified levels.

Search System Alerts -- You can search for specific system alerts. The search text must be more than two characters.

The spaces in the string that you pass in the Search System Alerts box are trimmed off. For example, entering HEALTH SERVER with a space at the end will search for all alerts starting with HEALTH SERVER, such as HEALTH SERVER-DISABLED and so on.

Click Reset to reset the applied filters to their default values. The Reset button is enabled when you make changes to any of the currently applied filters.

Alert Levels

The color-coded alert levels are described in the following table.

Level	Description
Critical (red)	Resource is in a critical state - there is no hard definition of this alert as the resource can also be offline or non operational.
At Risk (yellow)	Resource that is dependent on another resource that is critical generates an At Risk alert. Specifically: A cluster is At Risk if any of its resources is in a Critical state A group is At Risk if a host is Critical or At Risk if an App Server is Critical or At Risk A host is At Risk if a forest is Critical or At Risk if an App Server is Critical or At Risk (note the App Server At Risk alert will show up on every host) An App Server is At Risk if the database is Critical or At Risk A database is At Risk if a forest is Critical or At Risk if a replica is Critical or At Risk if a foreign replica is Critical if a schema or trigger database is Critical or At Risk if a sub-database is Critical or At Risk There are other At Risk scenarios, such as: A forest has low storage space A foreign replica is unavailable 30% of hosts in a cluster are failing would place cluster At Risk
Healthy (green)	Resource has no cluster-health-reports associated it is considered healthy.
Maintenance (dark grey)	Resource status has been intentionally disabled.
Offline (light grey)	Resource state cannot be determined or resource is in an Offline state (for example, result of setting enabled property to false for the resource).
Information (dark green)	Optional information and advice.
Unknown (white / hollow)	State of resource cannot be determined or an Ops Director application error has been thrown.

Point-in-Time Alerts

Point-in-time alerts are derived from logs.

Certain log messages in the server logs make Ops Director generate alerts. Generation of the alerts is based on rules configured in the debug.xml file in the Ops Director database. Each rule specifies a log level and a regular expression to which a log message of that level has to match. This log message is translated into an alert of the severity specified in the rules.

Event Logs

Use the EVENT LOGS tab to review events from all monitored App Servers throughout the enterprise. Narrow the results by specifying a time period, log type, and log file.

Access to log data is based on your access to hosts and clusters. For example, if you have a resource group of hosts containing host1 and host2, only logs associated only with host1 and host2 are displayed.

The columns displayed for EVENT LOGS are described in the following table.

Column	Description
Date & Time	Datetime of the logged event.
Level	The log level of the event. For a description of the log levels, see Understanding the Log Levels in the Administrator's Guide.
Cluster Name	The name of the cluster on which the logged event occurred.
Host Name	The name of the host on which the logged event occurred.
Message	The logged error message.
Log File	The full text of the logged event.

You may export data from the EVENT LOGS tab as a CSV (Comma Separated Values) file by clicking the Export icon in the upper right corner. The following rules apply:

The resulting CSV file will have the same columns as the Event Logs table in the UI.
All available columns are exported, regardless of whether they are visible or hidden in the UI.
All records that satisfy the set filter parameters are exported.
Data types of the fields in the CSV file correspond to those in the UI. In case of boolean values, Yes/No in the UI corresponds to TRUE/FALSE in the CSV file.

You may then import the CSV file into other applications (such As Excel) for further processing or analysis.

By default, all event logs are displayed. Use the Log Level filter to select which types of event logs to view. For a description of the log levels, see Understanding the Log Levels in the Administrator's Guide.

Select any log file to access its full contents, which may be downloaded for further analysis, or for archival and audit purposes.

The text for the event log is displayed in a popup window. The selected log entry is boldfaced. The window includes entries up to 10 minutes before and 5 minutes after the selected entry.

Task Console

Use the TASK CONSOLE tab for an overview of all tasks throughout the enterprise. The data table presents the following information:

the task name
the cluster, host, and database on which it is running
the user who initiated the task
the task priority
the current state of the task

The columns displayed for TASK CONSOLE are described in the following table.

Column	Description
Cluster	Cluster on which the task host is located.
Host	The hostname of the host computer on which the scheduled module is to be invoked.
Enabled	Whether the task is enabled (true) or disabled (false).
Task Root	The root directory (filesystem) or URI root (database) that contains the module.
Task Path	The module the task is to invoke.
Task Type	The task type: minutely specifies how many minutes between each invocation of the module. hourly specifies how many hours and minutes between each invocation of the module. daily specifies how many days between each invocation of the module and the time of day (in 24:00 notation). weekly specifies how many weeks between each invocation of the module, check one or more days of the week, and the time of day (in 24:00 notation) for the task to start. monthly specifies how many months between each invocation of the module, select one day of the month (1-31), and the time of day (in 24:00 notation) for the task to start. one-time specifies the start date in MM/DD/YYYY notation (for example, 07/29/2009 for July 29, 2009) and time in 24:00 notation.
Task Period	How often the module is to be invoked (every n months, weeks, days, hours, or minutes).
Created on	The datetime the task was created.
Database	The database to which the scheduled module connects for query execution.
Task Modules	The name of the database in which the scheduled module locates the application code. If set to (file system), then any files in the specified task root directory are executable (given the proper permissions). If set to a database, then any documents in the database whose URI begins with the specified task root directory are executable.
User	The user with permission to invoke the module.
Priority	The priority of the task: normal specifies the task is queued with normal priority. higher specifies the task is queued with higher priority.

You may export data from the Task Console tab as a CSV (Comma Separated Values) file by clicking the Export icon in the upper right corner. The following rules apply:

The resulting CSV file will have the same columns as the Task Console table in the UI.
All available columns are exported, regardless of whether they are visible or hidden in the UI.
All records that satisfy the set filter parameters are exported.
Data types of the fields in the CSV file correspond to those in the UI. In case of boolean values, Yes/No in the UI corresponds to TRUE/FALSE in the CSV file.