Version: 17.07
In the current version IMonitor-X provides two built in widget types. Out of those the space health widget is the one dedicated to providing real time node health statistics.
In order to define a space health widget first navigate to the dashboard view by selecting the Home icon from the main menu. Then click on the Manage Dashboard Widgets (1) button, followed by Define New Widget (2) button.
This will bring up the basic Add New Dashboard Widget form as follows.
On the name field (1) specify a name for the widget. Any sensible name to identify the widget would suffice.
Then in the Widget Type dropdown (2), select Space Health to add a space health type widget. Note that once you select the widget type, the Widget Configurations sections gets loaded. The widget configurations shows the configurations specific to a particular widget
This checkbox (3) defines whether the widget is publicly visible. If a widget is set to public, all the users can view this widget and pin it to their dashboards. But only the user who created the widget can edit or delete it. Note that you can make widgets public/private in Dashboard Management view.
This checkbox (4) defines whether the widget will be pinned to dashboard once it is created. Note that you can pin/unpin widgets in Dashboard Management view.
Cluster Space (1) |
Select a cluster space for which you would like to see health statistics. |
Max File Descriptor Count (2) |
In Unix and related computer operating systems, a file descriptor (FD) is an abstract
indicator (handle) used to access a file or other input/output resource, such as a pipe or network socket. In the space
health widget, we show the current file descriptor count as a percentage and for that we require the max file descriptor
limit configured in the system. This value can be queries using the command |
Max Thread Count (3) |
Enter a value for max thread count the system can tolerate. |
Upon configuring all the required fields, click Save to save the space health widget.
A sample space health widget would be as follows.
The cluster space dropdown (1) can be used to view space health statistics of spaces other than the configured space. Note that this is just a view and upon refreshing the widget view goes back to show the configured space. (2) shows the number of instances in the space and there will be accordion panels for each node in space. In the title of each accordion panel the node name (3) and the service url (4) will be shown.
Then there is the progress bar (5) which shows the overage health. The progress bar will be green in Normal state, orange in Warn state and red in Critical state. In the four circular pie-charts, percentage values of memory usage (6), cpu usage (7), open file descriptor count (8) and thread count (9) is shown.
In case where any of the above for percentage is above 90%, the state will be marked as Critical and similarly in a case where any of the above percentage is above 60%, the state will be marked as Warn and other cases will show the state as Normal.