Opsview Knowledge Center

Dashlets

Learn how to use Opsview Dashlets

Dashlets Overview

There are 19 different Dashlets available in Opsview Monitor (not including the seven extra Dashlets available to Users with Network Analyzer, which are covered within the Network Analyzer section).

These Dashlets are:

In the following sections, each Dashlet will be outlined and discussed, along with an example of a configured Dashlet and explanation of each of the configuration options.

Process Map

A Process Map is a Dashlet that allows you to upload an image to Opsview Monitor and overlay the monitored data onto the image. This data includes:

  • Hashtags
  • Host Groups
  • Host
  • Service
  • Performance Metric

You could upload an image of a rack, for example, and create a 'live status' map of the rack using the health of the monitored Hosts that are within the rack, as below:

In the above example, we have a real-life image of a rack that has been photographed (or generated via an IP camera) and uploaded to the process map. Atop of this image, various metrics have been added to reflect the health and performance of the objects within the rack, i.e. the two shelves of hard drives are in a problem state, and the performance of one of the servers is in a warning status, i.e. it may be overloaded. This allows you to not only see 'datashelf1 is down', but where that is within a given rack ' thus reducing the time-to-resolution, should it need physical intervention.

Another example is using a Hashtag per data center/office, and using a map of the country to show the health of each location based on objects within the given Hashtags:

Using a Process Map showing the map of the U.S., we can see immediately that there are problems within our Washington data center, again reducing the time-to-resolution and giving our Sysadmins a head start.

Adding and configuring a Process Map

To add a process map to your Dashboard, simply drag the Dashlet 'Process Map' onto the Dashboard as outlined in Creating a new Dashboard:

Once you have added the Process Map to the Dashboard, you will be presented with a view similar to the one below (this may be different depending on which Dashboard layout you chose):

To upload your own image, click on the Settings icon in the Dashlet header (the 'gear cog'), which will load a configuration modal window similar to the one below:

There are two methods of adding an image to a Process Map:

  • Upload Image: The simplest way, you simply select a saved .jpg or .png via the 'Browse' button and click on 'Upload' which will upload and display a preview within the 'Image Preview' window, as shown below:
  • URL Image: The 'URL Image' option allows you to enter the full path of a web-based URL, i.e. http://mysite.com/myimage.png. This image will then be downloaded on a regular User-defined interval, meaning if the image is constantly changing, then the Process Map image will reflect the changes. Once you have entered your URL, click on 'Preview' which will download the image and display it within the 'Image Preview' box, as shown below:
  • If you enter an invalid URL, or one that cannot be accessed, then you will see the error message 'Image cannot be previewed'.

Once you have uploaded an image via one of the outlined methods, you will have a process map similar to below:

You can now begin to overlay metrics onto the image. To do this, simply right-click on the image and you will be presented with a menu as below:

To add a Hashtag onto the Process Map, click on the 'Hashtag' row ' similarly click on 'Service' to add a Service Check, Host to add a Host, and so forth.

If adding a Hashtag, you will be presented with a configuration modal window similar to the one shown below:

Click on the 'Hashtag' text box and a drop-down list will appear, allowing you to select one Hashtag. After selecting the image size from Small, Medium and Large (the size of the icon), click 'Save' and the new Hashtag status icon will appear:

You can then drag and drop this icon anywhere within the process map, i.e. place it on the relevant item with the image. The addition of 'Host', 'Service' and 'Host Group' is very similar (select your object, select size of the image (icon) and click save).

The addition of 'Performance metric' indicators is slightly different. After clicking on 'Create > Performance Metric', a new modal window will appear similar to the one shown below:

This modal window allows you to choose a specific Service Check on a specific Host, and determine a custom label along with other options including arrow direction, arrow size ('Image Size') and more. The modal window also allows for the configuration of thresholds for the numbers specified, i.e. if you are displaying network throughput with this Performance Metric, you can say 'If traffic goes about X, change the arrow to yellow/ WARNING, if it goes about Z, change the arrow to red/CRITICAL'. This provides a visual indicator as to the performance metrics status.

In the window above, we are configuring the performance metric to display the time it takes to get a response in milliseconds (ms) from the Opsview server via HTTP. We want to set the threshold so that anything bigger than an 55ms response will set the performance metric to the WARNING status (yellow), and anything bigger than an 85ms response will set it to CRITICAL (red):

Clicking 'Save' will now add our new performance metric to the Process Map:

As with the other visual indicators, this performance metric can be dragged and dropped, and also hovered-over with the mouse to display the Service Check itself, as shown below:

Visual indicators can also be edited and removed from the Process Map. To do this, simply right-click on the 'performance metric', for example, and click on Edit/Delete as below:

On press of the 'Delete' option, a modal window will appear asking you to confirm the action:

Click on 'Yes' and the visual indicator will be removed from the Process Map.

Hashtag Tree Map

The Hashtag Tree Map displays a visual representation of the health of the Service Checks within one or more Hashtags within the Opsview Monitor systems:

The size of the cell is directly proportional to the number of Service Checks within the given Hashtag in a certain state, i.e. a large block of red indicates a large number of Service Checks in a CRITICAL state, as with the 'business-systems' Hashtag above.

The configuration of the Hashtag Tree Map Dashlet is very easy; simply click on the Settings icon in the Dashlet header which will load a configuration window. You can then choose which Hashtags you wish to display in this particular Dashlet, as shown below:

Once you have saved the settings, your Hashtag Tree Map will re-load with only the selected Hashtags:

Finally, you can hover the mouse over each Hashtag section to view how many Service Checks are in the given state, within the specific Hashtag, as shown below:

Hashtag List

The Hashtag List Dashlet provides a cut-down version of the summary view found within the Hashtags section of 'Monitoring':

The Dashlet will display the 'Description' field of the Hashtag, along with the status of the Hashtag, in a tabular format that can be sorted A-Z or Z-A. Sorting on the status column will sort the table based on severity, i.e. CRITICAL, WARNING, UNKNOWN or OK (or vice versa).

The configuration of the Hashtag Tree Map Dashlet is very easy; simply click on the Settings icon in the Dashlet header which will load a configuration window. You can then choose which Hashtags you wish to display in this particular Dashlet, as shown below:

Once saved, the Dashlet will update to display only the selected Hashtags:

Hashtag Cells

The Hashtag Cells Dashlet provides a cut-down version of the Grid View found within the Hashtags section of 'Monitoring':

The Dashlet will display the 'Description' field of the Hashtag in individual cells. The cell's color will indicate the status of the Hashtag, i.e. OK is green, CRITICAL is red and so forth.

The configuration of the Hashtag Tree Map Dashlet is very easy; simply click on the Settings icon in the Dashlet header which will load a configuration window. You can then choose which Hashtags you wish to display in this particular Dashlet, as shown below:

Once saved, the Dashlet will update to display only the selected Hashtags:

Performance Gauge

This Dashlet is used to display one or more Service Checks in a 'gauge' style, with the ability to show the WARNING and CRITICAL levels. Below is an example of a performance gauge used to measure available file system storage:

The configuration for this Dashlet is as follows:

In the Dashlet we only have one series, 'Opsview::disk: /::root' (Host::Service Check::Metric), however more than one series can be added. For example, we could add seven file systems and the number returned will be total of the seven metrics.

The minimum and maximum value are User set, i.e. the left and right of the performance gauge, as is the WARNING and CRITICAL threshold fields. You can choose to have thresholds defined when the value is greater than WARNING/CRITICAL, or conversely you can have thresholds defined when the value returned is lesser than WARNING/CRITICAL (i.e. space remaining, etc). If 'treat threshold as percentages' is ticked, the warning and critical values must be within 0 to 100%, otherwise they must be between the minimum and maximum values. If '>=' is chosen, then the critical value must be higher than the warning value. If '<=' is chosen, the warning value must be higher than the critical value. The 'Humanise Values' check box tells Opsview Monitor to try and 'humanise' the returned value, i.e. if the performance data for the Service Check's metric is '6001' with a UOM (unit of measurement) of 'Mb', then without 'humanize..' checked, the performance gauge will display:

With 'Humanise..' checked, Opsview Monitor will convert '6001Mb' to '6.0G', as below:

'Humanise values' also converts the minimum and maximum values, meaning '10000000000' becomes '10G' (Gigabytes). Unfortunately, even with 'Humanise..' checked, the values need to be entered in a byte format, i.e. 10000000000 for 10.0GB, 10000000 for 10MB, and so forth.

Performance Graph

This Dashlet allows Users to add a graph to their Dashlet, in a similar view and format to that within the Graph Center and within the 'Investigate Mode' in 'Host Groups, Hosts and Services':

The performance graph dashlet allows you to determine which series to display on the graph, where a series is a metric returned from a Host's Service Check, i.e. 'Opsview::disk: /::root' (Host::Service Check::Metric. The Dashlet also allows you to choose the type of graph (line/area), the thickness of the lines, and also the duration ' i.e. show the last 1 hour, 3 hours, 3 days, etc.

The configuration window for the performance graph Dashlet above is:

The series were added using the 'multi-add/bulk add' functionality. To use this functionality, simply select a field, such as 'Service' and choose your Service Check, i.e. 'Unix Load Average'. Then, choose your metric, i.e. 'load1'. After you have done this, the 'Host:' field will list only Hosts who have this Service Check/Metric combo.

You can then simply multi-select the Hosts, click 'Add' and the combination will be added, as below:

As can be seen below, there are three metrics, which when combined with one Service Check and ten Hosts results in 30 series that are added via the 'Add 30' button:

The options drawer contains three settings:

  • Duration: The range of data the graph will show: 1 hour, 3 hours, 6 hours, 12 hours, 1 day, 7 days, 14 days, 28 days, 1 month, 3 months, 6 months, 1 year, 2 years, 3 years.
  • Chart style: The display style, either Line, Area or Stack.
  • Line thickness: The thickness of the per-series lines, used for all three styles.

Below are some examples of graphs with different display styles:

Duration: 1 day, Chart style: Area, Line thickness: Thinnest

Duration: 1 day, Chart style: Area, Line thickness: Thinnest

Duration: 1 day, Chart style: Line, Line thickness: Medium

Duration: 1 day, Chart style: Line, Line thickness: Medium

The data on the Performance Graph can be viewed by clicking on the 'up' arrow, located at the bottom of the Performance Graph Dashlet:

When this arrow is clicked, the drawer will load. This drawer contains the series data including hostname and Service Check, as shown below:

The performance graph can also be 'focused'/minimized. To do this, simply drag your mouse across the area of the performance graph you wish to focus on/zoom into:

Once released, the Performance Graph will redraw to focus on the selected area:

To reset the Performance Graph, simply click on the 'Show All' icon, located in the top right-hand corner of the Dashlet.

Network Map

The Network Map Dashlet takes the dependencies between Hosts (as defined by you, the administrator), and draws a spider diagram outlining the relationship between Hosts, along with a status indicator for each Host, as shown below:

In the example above, we have focused the network map on the Host 'firewall1'. This Host is connected to a router, 'Cisco2611', which subsequently connects to a 'switch3' which has four Hosts connected to it. The Network Map allows you to see the dynamic, 'true' hierarchy of the Hosts within your environment.

The configuration window can be seen below:

The first drawer, 'Filter from Host', allows you to select a Host of which you wish to see the expanded network map (as shown above).

The second drawer, 'Filter by Host group', allows you to filter the list of Hosts available within the 'Filter from Host' by a given Host Group, and also limits the network map Dashlet to only show Hosts that are within the specified Host Group.

The final drawer, 'Advanced filtering', contains two options:

The first option, 'Only display Hosts that possess one or more children' essentially removes the 'status' from Hosts who have no further children, i.e. 'buttermere' has no children. With this option checked, the status (green/red/yellow) is hidden:

However, with this option unchecked, it displays the Host status:

The second option, 'Delete Saved positions', will reset the network map on each load if you have dragged them into different positions (by using their status icon, i.e. click and drag the green check, etc).

Metric Pie Chart

The metric pie chart allows you to create a pie chart view of the perfdata gathered within Opsview Monitor for Service Checks metrics:

Pie charts are great for comparing Hosts that are using the same Service Check and metric, i.e. comparing which Hosts are using the most network bandwidth, which Hosts are using the most disk space, etc.

The pie chart segments can be clicked on, which will load a contextual menu displaying the Host, Service Check, metric and value fields ' along with the percentage of the pie chart each slice has, and a small sparkline graph displaying historical performance:

There is a 'legend' icon in the top right-hand side, that allows you to view what each color is, i.e. 'Blue is 'warninghost'', etc.

There is also a data drop-down in the bottom left-hand side, which allows you to say 'What did this pie chart look like a week ago?'.

The data in the top right hand side will list the number of slices, and also the total data ' i.e. if each slice is 'Number of connections', then the number will be total number of connections across all slices.

Metric Pie Chart configuration allows you to choose whether you wish to have slices based on Host, Service Checks or Metrics:

In the example above, we selected 'Slices based on: Host', chosen the service as 'Unix Memory' and the metric as 'utilisation'. We can see there are 14 slices, i.e. 14 Hosts have that Service Check/Metric. As such, we have set 'Show top: 15' which will show all of the slices. If I only wanted to see the top five Hosts, in the context of memory utilization, I could choose 'Show top: 5' instead.

Host Group List

The Host Group List Dashlet provides a cut-down version of the view available within 'Monitoring > Host Groups, Hosts and Services':

This Dashlet provides the ability to view the 'Downtime, Host Handled, Host Unhandled, Service Handled and Service Handled, Total Hosts, Total Services' columns on a per Host Group basis.

By default, only three status columns are displayed ' Downtime, Host Unhandled and Service Unhandled, and all Host groups are shown. To add or remove columns, hold the mouse cursor over the column header row, click on the contextual menu icon that appears and then select or deselect column names within the 'Edit Columns' menu, as shown below:

In the example, we have added only the 'Services' columns:

The Host Group List can be sorted on the 'Host Group' and'Total Services' columns, as above, by clicking on the column name.

The configuration for the Host Group List Dashlet comprises of two sections:

Filter by Host group: Filter the Dashlet to only display selected Host groups.

Filter by State: Filter the Dashlet to display only Hosts or Service Checks in a certain state, i.e. only show 'non-OK/non-UP' items.

Once you have selected your desired filters, click 'Save' and the Dashlet will repopulate with the relevant Host groups and data.

Host Group Tree Map

The Host Group Tree Map is very similar in behavior to the Hashtag Tree Map, in that it displays a visual representation of the health of the Service Checks within one or more Host Groups within the Opsview Monitor systems:

The size of the cell is directly proportional to the number of Service Checks within the given Host Group in a certain state, i.e. a large block of red indicates a large number of Service Checks in a CRITICAL state, as with the 'Linux servers' Host Group above.

The configuration of the Host Group Tree Map Dashlet is very easy; simply click on the Settings icon in the Dashlet header which will load a configuration window. You can then choose which Host Groups you wish to display in this particular Dashlet, as shown below:

You can also choose to 'Filter by State', which allows you to tell the Host Group Tree Map to only display Hosts or Service Checks that are in a certain state, i.e. OK, CRITICAL, UP, etc:

Finally, you can choose to 'Only show unhandled Hosts and services', which will only display Hosts or Service Checks that are in an unhandled state, i.e. CRITICAL/WARNING/UNKNOWN or DOWN/UNREACHABLE and not acknowledged by an administrator.

In the example below, I have filtered the Dashlet to only show non-OK states (I.e. hide 'UP' and 'OK'), and also checked the 'Only show unhandled Hosts and services' box:

Event Table

The Event Table Dashlet allows you to view the events (changes of state) generated within Opsview Monitor in a Dashlet format. This table is a smaller version of the table within the 'Events Viewer' section of 'Monitoring':

The Dashlet will display, in a time sorted order (most recent at the top), the events that have been generated within Opsview Monitor. These events can be filtered in a range of different ways, as per the settings modal window for the Dashlet:

You can choose to filter the Events Table to show events that have occurred within specific Business Services, Components, Host Groups, Hashtags, Hosts, Host Templates or Service Checks ' and also filter the events based on their state, i.e. only show CRITICAL events, or only show 'HARD' events, etc. These filters are applied using an 'AND' logic, i.e. 'Only show events within 'Host Group A', on Hosts 'B' and 'C', only within the Host Template 'Unix Base', and only show CRITICAL events, that are in a HARD state'.

Finally, you can also choose to restrict the number of events shown in the table via the 'Filter by Limit' drawer, which allows you to choose between showing 25, 50, 100 or 200 most recent events.

Event Graph

The Event Graph Dashlet allows you to view the events (changes of state) generated within Opsview Monitor in a Dashlet format. This table is a smaller version of the bar chart within the 'Events Viewer' section of 'Monitoring':

The Dashlet will display, by default, data from the last six hours, for all events generated within the Opsview Monitor system. To see what the events are, simply click on a section, i.e. the 'Red bar' above with the '9' value in it, will load a new modal window, as shown below, displaying those 9 CRITICAL events:

Unlike in the main 'Event Viewer' section within 'Monitoring', only one bar can be selected at a time.

The Event Graph Dashlet has a range of filters, exactly the same as the Event Table Dashlet, which allows you to filter the events displayed on the Event Graph by a range of different ways, as per the settings modal window for the Dashlet:

You can choose to filter the Events Graph to show events that have occurred within specific Business Services, Components, Host Groups, Hashtags, Hosts, Host Templates or Service Checks ' and also filter the events based on their state, i.e. only show CRITICAL events, or only show 'HARD' events, etc. These filters are applied using an 'AND' logic, i.e. 'Only show events within 'Host Group A', on Hosts 'B' and 'C', only within the Host Template 'Unix Base', and only show CRITICAL events, that are in a HARD state'.

Finally, you can also choose to define a different timeframe from the default '6 hours', with options for '3 hours', '6 hours', '1 day', '2 days'.

Top Service Issues

The Top Service Issues Dashlet allows you to choose a Service Check, i.e. 'Unix Load Average', and view all of the Hosts who have that Service Check in a non-OK state (WARNING/CRITICAL), as shown below:

In the example above, the Top Service Issues Dashlet is displaying all Hosts who have the 'Unix Load Average' Service Check in a CRITICAL/WARNING state. The configuration for this Dashlet is:

In the above configuration window, you can choose the service, i.e. 'Unix Load Average' (just click in the field and a drop down window with all available Service Checks will appear). You can also choose to determine a WARNING/CRITICAL threshold, which will change the color of the row, and also choose between showing the 'Top 10' or 'Top 25' Service Checks in a non-OK state.

There are also options to further filter the list, i.e. only show Hosts that have 'Unix Load Average' in a non-OK state, within a certain Host Group ('Filter by Host group'), or only show Hosts that have 'Unix Load Average' in a non-OK state if that state is CRITICAL and HARD ('Filter by State'):

Top Services by Metric

The 'Top Services by Metric' Dashlet is different to the 'Top Service Issues' Dashlet, in that it allows you to choose not only a Service Check but also a metric of the Service Check, i.e. the metric 'load1' of the Service Check 'Unix Load Average', as shown below:

The other difference is that it lists all Hosts who have the specified Service Check & metric, instead of only ones who have the given Service Check in a non-OK state.

In the example screenshot above, we are displaying the top ten Hosts with 'Unix Load Average' and 'load1' applied, using the value of 'load1' as the sorting. The configuration for this Dashlet can be seen below:

In the example above, you can choose to define WARNING/CRITICAL threshold values ' i.e. of the 'load1' metric is above '3', change the row to yellow/WARNING, and if it is above 6, change the row to red/CRITICAL. You can also choose to define the number of issues to display, i.e. 'Top 10' or Top 25'.

There is also the option to further filter the list, i.e. only Hosts that have the 'Unix Load Average' / load1 metric, within a certain Host Group ('Filter by Host group'),

Note Pad

The 'Note Pad' Dashlet allows you to add text via a WYSIWYG editor, allowing you to add text, notes, hyperlinks, contact phone numbers and more to a Dashboard:

The configuration is very simple for the Note Pad, simply click on the 'Settings' icon with the header tab of the Dashlet and enter your desired text via the WYSIWYG editor, as shown below:

Simply click on 'Save' and the newly-entered text will be saved and displayed within the Dashlet.

System Status

The 'System Status' Dashlet will display the health of the Opsview Monitor system, as a cut-down version of the data stored within 'My System':

The Dashlet displays data regarding the performance of the Opsview Monitor system including average reload duration, last reload time, whether there are any uncommitted changes (i.e. a reload is required), and more.

There are two configuration options for the Dashlet:

'Enable configuration state indicator' will display, when checked, the 'Configuration status' row of the Dashlet, and the 'Enable passive Service Check metric' option will display, when checked, the 'Passive Service Checks in last hour' row of the Dashlet.

Dashlets

Learn how to use Opsview Dashlets