9 April 2015 blogs Guest Blogger 8 min read
Savision has made a huge addition to the System Center Operations Manager community with the launch of their new “Dashboards by Savision” solution. Dashboards by Savision is an HTML5 based dashboard solution which reads directly from the Operations Manager data warehouse. Additionally, their solution contains pre-built dashboards for a series of different products which are monitored with Operations Manager. The result is a simple to install solution which provides pre-built dashboard solutions that can be used on any device.
Daniel Örneling has provided a blog post which discusses how to install Dashboards by Savision, an overview of what it provides and details on how to build your own custom dashboards. Check out his blog post ”Savision Dashboards”
The focus of my blog post will focus on the following topics:
- What pre-built dashboards are available in Dashboards by Savision?
- An exploration of each of the pre-built dashboards which are added without any customization to an environment
- A discussion of two custom dashboards included within my environment
- One general navigation tip and a summary of this blog post
What pre-built dashboards are available in Dashboards by Savision?
The screenshot below shows the variety of pre-built dashboards included in this solution including:
Additionally, I have two custom dashboards which were created for my environment which I will also discuss in this blog post:
The dashboards which appear as non-greyed are available in your environment, the greyed out ones if you highlight and click on them you can see why they are not enabled such as the example below for my SharePoint dashboard:
Opening any of the pre-built dashboards is as simple as clicking on the icon. As a next step we will go through and show examples of out of the box dashboards and their functionality in my environment with no customization or configuration other than installing the solution and adding a license.
The Active Directory dashboard contains a variety of information including:
- A heatmap for DC health
- Top DCs by active alerts
- Top DCs by bind times
- Top DCs by authentications
- Top DCS by CPU
- Top DCs by memory usage
- Top DCs by replication traffic
- Top DCs by TCP connections
As with many of the other dashboards you can change the duration of how much data you will display by clicking on the duration (One week by default).
This updates all of the counters shown on the dashboard (the example below shows one year of data). Since Dashboards is retrieving data from the data warehouse you can provide dashboards which showcase long periods of time even though the solution may have been just installed. The health state is determined by the health of the Active Directory object so each of the domain controllers in my environment is currently healthy from this perspective.
You can drill into the individual servers shown above and get further detail as shown below for my Cloud-DC1 system (note the “Related Services” section which shows that this server is part of the Active Directory Topology Root and a distributed application – LOB Application).
The Computer Health dashboard shows the health of the Windows Computer objects in Operations Manager in an intuitive list ordered by their health state. The example below shows that the order if Critical servers (Red), Offline servers (Grey), Warning servers (Yellow) and Health servers (Green). The ordering is important as it’s easy to identify systems which need investigation at a glance from the perspective of the Windows computer.
You can drill into the various objects and see what is occurring within them. The health state is determined by the health of the Windows Computer object so we can see the active alerts which may be impacting this server’s health from a Windows Computer perspective.
The Exchange dashboard contains a variety of information including:
- Exchange servers with most active alerts
- Exchange organization active alerts
- Top Exchange servers by CPU usage
- Top Exchange servers by memory usage
- Top Exchange servers by client connections
- Top Exchange servers by Disk I/o
- Top Exchange servers by message traffic
From this level we can dig into the Exchange server (EXCH in the example above) to see further details about the server from an Exchange perspective as shown below (including the fact that in the Exchange portion of the lab I’m demonstrating from isn’t really working well and isn’t being utilized now based upon the metrics shown below). Note the “Related Services” section again below as this makes it easy to see what services this particular object is associated with.
The Internet Information Services dashboard contains a variety of information including:
- A heatmap for IIS health
- Top IIS servers by active alerts
- Top IIS Servers by web service connections
- Top IIS servers by CPU
- Top IIS servers by memory usage
- Top IIS servers by web service traffic
- Top IIS servers by web service requests
You can also drill in to the IIS server and see details related to IIS performance, connections, OS performance, server details and any related services.
The SQL Server dashboard contains a variety of information including:
- A heatmap for SQL server health
- Top SQL servers by active alerts
- Top SQL servers by transactions
- Top SQL servers by average wait time
- Top SQL servers by CPU
- Top SQL servers by memory
- Top SQL Servers by user connections
The SQL dashboard makes it easy to see what servers are experiencing performance issues such as my DB03 server shown above. Opening this server in the dashboard gives us more details as shown below.
And from the first dashboard we can drill into more details as shown below.
There are two non-standard dashboards that I wanted to include as Savision has done some very interesting things with these and they have helped to resolve a client issue that I’ve been running into. These dashboards provide an easy way to visualize specific objects as well as the health state of specific monitors. This is a huge area of interest for me personally and has been since 2011. In 2011, I blogged on a question about why we can’t add a monitor to a distributed application. The reason is that while a monitor has a health state, it is not an object, and since it’s not an object it can’t be added to a distributed application or to a dashboard – until now.
Basic Computer Health & Monitors
In the dashboard shown below, Savision has provided a method to show the health of specific monitors and is rolling up these monitors to the health of the server column shown on the left. As an example, the right column shows the Operations Manager monitor for memory health, the CPU column maps to the CPU health. If either of these monitors go to a non-healthy state they impact the performance column as well. If any of the columns are unhealthy they roll up to the server column on the left.
The details on these columns are provided below by Savision:
- Server: Windows Operating System (the name from the related Computer object).
- Ping: Computer / Availability / Ping Status
- Disk Space: Computer / Hardware Availability Rollup / worst of Windows 2012 Logical Disk Free Space Monitor, Windows Server 2012 Logical Disk – Free Space Rollup Monitor, Logical Disk Free Space, Windows Server 2012 Logical Disk Free Space (%) Low across all Logical Disk (abstract)
- CPU: Computer / Performance / Operating System Performance Rollup / Total CPU Utilization Percentage
- Event Log: Computer / Availability / Operating System Availability Rollup / Availability / Core Windows Services Rollup / Windows Event Log Service Health
- Services: Computer / Availability / Operating System Availability Rollup/ Availability / Core Windows Services Rollup
- Performance: Computer / Performance / Operating System Performance Rollup
- Memory: Computer / Performance / Operating System Performance Rollup / Perfohrmance / Available Megabytes of Memor
How many times has it been an issue where your Windows Computers appear as unhealthy due to a monitor which doesn’t really impact the functionality of that server? As an example, if a disk is highly fragmented, that will change the server’s health to warning. Using this type of a dashboard we can quickly visualize exactly what we want to visualize using either objects or monitors to drive the health state shown in the dashboard.
Basic Computer Health
The Basic Computer Health dashboard shows only the health of the computers displayed from the perspective of the Basic Computer Health & Monitors dashboard. This provides a top level NOC view of server health which can then in turn be drilled into to determine what underlying object or monitor caused the health state displayed on the dashboard to occur.
|General navigation tip: If you want to get back to the home page, use the icon in the top left corner of the dashboard:|
Summary: If you are an existing customer for Operations Manager you should definitely check out the new Dashboards by Savision solution (Hey why not: It’s free, It’s fast & It’s HTML5!)
About: Cameron Fuller
Cameron Fuller is an author, speaker, Microsoft MVP and principal consultant for Catapult Systems, a globallyfocused Microsoft IT consulting firm that provides application development, enterprise solutions and infrastructure services. He’s a System Center Cloud and Datacenter MVP and co-author of System Center 2012 Operations Manager Unleashed and System Center Operations Manager 2007 R2 Unleashed (Sams).