I have had a couple of requests to do a how to guide on setting up the dashboard I showed in THIS post. So here it is.
First step is to look at the layout in my dashboard I use the following widgets:
- 1 x Metric Graph
- 3 x Heat Maps
- 1 x Health Status
- 1 x Health Tree
- 5 x Generic Scoreboards
1) Create a new tab and select edit, this will bring up the widget screen you see below, drag from the left and drop into the right the widgets I have listed above. Make sure you select 3 column layout, up in the top right you can see a little image of the layout here you can drag the columns to make the widths you need for each.
*Note* I will be doing the widgets that dont reguire any special configuration first, then move onto the ones that require groups or supermetrics.
2) Health Status Widget – click on the little gear icon on the widget to edit the settings, you will get a properties box similar to the below image,
- under the tags locate “Cluster Compute Resource” and select the cluster that you are setting this dashboard up for.
- Make sure you select self provider as on and if you want the widget to refresh select refresh as on and the time I use 300 seconds which is 5 minutes. since the collection interval of vCOPS is 5 minutes I used the same value here.
- Click OK
3) CPU Core Heat Map widget – click the little gear icon on the Heat Map Widget that you are using for CPU core utilisation. You will get a properties box similar to the image below. I have circled the part to pay attention to.
- first change group by to “Cluster Computer Resource” and Resource Kinds to “Host System.
- Under Attribute Kinds select “Core Utilization(%)” then under the tags Select the cluster you are setting up located under the “Cluster Computer Resource” tag heading.
- Under the colors is where you put in a value to a color I just used the default 0 – 100 as we are dealing with a metric we want to know the full range.
- if you want this widget to refresh select On for the Refresh widget content and choose your refresh interval in seconds
- Under the configuration heading in the top right you will need to enter a name and save it, here you could make multiple Configurations and a drop down is available on the widget to choose which one you want.
- Click OK
4) Memory Swap Heat Map Widget – Again click the gear icon to open up the settings box which will be similar to the image below.
- This is almost identical to the above configuration except under Attribute Kinds select the Swap in Rate
- The colors will also have to change, As any swapping is not a good thing I used the value of 5 as the max
- Click OK
5) Metric Graph – When clicking on the edit icon you will be prompted with something similar to the below image.
- Set the Self Provider as “ON” and set the refresh widget as “ON”
- Under the Resource Tags box on the right there is a “LIST” tab in here is a search bar. type in the name of you cluster and hit enter. Select the cluster in the results directly under the search bar.
- the clusters metrics will then be displayed on the right hand column, double click on a metric and it will be listed under the selected Metrics box on the bottom. Select the metrics you think will help your environment. or use the ones I chose to display.
- Click OK
6)General information Generic Scoreboard Widget – When clicking the the edit widget icon you will be prompted with a properties box similar to the one below.
- Set self provider as “ON”
- in the search field in the center enter the cluster name and hit enter, Select the cluster from the results listed directly under the search field.
- on the right double click the metrics you would like to display. I have a couple listed there but this can be anything you like I just use it for general info like vCPU to pCPU ratios etc.
- Remember to change the box label and the measurement unit, also the yellow/orange/red bound numbers. Basically the box will be green until it matches the yellow number entered, it will be yellow until it matches orange and it will be orange until it matches or exceeds the red number.
- Click OK
7) Datastore Threshold Heat Map Widget – Please see a previous post At a glance datastore usage by cluster or host for custom dashboards this is how this widget is setup.
8) Datastore Health Generic Scoreboard Widget – Again edit the setting of the widget and you will be displayed something similar to the image below.
- Set self provider as “ON” and enable the refresh content
- Like previous generic scoreboards above search for your cluster and select it.
- Select the metrics you want, the image shows the metrics I use, they are all standard and nothing special
- Remeber to change the Box label and Measurement and these ones will need the threshold you want to set. The ones in the image are what I use but your thresholds may differ
- Click Ok
9) CPU Health Generic Scoreboard Widget – This widget has 1 metric which is a super metric. That is the CPU Ready %, as some of you may know this is a host and VM metric not a cluster metric so to get this to work we have to work a little magic.
- Open up super metrics and create a new one (if you are unsure how to do this I have explained it before in “vCOPS Super Metrics making more readable dashboards”)
- The function we want to use is “avgN” of resource kind virtual machine as per the below image
- Make sure the super metric is part of a group that is added to at least the cluster your setting this up for. (this is show in the link provided in the fist dot point)
*Note* because a cluster is a grandparent to a virtual machine and a host the parent we need to go down 2 levels, this is what the avgN is bascially doing. it doesnt give you huge amount of info but it gives a general view of CPU health in regards to CPU ready times.
- Make sure at least 5 minutes has passed so that the new metric comes into effect
- Edit the widgets settings and a properties box will come up like the image below
- Like all previous generic scoreboard widgets search for your cluster and select it.
- Choose the appropriate metrics mine are listed in the image and you will notice the super metrics are under their own metric heading
- Configure the units and color numbers that is appropriate to your environment
10) Capacity Remaining Generic Scoreboard Widget – Hopefully by know your getting to know the generic scoreboard widget, open the settings and will be displayed with something similar to the image below.
- Enable self provider
- search for your cluster and select it
- under the time remaining metric select the ones listed in the image below
- make sure you set the thresholds for colors for these in line with your capacity management policies
11) Memory Health Generic Scoreboard Widget – This widget has a number of super metrics associated with it, They are simple math ones converting KB to GB I have covered this in a previous post “vCOPS Super Metrics making more readable dashboards” Instead of me going over everyone of the super metrics I will show some screenshots of my settings for a couple. They are all identical bar the metric im converting.
- once the super metrics are setup and have been assigned to the cluster(s) and given 5 minutes to generate open up the settings for this widget
- Like all the other generic scoreboard widgets set the self provider as on and search and select your cluster
- Select all the memory metrics you want to display, The list I used in mine are visible on the image below I used super metrics on everything that is measure in KB as it makes it more readable, for larger environments you could up this to TB