We would like to know incident pattern in production and their type, is there any graph available to show incident pattern in specific period or weekly. Apart from that anythig which we can compare from last week incident patterns.
If you're looking for charts of Incidents being opened then I think the only option - and it isn't particularly elegant - is to use the Incidents Dashlet's History column which visualizes the opening and closing of Incidents over whatever timeframe the Dashboard/Dashlet is set to. You could set the timeframe to last 7d or "this week"; you can open multiple Incidents Dashlets on a Dashboard and set each to a different timeframe (one could be this week, the other could be last week, etc.) to get a visual comparison. You can also filter the Dashlet to specific Incidents and you can remove any columns you don't like. And then, you could save it as a Dashboard for future use.
Another option is to chart Threshold Violations. There is a Measure for Threshold Violations while there is not one for Incidents being opened. So, it's not exactly the same thing as charting Incidents as you are charting when the Threshold was violated, not when the Incident is opened.
The Business Transaction Hotspot Dashlet will let you see counts of Incidents and Violations for different time periods (this week, last week, etc), but it is only counts (not patterns) and only for your Business Transactions.
There is an "Incident Chart" which will visualize the status of an Incident (Open or Closed) as a Traffic Light (it shows both current status and worse case status for the timeframe), but it can't be changed to a Line Chart or the like to show the pattern.
So, I don't have a perfect solution for you but I hope these thoughts help.
As dynaTrace is APM targeting to production proactive monitor, in order for us to define any potential problem we must understand Trends, That's why I was expecting any charting option where we can visualize trends and reactify any potential problems.
We have various graphs and other measures in place to plot some pattern on CPU usage however not able to get anything for support team related to incidents where they can analyze pattern and development team can look for any optimization on that area.
I have a similar situation and I like the Threshold Violations measure. However, it looks like I need to create this measure for every single measure that I may want to get this count for. Is there a "global" option to set this up for all measures easily or do I really need to create a threshold violation count for ever single measure that I care about? Can I get this information by querying the database some how?
I wanted to add the following email conversation I had with Derek on a similar topic. This may help here as well:
Is there a way to graph how many times a threshold was violated, regardless if it generated an incident or not? Most of our incidents are set to alert after two consecutive threshold violations within the evaluation window of the incident. I want to see how many times a threshold may be violating off and on which means it wouldn’t generate an alert because it didn’t fail two consecutive times. I know you can just graph out whatever measure(s) that are defined as the condition of the incident, but we have hundreds of different measures we would need to do this for. Just curious if there was a better way or not.
For Business Transactions this is possible. Every result measure that has a threshold defined will automatically produce an additional result measure that tells you the number of violations. Would that help?