Showing results for 
Show  only  | Search instead for 
Did you mean: 

Baseline alerting for low traffic transaction

We have several critical business transactions in place and I'd like to get alerted if it deviates from the baseline. Well, I thought that it would be easy to accomplish, but these critical transactions do not have a high throughput... Only a few requests per minute, but are considered important for the business

The thing is that it never reach the 'significant measure' status, so we never get alerted for these BTs!

I think that there is a lack of flexibility regarding baseline monitoring. I really believe that baseline monitoring is the way to go, but I cannot make it work as desired...

I don't want to waste time setting manual threshold, I think that this is not the best way to go. The system should have the capability to let me know when a situation becomes abnormal

Did I miss something? Any suggestion?


Dynatrace Pro
Dynatrace Pro

Baselines work great for high throughput transactions, but low throughput transactions will not be able to use the throughput baselines. The best thing to do is to create fixed thresholds. What is the reason that you don't want to do that?


Setting individual threshold per transaction is time consuming. Threshold can be too low or too high, who knows? It takes some time to know what the right threshold should be... I don't say that this is not feasible, but I think that in 2016, we should have the ability to leave the system learn what a normal behaviour is and get notified when it becomes abnormal

Dynatrace Leader
Dynatrace Leader

Unfortunately AppMon doesnt provide automatic baseline on throughput - at least not alerting. We do calculated an "expected" throughput funnel if you put the BT on the Monitoring Dashboard - but we are not alerting on it as we do for Response Time or failure rate. It is a missing capability that we are aware of and that we are trying to address in upcoming releases


Automatic baseline on throughput would be great, but having the possibility to generate response time or failure rate incidents for low traffic BTs would be great too. The concept of significant measurement is great, but doesn't allow enough flexibility to monitor critical transactions that do not have a high throughput. Hopefully there gonna be improvements for that too...

See here what a chart looks like.