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

SLA for synthetic monitor which consists of Private Synthetic Robot as well as Public Synthetic Robot

By 'Private Synthetic Robot' I mean the Synthetic Engine running inside Synthetic ActiveGate that is hosted inside customer's On-Premise, whereas 'Public Synthetic Robot' of course simply means Synthetic Location hosted by Dynatrace around the globe.


Say, if customer want to has the SLA of 99.5% for whatever website they want to monitor, but sometimes the website maybe can only be accessed from private and/or public, which means availability would be different.


In those case, just want to know the opinion/experience of how do you measure the availability then?

Get two SLA measurement (i.e one is 'internal-SLA' so-to-speak while another is 'external-SLA')? Combine into one? (and if that's the case, combine by average? min? max? or use some custom-formula to combine the two into one)


Thanks for your kind sharing in-advanced.


Best Regards,

Wai Keat



My general approach would be that if I have two monitoring locations available, I would only report the downtime if the issue is detected from both. That makes the results a bit more reliable. Of course there can be exceptions, for example if the app is only used by external users, then it doesn't really help if it was only available from the internal network (but down for everyone else).

How to handle the reporting in practise? Maybe use the availability MAX value, which means that it's 100 % as long as at least 1 location was successful. Another way which may work in some cases (if you just need to report downtime as individual breaks, with start time + end time) is to configure Problem detection to alert only when the monitor fails from all locations, and then you can report the Problem durations as downtime.

Featured Posts