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

Fixed transactions threshold

ckc
Organizer

Hi,

Anyway I can set fixed thresholds on transactions when setting up applications, or even better on operations. On our Application Health Status we have now set it to Operation time as primary metric, and would like to use defined thresholds not trends over xx:xx time ranges.

Regards,

Claus

4 REPLIES 4

Hi Claus

Static thresholds to determine if an operation is considered fast or slow can be configured in the software services rules. This however would not affect the AHS if you choose operation time as the primary metric.

If you just want to change the threshold on the AHS simply refine the section, select the threshold tab and click on custom

Regards

Alasdair

ckc
Organizer

My problem with that is it is a general threshold for all operations, I would like to define thresholds pr operations. Our operations related to our main portal is from 20ms to 6sec.

Where do you want to show this Claus?

The top AHS screen is the total of all underlying applications?

The values from the various operations rolls up to the level above so if your portal only has two operations, one 20ms and one 6 sec, then on the applicationlevel would be the average of these two. The response time as a numeric value don't serve much use here so instead you should be looking for the fast/slow/failed operations as they would indicate the underlying 2 operations statuses.

ckc
Organizer

I would like this to be shown on AHS.

Our portal uses 28 operations on login, a login on average takes 9 seconds (I know its horrible)

I do think that if we define operations with fixed values it would make a lot of sense for us since fast/slow operations are based on a global threshold or on SS, and our operations time goes from 20ms to 6 seconds then global threshold or SS threshold doesnt seem to do it.

A average based on last 24 hours trend would be a problem as well since the fast operations could run horrible and still be within average since avererage when I set it up now is around 4 seconds on most transactions.

Thinking im off course regarding the use of the applications and should make the AHS for operations with average trends baselines so general error, tiers going bananas are caught at a glance, and if we want to be alerted if specific operations time go up after releases of some banana started indexing on a prod database we set this up in Dynatrace.