-
Incident report
-
Resolution: Incomplete
-
Trivial
-
None
-
3.2.5
-
None
-
Linux
Recreation of the issue:
Create item I1 numeric unsigned updated in 60 seconds
Create trigger T3 severity average I1.last() > 3
Create trigger T2 severity warning I1.last() > 2, dependent on T3
Create trigger T1 severity information I1.last() > 1, dependent on T2
Item I1 goes from 0 (OK) to 4 - only trigger T3 fires, that's fine
Item I1 goes from 4 back to 0 - trigger T3 returns to OK, and that should be the end of the chain, but instead trigger T2 fires (actually continues to be in PROBLEM state)
T2 becomes visible in dashboard, actions are taken on it (notifications for PROBLEM state sent), in 60 seconds it goes to OK state and T1 fires.
In 60 seconds T1 goes to OK as well.
So instead of supposed 2 notifications (problem of severity level average start - problem recovery) I get 6, two for each severity level.
I tried to change item update interval from 60 to 10 seconds, and triggers T2 and T1 fired and recovered in 10 seconds as well.