First of all thanx a lot for the new statuses "normal" and "warning" which are very helpful for us and propably many other admins, too!!!
In our first tries we realized that a test will have the status "normal", if both expressions (for warning and normal) are true.
Can I expect this behaviour for all tests?
Regards
Thomas
New statuses
-
- Posts: 166
- Joined: Sat Apr 15, 2006 2:14 pm
- Location: Germany, Bavaria
Doesn't it make more sense, to change that behaviour ?KS-Soft wrote:You are welcome![]()
CorrectIn our first tries we realized that a test will have the status "normal", if both expressions (for warning and normal) are true.
Can I expect this behaviour for all tests?
Regards
Alex
Warning is more serious and if a warning expression is true it should overrule the status to be warning instead of normal.
Not in case of escalation - what is my idea of the new statuses.KS-Soft wrote:Normally you should use different expressions and only one should return True
Regards
Alex
OK - Normal - Warning - Bad.
And of course you can build expressions that are unique - but that are more complex.
In most cases both expressions are TRUE and the more specific "Normal" overrules the "Warning". And the warning expression can be kept very simple.CPU test with 75 %
enable “Treat Warning status as Bad” option
enable “Use Warning" status” option and provide ("%SuggestedSimpleStatus%"=="DOWN") and (%FailureIteration% < 10)
enable “Use Normal status” option and use expression like ("%SuggestedSimpleStatus%"=="DOWN") and (((%FailureIteration% < 5) and ("%Reply%" <= "90 %")) or ((%FailureIteration% <= 2) and ("%Reply%" > "90 %")))
So it's OK to allow overlapping expressions and define a priority, from my pov.
I only think it's more logical to give the priority to the "Warning" expression.
But it's a change of program behaviour - so it has to be decided before the final release. That's why I'm talking about that.
But of course it's your decision.
And don't worry, I'll keep using HostMonitor in any case

Regards
Juergen