All questions related to installations, configurations and maintenance of Advanced Host Monitor (including additional tools such as RMA for Windows, RMA Manager, Web Servie, RCC).
Hi
I setted up some ping tests in HM.
Lately, I got a mail from HM triggered by an alarm that says that a certain server went up. That´s not my problem but it says that the server has been down for 42 days. That is not true. In the log analyzer the server was down since 6 days. How can that be?
Furthermore some entries in the log analyzer say that the alive time of some servers have the value e.g. -22,11% or 106,79%.
Did I anything wrong or is it a known bug? Or any other Explanation?
In my case it is absolutely necessary that I can rely on the results that the HM provides, so please help me!
Could you please provide some more information:
1) what version of HostMonitor and Log Analyzer do you use?
2) what kind of log file do you use (html, dbf, text)?
3) does HostMonitor display alive%, dead% counter correctly?
4) have you changed system date/time format?
Concerning the information in my alarm mails that some servers have been up/down in meantime for more than 50 days, I want to add that I use the macro variable %previousstatusduration% in the mail body.
Thank you for the file but I cannot reproduce this problem. All % counters looks correctly, their values between 0 and 100.
May be you analyzed several log files?
Concerning the information in my alarm mails that some servers have been up/down in meantime for more than 50 days, I want to add that I use the macro variable %previousstatusduration% in the mail body.
I checked how counters work... the only explanation I have - system time was changed (e.g. someone set date to January 2004, and after a minute changed time back to normal). It that possible?
there is only one log file in HM folder.
But I found some entries in the system event log. Several strange W32 Time messages, containing information about failed time synchronisations and something like: "system time is 3456342 seconds"
So it could have something to do with changed system time...
I am going to send you a screenshot of my log analyzer, so you will see the negative values.
it's not 2003 anymore and the actual version we use is 9.18, log is html, .
Still (again?) the same issue: Negative alive times.
What might be a reason at our installation: Very high test counts (>18.000.000).
Out of ~50 Tests over here only (but all of them!) the tests with very high counts show negative values.
All Tests with negative alive% are (alive% + dead% + unknown% <100), even if you do not count the negative sign of the alive% as negative.
As you asked Rapthor to send the logfile: This would be of no help as deleting the log file does not change anything. Is that counter in the HML file or any other?
Regards
P.S.
negative Alive% with smallest 'Total Tests' number is 22495312 total tests. All other negative Alive% have higher number of total tests.
highest 'Total Tests' with positive alive% is 18047963, all others are lower.
What application shows negative values? HostMonitor or Log Analyzer?
>What might be a reason at our installation: Very high test counts(>18.000.000).
I assume you are talking about TotalTests counter?
What option do you use for HostMonitor?
- Display Alive/Dead ratio of passed/failed tests
or
- Display Alive/Dead ratio of alive/dead time
(option located on Misc->Reports&Statistics page in HostMonitor Options dialog)
>What application shows negative values? HostMonitor or Log Analyzer?
It's HM. We don't use LA (much) as it takes too long to load due to the huge size of our logs.
HM shows the same negative value list view as in Test info whilst test history shows positive values which (roughly estimated) will show the correct value.
>I assume you are talking about TotalTests counter?
yes, see P.S. of OP
>What option do you use for HostMonitor?
we use Display Alive/Dead ratio of passed/failed tests,
display is correct (again, roughly estimated) when switching to option Display Alive/Dead ratio of alive/dead time (BTW: thanks for pointing to the right menu )
I think we found mistake. Will be fixed in new version (begining of July).
Then both options will show correct value (you will not need to reset stats).
