IT services SLA calculation incorrect

[Michael0]

When I check the IT services for SLA calculation I find out that something is wrong with the calculation.
For example the SLA calculation for the "this month" filter:



As you can see the second from last value there was a problem: 0.1625%
But the overall SLA calculatin shows a problem from 4.6289% <-- where is zabbix getting this percentage value from?

another example is a correct SLA calculation with the filter "last7days"



It shows a problem from 0.3212% and the overall calculation shows also a problem with 0.3212% for the last 7 days which is a correct calculation

Thank you for infos!

[Heilig]

Use the API service.getsla method (https://www.zabbix.com/documentation...service/getsla). In addition to the SLA level, it shows the problems that have been for a specified period. So you will be able to check the correctness of the SLA calculation.

[Michael0]

Thank you for your tipp Heilig!

I´ve already installed the json rpc client but I got the following error message back, when I try to communicating with the zabbix api:

Can't locate JSON/RPC/Client.pm in @INC (@INC contains: /usr/local/lib64/perl5 /usr/local/share/perl5 /usr/lib64/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5 .) at ./auth.pl line 6.
BEGIN failed--compilation aborted at ./auth.pl line 6.


I´ve also found another problem with the SLA calculation...

This SLA calculation is correct in the past, 19h0min OK = 0% Problem = 100% SLA:



and currently SLA calculation:



It only shows 18h59min "OK" but 0h0min problem, so 1 min is missing every day and thats why I didn´t get the 100% sla...

[Michael0]

The API is now working, but it shows me the same result as in the GUI and I don´t kow why the SLA calcualtion is not correct

[Michael0]

ok finally I found the problem for the wrong calculation:

For some days (21st and 31st) it shows 100% problem, which is not true:



Would could be the problem in this case?

[Herwish]

Hi, we've got similar situation running Zabbix 2.4.2.
Anyone found a logical explenation of that? Is this a bug or we simply miss something in here?

[[email protected]]

This problem is happening yet...

Hi...

I am getting the same same problem. I am using Zabbix 2.4.5, so look on this print screen, it is possible see some strange things in this query:

In lines 108 and 109, I have the same status(4) in different times, but ServiceAlarmId 50519 report about a problem happened n 13:55:14, and ServiceAlarmId 50523 report about 13:55:10, a moment in time line before.

ServiceAlarmId Time Value
========= ================ =====
50519 2016-08-03 13:55:14-03 4
50523 2016-08-03 13:55:10-03 4

How is it possible the same Service(Integradores) get the same status in different time??

[[email protected]]

Other question about SLA problem

hi...

How is it possible a root show triggered if their leaves are OK??