Hello everyone,
I have a weird behavior that I am encountering and i hope to find some answers with you here.
The way I am making use of the Zabbix mechanisms is a bit peculiar, but please bear with me as i try to explain it to my best effort.
The reason why i decided to take such an approach is because I am already monitoring a lot of other things for my organization using Zabbix and it resulted as a collaboration with some of our data engineers.
They are generating some basic JSONs with 5-6 key-value pairs which contain various reporting/alerting valuable information for our organization. I am generating the JSON with one external script and creating dependent items based on the JSON Path processing to extract each other value. Most of the fields contain general text information that are not used for logical expression evaluation, except one that is called generateAlert which may be True or False , depending on the channel i want to send the message to and in which format. All of the dependent items are part of my evaluatiion expression so that i can use the ITEM.LASTVALUE# functionality in order to create a predefined message using all of the collected info. I also added a script in the trigger action to automatically close the incident generated for the host id.
The evaluation expression looks something like:
1) last(/blablab/bla#1)="true" and length(last/blablab/abc,#1))>0 and length(last/blablab/abcd,#1))>0 ..... and so on
2) last(/blablab/bla#1)="false" and length(last/blablab/abc,#1))>0 and length(last/blablab/abcd,#1))>0 .....and so on
Although in most of the cases implemented after this model everything works as expected, for two of these hosts i sometimes get both of the triggers issuing at the same time. This happens only when i have a true evaluation. What happens is that for the evaluation and the next first time it is false, both triggers are getting shot. This never happens when it is only false.
I have tried querying GPT about it and somehow it told me that it could be related to the alphabetical order of the trigger evalation(did not find that info in the documentation) but it did not solve my issue.
I also tried implementing the Depends on mechanism in both ways( a depends on b, b depends on a) , still did not resolve.
Could you please let me know what could be causing this dual triggering of my abstract behavior. Any info is welcome.
Thank you!