If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to REGISTER before you can post. To start viewing messages, select the forum that you want to visit from the selection below.
If you want smart monitoring like Dynatrace then ZBX is not the right way. They are mostly commercial systems. You can get remotely close to it with correlations, predictions, but only after you collect enough data in cycles given by the measured system (day, week, month, year?) You will also need a perfect analytical tool for logs, I don't know if ELK is the right way. I would personally choose PostgreSQL and TimescaleDB addon for DB.
Moreover, if Prometheus and Grafana wouldn't be better for your monitoring.
Previously i was using Zabbix 7.0 and as a database, i was using Mariadb Galera cluster for HA and garbd for quorum,
my new requirement is i , shifting from Mariadb to Postgresql, the Zabbix version stays same.
Question 1 ??
Which is more feasible at this stage: migration from mariadb to postgre / starting from scratch again .
Question 2 ??
AS i was using galera and garbd while using Mariadb, what can i use while using Postgre so that it helps in my setup.
My deployment is done for zabbix 7.0 i am having 2 HA Servers where zabbix monitoring is going on. Now there is a use case : For ex : when a cpu utilization goes high i get a alert after the cpu goes high, Is there any way that i can know before a cpu goes high, i mean proactive alert that cpu might go high in 20 minutes and due to these reasons. Does zabbix has any feature as such and if it isnt , what are the other ways to do
A CPU is build to go high when needed, you might want to trigger on sustained high usage. Or average over time based, lets say 20m average compared to 20m average x time ago. And if you want to go even smarter also compare this to lets say day ago to filter out a delay running routine which causes a temporary higher load.
As these are completely installation specific they are not out of the box inside a template, you need to build them yourself.
For usage as diskspace etc you could also go the route of trend triggers when the storage is getting more occupied during it's age
Comment