I have a bunch of graphs that have issues with outliers from bad data input, like this:

I need to cut the outliers (the spike) from the graph. Displayed is a case where the spike is only about 6x the normal activity, but when the spike is 500x the normal activity, the meaningful data cant be seen if viewing a timeframe which includes the spike; In a case like this the spike is the only thing visible if it is high enough.
I'm assuming I can go into the zabbix database and delete the spike datapoints, or update them to set them to zero or some more reasonable value. Which tables am I looking to do this in? I notice there are a handful of tables that have to do with graphs, don't want to screw anything up. Any pointers?
The data issue that caused the spike is a separate issue, just looking for a process to remove outliers.

I need to cut the outliers (the spike) from the graph. Displayed is a case where the spike is only about 6x the normal activity, but when the spike is 500x the normal activity, the meaningful data cant be seen if viewing a timeframe which includes the spike; In a case like this the spike is the only thing visible if it is high enough.
I'm assuming I can go into the zabbix database and delete the spike datapoints, or update them to set them to zero or some more reasonable value. Which tables am I looking to do this in? I notice there are a handful of tables that have to do with graphs, don't want to screw anything up. Any pointers?
The data issue that caused the spike is a separate issue, just looking for a process to remove outliers.
Comment