Not even sure I’m asking this right or in the right category but here goes.
I’ve got a multi sensor and for at least one sample of humidity the sensor read as 868657 instead of a reasonable percentage. Readings on either side of that one are in the mid 40s to low 50s (%). That one sample way out of range has caused the graph to get very out of whack.
Is there a way to delete that one sample or flush the entire list of humidity readings?
As far as I know I’m using rrd4j. I hadn’t set up any other persistence besides what happens default. I’ve poked around to add mariadb but not having much luck.
Tried fiddling around with the rrd file using rrdtool but that tool won’t work with those files apparently.
In this case you are probably not using the correct version of the tool. ATM I can’t say which is the correct one for a 4.0.x installation asI’m on the road…
The linked post of mine is rather old, you would the RRDTool correlating the actually used rrd4j verrsion.
Even better, the persistence config in OH 4 now has built in filters. You can handle this in the persistence config.
I think you can delete an entry or range of values through the REST API, but I don’t know what state that would leave rrd4j. Better would be to figure out the right rrdtool to remove the bad value. Set up your persistence to prevent bad values in the future. In a pinch, if you do want to remove them all, you can delete the file matching the Item name under $OH_USERDATA/persistence/rrd4j. You probably don’t want to do that while OH is running.