Hi, I’m facing issue with graphs from some Influxdb. During memory upgrade we rebooted server and fter reboot of we have huge peaks in graph as you can see in attached picture.
Data are sent to this meassurement every 10 second from collectd. I don’t see any errors related to this meassurement in collectd or influxdb. Only that there is as well higher disk utilization as well.
Graphs in collectd looks nice and are not affected at all.
SELECT 8* non_negative_derivative(mean("value"), 1s) FROM ..
Group by $interval always display data with huge peaks, Only static 1m show graph usable but doesn’t make sense for huge data ranges.
Any idea what’s causing this?