Hello.
I have noticed a strange behavior and I'm not able to see if it came from RPIMonitor or from RRD.
When I set an interval in a metric extraction with a rrd typ of DERIVE, it the graph indicates only spikes of data : the graph is at 0 when not in the interval, and at the proper value in the interval. If I understood well how interval works, it should not be the case.
According to the doc https://xavierberger.github.io/RPi-Monitor-docs/22_extraction.html setting an interval is supposed to extract a value every n loop of the daemon. It is not indicated what happens when we are not in the interval, but I suppose no data should be extracted.
To reproduce :
- I set
daemon.delay=60 in daemon.conf to get data every minute
- in
network.conf, I set dynamic.10.interval=10 so there is an interval only for the 1st metric (net_received = download), which is a RRD type of DERIVE.
So it should monitor the download bandwith every 10 minutes. The upload/send bandwidth is still monitored at 1 minute.
See the graph :
I activated the interval at 11:00. You can see the green area (download bandwidth) change from a smooth average around -500 to spikes with value at 0 outside the interval.
Zoom between two green spikes :
We can see the grren line at 0 and some spikes.
It looks like data are acquired outside the interval and are set to 0. Is there a way to change that ?
It did not appends with other type, only with DERIVE . But bandwidth (in bytes/s) can only be extracted by derivating the absolute value of /sys/class/net/eth0/statistics/rx_bytes .
Hello.
I have noticed a strange behavior and I'm not able to see if it came from RPIMonitor or from RRD.
When I set an interval in a metric extraction with a rrd typ of DERIVE, it the graph indicates only spikes of data : the graph is at 0 when not in the interval, and at the proper value in the interval. If I understood well how interval works, it should not be the case.
According to the doc https://xavierberger.github.io/RPi-Monitor-docs/22_extraction.html setting an interval is supposed to extract a value every n loop of the daemon. It is not indicated what happens when we are not in the interval, but I suppose no data should be extracted.
To reproduce :
daemon.delay=60indaemon.confto get data every minutenetwork.conf, I setdynamic.10.interval=10so there is an interval only for the 1st metric (net_received = download), which is a RRD type of DERIVE.So it should monitor the download bandwith every 10 minutes. The upload/send bandwidth is still monitored at 1 minute.
See the graph :
I activated the
intervalat 11:00. You can see the green area (download bandwidth) change from a smooth average around -500 to spikes with value at 0 outside the interval.Zoom between two green spikes :
We can see the grren line at 0 and some spikes.
It looks like data are acquired outside the interval and are set to 0. Is there a way to change that ?
It did not appends with other type, only with DERIVE . But bandwidth (in bytes/s) can only be extracted by derivating the absolute value of /sys/class/net/eth0/statistics/rx_bytes .