Hi guys! please help me some of my graphs are broken and im seeing these everytime i logged in. that i have devices unpolled.
i have devices that have these that the polling duration is 200 seconds. how can i fix this. or how to debug it. im using docker on it
I think this means that your polling is taking too long and not completing in the the 5-minute time slot. That is a lot of long device poll times. You should look at reducing that time.
The performance documentation might help you.
How can i reduce the poling time can you give me example?
Start here: Performance - LibreNMS Docs
I tried making 2 more pollers and i dont have broken graph and all are polling but my cpu spikes up
I tried making more 2 pollers, no broken graphs and no device unpolled but cpu spikes up
LibreNMS is CPU heavy, that doesn’t mean it’s an issue. Just see how you get on.
If you have CPU spikes that is also a sign of setting poller workers too high.
Hi i have a problem now
i setup a 2 new poller since monday from monday to wednesday this poller is great then comes today the devices are polling slow again, how can i make it the device to have like per poller group. i create 3 groups now
how can i assign each device to this group . I cannot saw a setting when i click the device to assign to a designated polling group
I suggest setting all pollers and devices to group 0. Pinning devices to a poller is bad practice.
I already did setting all devices to group 0 and still got unpolled device and causing broken graph
this is my validate.php file. i noticed that python wrapper pollers found is no. ive set my poller workers now to 32 for 3 dispatchers. should i just use 1 dispatcher or deploy 3 dispatcher in a single instance?
I had the same Problem. Have a look in the logs
/opt/librenms/logs/*
Had an error with the InfluxDB integration. I turned it off. no everything works fine again