I’m running the docker stack as provided in the example directory. All works fine, for the 7 items I do monitor in my home LAN.
I prefer to monitor with a 1 minute interval and that creates challenges. I think the main issue is that I’m unable to find the crontab which I need to edit in this dockerized environment.
Steps to reproduce:
-Changing the values from 300 to 60 seconds
bash-5.1$ php rrdstep.php -h all
Converting /data/rrd/apbeneden.high.local/ping-perf.rrd: [OK]
Converting /data/rrd/apboven.high.local/ping-perf.rrd: [OK]
Converting /data/rrd/apgarage.high.local/ping-perf.rrd: [OK]
Converting /data/rrd/esx1/ping-perf.rrd: [OK]
Converting /data/rrd/esx2/ping-perf.rrd: [OK]
Converting /data/rrd/nas/ping-perf.rrd: [OK]
docker-compose exec --user librenms librenms php validate.php
====================================
Component | Version
--------- | -------
LibreNMS | 21.11.0
DB Schema | 2021_25_01_0129_isis_adjacencies_nullable (224)
PHP | 7.4.25
Python | 3.9.5
MySQL | 10.3.29-MariaDB
RRDTool | 1.7.2
SNMP | NET-SNMP 5.9
====================================
[OK] Installed from the official Docker image; no Composer required
[OK] Database connection successful
[OK] Database schema correct
[WARN] Updates are managed through the official Docker image
If it’s an issue with the WebUI then please consider including a screenshot and the browser version you are using.
If you are having troubles with discovery/polling include the pastebin output of:
When changed al these parameters the output of the graphs showing no data, which is also confirmed by the orange banner in the corner that there are issue with polling.
One minute polling is always going to be a challenge because if you can’t guarantee your polling of all devices will finish in significantly less than a minute then the previous polling session won’t be finished before the next one is scheduled to start.
Depending on how fast your devices are, how many devices you have and how many threads your poller is set up to use it might not be possible to finish polling in under a minute.
Looking at our own system the total polling time is 103 seconds for 341 devices so too slow to set up one minute polling intervals (at least without increasing poller threads) and I suspect you’ll find the same.
With your polling put back to a normal 5 minute interval for half an hour or so go into Settings->Poller->Poller and check “Total Poll Time” - if it’s not well under 50 seconds forget about trying to enable one minute polling.
You can go to Poller->Log and sort by Polling duration to see how long each individual device takes, keeping in mind that multiple devices are polled in parallel depending on how you have the poller set up.
If the individual device polling times are all well under a minute you might be able to increase the number of poller threads in cron to get the total polling time down further, but keep in mind this increases the intensity of network traffic bursts generated by LibreNMS, and increases the memory/network/cpu demands on the LibreNMS server.
If you have individual devices that take more than a minute to poll (and we used to have some, but not any more) then there is nothing you can do to get polling down to one minute.
Not sure. Back when I first set up LibreNMS I tried to get one minute polling working and while I can’t remember what the issues were exactly I couldn’t get it working so I decided it wasn’t worth the hassle and just stuck with the default 5 minutes.
