I am running TrueNAS ElectricEel-24.10.2.2 on a FREENAS-MINI-XL and have installed an APC Back-UPS RS 1500MS2 and it is connected to a Linux system running the NUT service and I can query it just fine using the upsc command from the NAS. In the TrueNAS UI I have configured the UPS as a slave and put in the UPS name, IP address and port of the NUT service along with the NUT monitoring user and password. I am using a user defined as upsmon ability.
Saving a packet capture I am seeing requests from the TrueNAS to the Linux system running NUT querying for ups.status and getting a status returned but it is not querying for anything else so in the UPS reporting page in the TrueNAS UI there is nothing being reported.
Is there a way to get the data to show in the UI report for the UPS?
You could try the peaNUT app. I have found the TrueNAS remote NUT implementation to be incredibly fussy and intolerant of NUT installed on a raspberry Pi. Would work fine for a while and then an auto update or like event would cause the TrueNAS NUT agent to report the Pi as unreachable, etc.
My report of this issue was closed as a duplicate, which is of course perfectly ok.
Unfortunately the other issue in JIRA was also closed without any further action. They claim they have to wait for an upstream update of Netdata in Debian.
It is a bit sad that there is no solution for this but as long as it is properly monitoring the UPS status I am OK with it. I have other ways of monitoring the UPS but fingured that it would be nice to having the info in the TrueNAS UI.
I have worked for large companies in data center infrastructure for decades and I am not used to have to directly monitor the UPS from compute systems. It has been a luxury to have reliably redundant power to every rack. I guess that is one of the perks for in the data center for a power company.
There is a relatively simple and safe fix for UPS SLAVE stats reporting if you don’t mind having to apply such hacks in order to fix TrueNAS.
It absolutely blows my mind that TrueNAS back themselves into a corner by saying they have to wait for upstream Debian to fix Netdata - which may or may not ever happen - when it is so obviously broken in their own product and the fix is relatively trivial. Why bother fixing anything - just wait for Debian, right?
