Running Mimosa 5ghz equipment the frequencies are showing NAN in the wireless graphs. Not sure what Librenms is looking for but when I snmpwalk it, the OIDs are the same as their documentation;
SNMPv2-SMI::enterprises.43356.2.1.2.6.1.1.6.1 = INTEGER: 5760
SNMPv2-SMI::enterprises.43356.2.1.2.6.1.1.6.2 = INTEGER: 5760
SNMPv2-SMI::enterprises.43356.2.1.2.6.1.1.6.3 = INTEGER: 5630
SNMPv2-SMI::enterprises.43356.2.1.2.6.1.1.6.4 = INTEGER: 5630
Their object is mimosaCenterFreq.1, 2 ,3 and 4 for the above. Can we get that corrected?
Not sure if this also applies to their 11ghz stuff since I haven’t upgraded that firmware yet. Looks like they have fixed the snmp dying issue with the current firmware fyi.
tk
It already discovers those oids, can you post a discovery debug output?
Do you want it against all modules or a particular one?
Nevermind, I ran it against all. Looks like I can’t upload a .txt . Do you want it pasted in here or another way?
A couple of other like issues with Mimosa running the current firmware there.
Interface Non Unicast is reporting NAN.
In the Wireless reporting; SNR, TX/RX Power, and Noise Floor decimal place is off by one. Needs to move one more to the left, -872 should be -87.2.
Seems like I read some time they “corrected” their reporting which may have changed reporting here.
Looking over the discovery it is getting the values for the freq. so I guess its not making it into the Freq graphs.
In the Wireless reporting, the values look correct in the discovery just being reported and graphed incorrectly.
Not sure on the Non Unicast as it is not shown or I can’t find it in the discovery output.
Pastebin it somewhere (non-expiring), like http://p.libren.ms or https://gist.github.com
Thanks @tlkhorses
Looks like there are some silly mistakes in the code.
Can you please send test data from your B5?
./scripts/collect-snmp-data.php -h HOSTNAME
Murrant,
Frequencies are now showing correctly.
Don’t know if you want this in a separate thread but in the above I mentioned that in those other graphs, the decimal is in the wrong place. For instance it is graphing -872 instead of -87.2. From the file I provided it looks like it is providing the correct info when snmp queries.
I made that change so you are going to have to be more specific.
Looks like Noise Floor is 3 digits in the graphs, the other two are fixed.
Interface non unicast is showing NAN in the graphs.
EDIT:
Have to correct that. Looking at two different APs, both B5c and running same software, SNR is 3 digits on one graph and 2 on the other’s graph. The only difference I can see between them is the one showing three digits SNR is running two frequencies where the other is running one.