search

zha-ng / zha-map

Build ZHA network topology map.
MIT License
57 stars 17 forks source link

LQI reported false in neighbor files #41

Closed Repsionu closed 3 years ago

Repsionu commented 3 years ago

HA latest on real core (venv), coordinator is Tasmotized ZbBridge in ZHA mode. Neighbor files report false LQI versus ZHA data (and real world distance) e.g. device with LQI=255 reports 0 :-( No big problem of course but very nice ZigZag reports false colors for LQI :-) Done some zha_map service rescans but no luck... If needed can show LQI differences in neighbor file vs ZHA data. Best, JR

Adminiuga commented 3 years ago

Enable debug logging and compare what device actually reports vs what is in the file.

Repsionu commented 3 years ago

OK will do tommorow, too late today 👍 Debug logging You mean... All Zigbee stuff? Now no errors in logs :-)

Adminiuga commented 3 years ago

There won't be errors, but all lqi information is coming from zdo mgmt_lqi_rsp messages, so just compare that those contain right information

Repsionu commented 3 years ago

Did some further tests- still no go :-( Added my ZHA device info and some neighbor files: ZigBee LQI neighbours_60a423fffe1dbdc1.txt neighbours_00124b001e731acb.txt neighbours_00124b001e73198a.txt neighbours_00124b001e722554.txt Maybe :-) Best, JR

Adminiuga commented 3 years ago

  1. those are different LQI and unlikely to be the same. ZHA reports LQI as seen by the coordinator from the "last hop", eg Device -> [hop] -> [last hop] -> coordinator -- it shows LQI for the messaged relayed by the last hop.

  2. ZHA MAP reports LQI as reported by the device for its particular neighbor.

Unless you can show in the logs that ZDO Mgmt_Lqi_rsp contains different data/LQI for a particular neighbor then i

Repsionu commented 3 years ago

So for "multihop" or lots of routers to extend range we can not use :-( ? All graphing is uniformative... OK, then so it is :-) Best, JR