Hi,
zBSL is the switch to put the cc2652 module into bootloader mode.
zRST is the reset for the cc2652. (Basically restarts the module)
They are there to be able to put the module into the bootloader mode without needing to use the button on the board. The toggle for firmware update mode is just a script that works with these switches(pins)
@tube0013 - I’m not sure how to (or even if I can) edit the Github WIKI - but I lost that little piece of paper you sent with my POE coordinator. Thus I’ve been banging my head against the wall why I couldn’t ping my coordinator with tube_zb_gw_cc2652p2.local.
Finally stumbled across something somewhere (I don’t remember now) telling me I should have been using tube_zb_gw_cc2652p2_poe.local. This was extra confusing because the device name in my Unifi router was tube_zb_gw_cc2652p2
So - suggesting that the Wiki in Github be edited to:
-for CC2652p based gateways: tube_zb_gw_cc2652p2.local
-for CC2652p POE based gateways: tube_zb_gw_cc2652p2_poe.local
-for EFR32 based gateways: tube_zb_gw_efr32.local
Any chance you could post what was written on that piece of paper that you shipped with the coordinator? Thanks!
thanks for that, sorry for the confusion. I will update github with the suggestion. and what you have above is correct
the slip essentially says:
for Z2M the device is specified as the following:
serial:
port: tcp://IP (OR mDNS):6638
for ZHA, use Autodiscovery or specify:
socket://IP (OR mDNS):6638
Baud Rate 115200
Software Flow control
I’m trying to figure out the best way forward though as ESPHome now complains about the _'s in the names/mdns addresses. I may be moving to hyphens to be more compliant.
I’m also going to change wording to strongly advise using the IP address (reserved for the device with the router or with a static IP using a custom ESPHome build -as described on github). Auto Discovery in HA for ZHA handles this well, it will actually use the IP discovered via mDNS, and if it’s not there on startup fallback to the mDNS address to re-discover the IP.
Finally up and running with a little Discord help!
Got stuck at herdman not starting until I was told to replace mDNS with the actual IP. I thought I was OK as long as the IP reservation was static but as soon as I changed to numerical IP it started right up and was able to immediately connect my first switch.
Mosquitto and Z2M are both running in docker-compose containers.
Ahh! I had the mDNS address specified. That’s why my Zigbee2MQTT instance was taking forever to reconnect after a host restart (OS update).
Have changed to the (static) IP address. Thanks.
Version 2 of the CC2652 Ethernet Coordinator is up in the store and on the waiting list!
What’s new?
New look 
CH340B Serial to USB module for optional USB connection or debugging the ESP32 without extra Hardware.
LEDs which work with the 20210708 Z-Stack Firmware build
Boot (firmware recovery) and Reset buttons for the ESP32
The foundation is still the same CC2652p2 module and the WT32-ETH01 I’m now offering a variation without the WT32-ETH01 as it’s perfectly usable over USB without. I’m working on a removable insert for the case now to cover jack opening. and if circumstances change you can alway add it later, as it now takes the stock WT32-ETH01 which is available from multiple online sources.
The CH340B is the one version of the CH340 which supports eeprom programming, so when in USB mode this device will be compatible with HA’s USB discovery for ZHA to help new users get up in running (I still have to do the small PR but will get it in before beta next week).
Hey, first of all, really cool project and updates. Nice to see what you have created. 
I have three of your routers with the hope they have enough power to cover my house.
But I struggle already with short distances between two floors.
The coordinator is in the cellar, the first router is straight above and has a quite good connection to the coordinator (~20 lqi, I know lqi is not the best measure) but could be better.
I placed the second router another floor above and it is hard to get it paired to the first router.
A few tries it was poor connected to the coordinator two floors under it, but not to the router on groundfloor.
Another happening was, that a aquara sensor was paired to the coordinator one floor under instead to the router 5 meters away.
If I try to allow joining only throw the first router, nothing happens, only when the second router/sensor is near by the first one.
All testet with Z2M (v 1.18, 1.20, 1.21) and ZHA. Coordinator with newest firmware (20210708)
To reset the router it is enough to push the BSL button, right?
It feels like my sensors and bulbs have better connections to the coordinator than your routers and I have no idea why 
Edit: I also changed my Wifi to Channel 1 & 6 and set the Zigbee-Channel to 25.
Just received a coordinator last week. It was working fine till today when it started losing connection. I tried reflashing but that didn’t help. Ethernet port shows active but green led is flashing non stop. Ideas? It might got 5 mins then drops.
Can you dm your name or order number so I can look up when you ordered? Thanks
The led on the Ethernet port usually flashes continuously. Does it have an ip? Is it reachable in a browser to the esphome page?
Any changes to your network? You may want to try a static ip, which requires a esphome build, the details are on my GitHub or I can help and complile a bin file for you if you let me know the settings.
We can troubleshoot via DM too. Knowing exactly which coordinator you have is helpful for me too.
Thanks
Yeah had almost same problem. It was working great til today and it just stopped working out of nowhere. It doesn’t have an IP anymore and the devices aren’t reachable. Tried to reset but in the end I think I’ll have to go back to the sonoff.
But that’s really sad because it was working great til this afternoon.
which device do you have? are you able to try re-flashing the ESP32?
DM me with your name/order# for more support and options I’ll do my best to help.
My CC2652P2 (V1) just arrived today! 
I immediately encountered the same issue with Z2M, running as an add-on in HAOS in a VM. Stuck in a bootup loop with the error:
Zigbee2MQTT:error 2021-10-12 11:59:55: Error: Error while opening socket
at Socket.<anonymous> (/app/node_modules/zigbee-herdsman/src/adapter/z-stack/znp/znp.ts:198:24)
at Socket.emit (events.js:412:35)
at emitErrorNT (internal/streams/destroy.js:106:8)
at emitErrorCloseNT (internal/streams/destroy.js:74:3)
at processTicksAndRejections (internal/process/task_queues.js:82:21)
Swapping over to IP solved it immediately.
Is there any way to get mDNS to work with Z2M?
I don’t know, if anything it’s an issue with the container.
Got it, thanks. I’ll got my router to reserve an IP for it, no biggie. Working really well!
Just wanted to think @tube0013 for the great product and support. I had a device with a bad ethernet port and after some troubleshooting he said he was just going to replace it at his expense. Replacement device has been running strong for a week now. My mesh hasn’t been better and I came from conbee/ZBBridge. So many lines!!!
Not sure if this is the best place to ask the question… but here goes. I have a Tubes ZB coordinator and I am looking for a simple remote to use. I found these, but in the fine print it says it will only work with a Tuya Zigbee Hub?
Would appreciate some advice, and if these remotes are no good, perhaps there are other devices?
Working, see :
