Shelly HT ERROR: connect EHOSTUNREACH after update

[ht38f]

ShellyHT.json

Shelly HT ERROR: connect EHOSTUNREACH after update. see attached photo and flow

[ht38f]

This Shelly HT https://www.shelly.com/en/products/shop/shelly-plus-h-and-t

[ht38f]

It is still the same issue with 10.21.0

[windkh]

@ht38f I tested your flow and it works here... can you make sure that the shelly's IP is reachable from where you run node-red?

[ht38f]

Shelly should be reachable. I did a factory reset, gave it a new static ip based on MAC address in the router. The sensor is mostly disconnected but shows up a few seconds in the router and disappears again. I assume this is to preserve battery

[windkh]

So why do you get an error if it works here? Hm strange

[windkh]

Did you press the inject node while the shelly is sleeping?

[ht38f]

I tried. Same issue. I now watched the app and waited for it to update.23.19:25 it updated the Shelly app.

[windkh]

@ht38f the AxiosError occurs when you inject while the shelly is sleeping. That is ok, as the node can not talk to the shelly in this state. Devices that sleep should be run in "callback" mode. Nevertheless the connect EHOSTUNREACH is strange. Is something logged to the node-red console?

[ht38f]

Shelly firmware is 20240223-141926/1.2.2-g7c39781

[ht38f]

I am running node red on a Victron Cerbo and have no access to the files :(

[windkh]

Shelly firmware is 20240223-141926/1.2.2-g7c39781

0240223-141926/1.2.2-g7c39781 is installed on mine

[ht38f]

So it’s the same. I just received a debug showing data from the sensor!

[windkh]

Did you set a password?

[ht38f]

No password

[windkh]

@ht38f what node-red version and what nodejs version do you use?

[ht38f]

3.1.5

[windkh]

do you use different IP Address ranges in your home? what is the IP of the victron device?

[ht38f]

Think js is v18

[ht38f]

Victron is 192.168.1.138

[ht38f]

Shelly sensor is also in cloud. That’s how I see it updates.

[windkh]

what happens if you enter the following into the browser http://192.168.1.207/rpc/Shelly.GetStatus

[ht38f]

I need to check. I am not at the install but working remotely from victron VRM. I tried to connect to 192.168.1.207 earlier today. I did not manage to connect. I believe the device turns on and connects to the router and then goes back to sleep so it is not more visible

[windkh]

will go to bed now... maybe we should do a teams meeting where you can show it to me.

[ht38f]

I have been watching the flow and suddenly it says connected and is green. A few seconds after it goes back to ehostunreach

[ht38f]

Have a good night! Thanks for the help!

[windkh]

@ht38f hi... sorry I forgot to ask if the error is still there or not?

[ht38f]

Hi. The error is still there. When I try to get the status http://192.168.1.207/rpc/Shelly.GetStatus I get this: The server at 192.168.1.207 is taking too long to respond.

This makes sense as the H&T is only connected to the router a few seconds before it goes back to sleep. All other Shelly devices work normal.

[windkh]

The shellies that normally sleep like HT or smoke or door/window should not be polled. Instead you should use callback mode.