Wemo switches no longer working after upgrading from 35.x to 37.1 in Synology Docker

HI folks,

I recently (late December) setup Home Assistant 35.x on my Synology in a docker container. It automatically detected my Hue hub and two Wemo switches. I did not enter static IPs in the config file. Everything worked great with no discovery issues.

I upgraded to 37.1 yesterday morning, by downloading the 37.1 docker image on my Synology. After configuring the folder path for my config folder in Docker then and starting HA up neither my Hue or Wemo devices were recognized. There had been no other network changes during the upgrade.

I was able to get the Hue lights back by adding a specific entry for the Hue and setting a

static IP for the hub:
light:
platform: hue
host: 192.168.8.10

Reading another response it was suggested that I do similar for the Wemo switches with:

wemo:
  static:
    - 192.168.8.40 
    - 192.168.8.42 

This entry did cause the switches to show up and I see a

[homeassistant.components.wemo] Discovered unique device ############### in the log.

However, HA is neither showing the proper state for the switches or able to control them.

So then I tried fully resetting one of the switches from the iOS app and re-adding it to my network. Didn’t make a difference.

I also tried creating a new configuration file and folder, but it made no difference.

I see in the forum history that others have had issues with Wemo switches, but it looks like those issues were solved with earlier HA updates. There are also comments about the demo library being flaky, but I have not seen anything regarding this being an issue in 37.1.

I found this interesting post describing a way to fire off the commands from the command line, but the Docker image I am working with does not include this demo command line application.

I have put a large part of the weekend into trying to get these switches working again and have run out of ideas and Google results to work from. So I am hoping someone on the forum may have some ideas what might have happened in the upgrade or changed in 37.1.

As a side note, both the iOS Demo app and my Amazon Dot can both interact with the Demo switches without issue.

Try mapping WeMo discovery port to docker container

I had a lot of issues with WeMo discovery with docker. I remember that helped but no longer remember which port

Hmmm… thats makes a lot of sense. I see the port is now 49153 and I set it to that. Gave it the last 24 hours to discover the demo devices, but sadly it did not work.

Try port 8989
I added this port to fix this problem and believe it helped.
Haven’t used docker in a month so I’ve forgot if it was effective

I’m having the same issue; but on a docker container in a VM (not that it should matter). Tried 8989 as well with no luck. I tried hard coding too like you first suggested @TheOncomingAutomatio but I see a couple of switches, not all.

Not sure what could be causing this :confused:

Upon thought I never got WeMo working in docker.

I was pulling GPIO sensor data from 2nd HA instance on RasPi.
I sent WeMo state along with sensor data using eventstream component.

Apologize for wild goose chase.
I will look into this when I have time. I prefer docker and will likely move back.

@tmjpugh I did have Wemo working in Docker, so I know it can work. Just no idea why it stopped working after the upgrade.

I tried port 8989 and it didn’t work. I appreciate the suggestion even if it was a goose chase in the end.

Ive got wemo insight switches setup without discovery in my HA. I’ve separated the switches from my configuration.yaml into a switches.yaml with the code like this:

switch: !include includes/switches.yaml

  - platform: wemo
    static:
      192.168.1.101

  - platform: wemo
    static:
      192.168.1.102

They show up based on the names given in the Wemo app, which I then given friendly names to.

I replaced my wemo entry with what you suggested:

switch:
-platform: wemo
static: 192.168.7.60

-platform: wemo
static: 192.168.7.62

And this is what I see in the log:

17-02-10 05:11:19 INFO (MainThread) [homeassistant.loader] Loaded switch.wemo from homeassistant.components.switch.wemo
17-02-10 05:11:19 INFO (MainThread) [homeassistant.loader] Loaded wemo from homeassistant.components.wemo
17-02-10 05:11:19 INFO (MainThread) [homeassistant.bootstrap] Setting up wemo
17-02-10 05:11:20 INFO (Wemo HTTP Thread) [pywemo.subscribe] Listening on port 8989
17-02-10 05:11:20 INFO (Thread-4) [homeassistant.components.wemo] Scanning for WeMo devices.
17-02-10 05:11:25 INFO (MainThread) [homeassistant.core] Bus:Handling <Event component_loaded[L]: component=wemo>
17-02-10 05:11:25 INFO (MainThread) [homeassistant.bootstrap] Setting up switch
17-02-10 05:11:25 INFO (MainThread) [homeassistant.components.switch] Setting up switch.wemo
17-02-10 05:11:25 INFO (MainThread) [homeassistant.components.switch] Setting up switch.wemo

But the switches do not show up in the app or in the Current Entities list.

I’ve upgraded to 38.0 and my wemo switches took awhile to load but are still visible on the dashboard. I haven’t changed any settings in relation to them since I posted my previous settings. Are they not showing up in your wemo app?

I upgraded to 38.1 and gave HA some time to find the Wemo’s with no luck. My Wemo app on my phone can see and interact with the switches fine. I reset and restored one when this issue first came up to see if that helped. I have assigned the following ports in Docker in hopes that they would help: 49152, 49153, 8989. Still no luck.

Did you manage to resolve the issue?

Unfortunately, no. I ended up moving to hassio on an RPi3 and the switches show up now. So I ended up working around the issue in the end.

I’m having similar issues with WEMO and HUE lights in latest build (0.57.2).

Raised an issue: