Open aturri opened 1 year ago
This is odd...
I am running all my instances of HA behind a haproxy reverse proxy service in my routers. The proxy is taking care of the SSL processing. I have the same http:
setup as you have., except that trusted_proxies
specifically points to the gateway (e.g. 192.168.10.1
)
Is the 504 error generated by your reverse proxy?
Is your problem specific to the miele integration?
I cannot really see that we handle http requests differently when setting up the config flow than we do during normal operation.
I don't know who is generating the 504 error and neither which is the URL that is generating that error. However it's specific to the Miele integration.
I cannot really see that we handle http requests differently when setting up the config flow than we do during normal operation.
Exactly the same... I'll try to investigate
I've just installed this integration and just get a spinning circle when trying to configure it - Im accessing my home assistant via Nabu Casa (https://xxxxxxxxxx.ui.nabu.casa/) which I guess is a reverse proxy of sorts.
I'm not currently on my local LAN so can test further later on, but I suspect this might be the same issue?
@bradleykite This is the first report that there is a problem with setup through Nabu Casa. Please try to set up the integration from LAN.
Hi, OK I've been able to test a little more.
On clicking on "Add Integration" and then selecting "Miele", the UI just displays "Please wait, starting configuration wizard for Miele" - the pending request is a POST for /api/config/config_entries/flow
The POST data is: {"handler":"miele","show_advanced_options":true}
After waiting 21 mins, the browser closed the TLS connection and the UI never allows the integration to be configured.
I'm going directly to the Home Assistant IP - no reverse proxy or Nabu Casa.
At the same time, the main hassio log file displays essentially that it detected the browser close the TLS connection
2023-01-10 16:28:19.932 ERROR (MainThread) [homeassistant] Error doing job: protocol.resume_writing() failed Traceback (most recent call last): File "/usr/local/lib/python3.10/asyncio/transports.py", line 300, in _maybe_resume_protocol self._protocol.resume_writing() File "/usr/local/lib/python3.10/asyncio/sslproto.py", line 522, in resume_writing self._app_protocol.resume_writing() File "/usr/local/lib/python3.10/site-packages/aiohttp/base_protocol.py", line 31, in resume_writing assert self._paused AssertionError
Not sure what else I can provide to help troubleshoot this?
@bradleykite Moving your issue to separate issue #198
Hi, I've experienced the same issue with another custom integration. It looks like to be a common problem with NGINX ssl proxy and Let's encrypt certificate. As a workaround I was able to start config flow by using the local IP address of the Pi with HA installed.
https://community.home-assistant.io/t/504-gateway-time-out-during-update/171010
I think that we could close this issue, as it is not related to this integration.
Thanks Andrea
Hey there,
I try to setup my Miele washing machine. Unfortunately, I run into the same timeout error when using my Nginx SSL proxy. When accessing the HA instance directly without SSL proxy inbetween, I just get the loading circle without any changes even after minutes of waiting. I'm using a totally fresh install of the HAOS in a virtual machine hosted by KVM. Any other integrations can be set up without problems.
Is there anything I can try/set to acquire more information about what's going on?
Thanks! David
Hey again, guess I found the problem. I have a native IPv6 internet access and distribute a public IPv6 address to each device in my network. After disabling DHCPv6 for my network interface of the HAOS, I'm able to setup the integration successfully.
So, I guess there might be some problem with accessing the Miele cloud over IPv6 here.
I encountered the same issue as the original reporter, and fixed it disabling in configuration.yaml
:
#http:
# use_x_forwarded_for: true
# trusted_proxies:
# - 172.30.32.1
# - 127.0.0.1
# - 10.0.0.2
Unfortunately, the disabling IPv6 trick did not have any impact.
Now I had to revert the http
in the configuration.yaml
, hope it won't break the integration.
Bug ??
When setting up the integration in a new environment, the configuration flow cannot start and it shows the following error on frontend:
Impossibile caricare il flusso di configurazione: <html> <head><title>504 Gateway Time-out</title></head> <body> <center><h1>504 Gateway Time-out</h1></center> <hr><center>nginx</center> </body> </html> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page -->
I was not able to figure out why, but it happens in environments where there is a NGINX SSL proxy and with the following configuration:
http:
use_x_forwarded_for: true
trusted_proxies:
- 172.30.33.0/24
If I remove these lines from configuration file, then the config flow works correctly. If I try to start a new config flow after setting up the first one, it work too, even if there is the above configuration.What version of the integration are you using?
miele-0.1.4
Anything in the logs that might be useful for us?
No response
Additional information
No response