Unable to connect to Home Assistant. Retrying in __ seconds -

[mikehanrahan]

The problem

very similar to Issue 263, I am unable to login to Home Assistant via external URL. Internal URL remains functional. No Reverse Proxy Manager involved.

My domain name was registered thru CloudFlare, and I use CloudFlare's DNS & Dynamic DNS.
I have uninstalled, re-installed and re-set this add-on many times, with the result always the same. I have been researching on HA community, reddit, and here on GitHub. I have pasted in the bits you have requested from others in this thread and the HA Community site.
I am happy to work with you in any way you wish to help resolve this. Please let me know if you need any additional Info

What version of Cloudflared has the issue?

4.0.10 (and earlier)

What was the last working version of Cloudflared?

No response

What type of installation are you running?

Home Assistant OS

Add-on YAML Configuration

external_host_name: subdomain.domain.com     #(anonomized!)
additional_hosts: []

Anything in the logs that might be useful for us?

HA Log:  
WARNING (MainThread) [homeassistant.components.http.ban] Login attempt or request with invalid authentication from *** (xxx.xxx.xxx.xxx). Requested URL: '/auth/token'.   (Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36)

Cloudflared Startup Log: 
Add-on: Cloudflared
 Use a Cloudflare Tunnel to remotely connect to Home Assistant without opening any ports
 
 Add-on version: 4.0.10
 You are running the latest version of this add-on.
 System: Home Assistant OS 9.5  (aarch64 / raspberrypi4-64)
 Home Assistant Core: 2023.2.2
 Home Assistant Supervisor: 2023.01.1
 
 Please, share the above information when looking for help
 or support in, e.g., GitHub, forums or the Discord chat.
 
[15:21:51] INFO: Checking add-on config...
[15:21:52] INFO: Checking for existing certificate...
[15:21:52] INFO: Existing certificate found
[15:21:52] INFO: Checking for existing tunnel...
[15:21:52] INFO: Existing tunnel with ID *** found
[15:21:52] INFO: Checking if existing tunnel matches name given in config
[15:21:53] INFO: Existing Cloudflare Tunnel name matches config, proceeding with existing tunnel file
[15:21:53] INFO: Creating config file...
[15:21:55] INFO: Validating config file...
Validating rules from /tmp/config.json
OK
[15:21:55] INFO: Creating DNS entry ***...
2023-02-07T20:21:56Z INF *** is already configured to route to your tunnel tunnelID=***
[15:21:56] INFO: Finished setting up the Cloudflare Tunnel
[15:21:57] INFO: Connecting Cloudflare Tunnel...
2023-02-07T20:21:57Z INF Starting tunnel tunnelID=***
2023-02-07T20:21:57Z INF Version 2023.2.1
2023-02-07T20:21:57Z INF GOOS: linux, GOVersion: go1.19.3, GoArch: arm64
2023-02-07T20:21:57Z INF Settings: map[config:/tmp/config.json cred-file:/data/tunnel.json credentials-file:/data/tunnel.json loglevel:info metrics:0.0.0.0:36500 no-autoupdate:true origincert:/data/cert.pem]
2023-02-07T20:21:57Z INF Generated Connector ID: 6157411c-9374-42d2-932e-6693ecf9a0fd
2023-02-07T20:21:57Z INF Initial protocol quic
2023-02-07T20:21:57Z INF ICMP proxy will use 172.30.33.3 as source for IPv4
2023-02-07T20:21:57Z INF ICMP proxy will use :: as source for IPv6
2023-02-07T20:21:57Z INF Starting metrics server on [::]:36500/metrics
2023-02-07T20:21:57Z INF Connection f6e0fd34-0c6c-49d1-939e-757759fd2121 registered with protocol: quic connIndex=0 ip=198.41.200.13 location=ORD
2023-02-07T20:21:58Z INF Connection 3d316fa8-cc94-444c-adf8-b330000e1955 registered with protocol: quic connIndex=1 ip=198.41.192.7 location=EWR
2023-02-07T20:21:59Z INF Connection f5f7a0cd-c35a-4b53-8093-d55f36c2b584 registered with protocol: quic connIndex=2 ip=198.41.192.67 location=EWR
2023-02-07T20:21:59Z INF Connection 99e7e1f2-abe9-43b1-8d3a-74285555027f registered with protocol: quic connIndex=3 ip=198.41.200.43 location=ORD

Steps to reproduce the issue

1. Access external url via any device: android, desktop, etc, over WiFi, LTE, etc
2. See HA Login Page
3. Provide Username & password
4. Receive Error on screen and in log file
   

Additional information

No response

[brenner-tobias]

Are you using any Cloudflare Access as additional authentification layer? Which browser are you using?

[mikehanrahan]

I use Chrome as a browser, and yes, have GitHub Authentication enabled on Cloudflare. I have also tried with no Access control, but still getting the same unsuccessful login result.

…

On Tue, Feb 7, 2023, 6:10 PM Tobias Brenner ***@***.***> wrote: Are you using any Cloudflare Access as additional authentification layer? Which browser are you using? — Reply to this email directly, view it on GitHub <#318 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABGK67PIZVRDJSFA6KFTCJ3WWLI5ZANCNFSM6AAAAAAUURDCGY> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[maramazza]

I have the same issue with 2 different fresh HA installations

[mikehanrahan]

I just tried 2 more browsers (edge, duckduckgo), getting the same result

[jeefm]

Same here, HA unable to connect retrying xx seconds. uninstalled and other recommandations done as well.

[Tekonsite]

I am having the same issue on my HomeAssistant. Mine is a vm on Proxmox. I had to do a restore back 2 days to the version before this recent update. I now have no issues. I turned off auto updates until this is resolved.

Logger: homeassistant
Source: components/cloudflare/init.py:95
First occurred: 12:40:35 AM (9 occurrences)
Last logged: 8:40:35 AM

Error doing job: Task exception was never retrieved
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/components/cloudflare/init.py", line 57, in update_records
await _async_update_cloudflare(session, cfupdate, zone_id)
File "/usr/src/homeassistant/homeassistant/components/cloudflare/init.py", line 95, in _async_update_cloudflare
records = await cfupdate.get_record_info(zone_id)
File "/usr/local/lib/python3.10/site-packages/pycfdns/init.py", line 105, in get_record_info
record_information.append(CFRecord(data["result"][0]))
IndexError: list index out of range

[AlpineWhite]

I am having the same issue on my HomeAssistant. Mine is a vm on Proxmox. I had to do a restore back 2 days to the version before this recent update. I now have no issues. I turned off auto updates until this is resolved.

Logger: homeassistant Source: components/cloudflare/init.py:95 First occurred: 12:40:35 AM (9 occurrences) Last logged: 8:40:35 AM

Error doing job: Task exception was never retrieved Traceback (most recent call last): File "/usr/src/homeassistant/homeassistant/components/cloudflare/init.py", line 57, in update_records await _async_update_cloudflare(session, cfupdate, zone_id) File "/usr/src/homeassistant/homeassistant/components/cloudflare/init.py", line 95, in _async_update_cloudflare records = await cfupdate.get_record_info(zone_id) File "/usr/local/lib/python3.10/site-packages/pycfdns/init.py", line 105, in get_record_info record_information.append(CFRecord(data["result"][0])) IndexError: list index out of range

I have the exact same symptoms and errors.

Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/components/cloudflare/init.py", line 57, in update_records
await _async_update_cloudflare(session, cfupdate, zone_id)
File "/usr/src/homeassistant/homeassistant/components/cloudflare/init.py", line 95, in _async_update_cloudflare
records = await cfupdate.get_record_info(zone_id)
File "/usr/local/lib/python3.10/site-packages/pycfdns/init.py", line 105, in get_record_info
record_information.append(CFRecord(data["result"][0]))
IndexError: list index out of range

Was working well before.

[mikehanrahan]

Could a common thread be that we all have multiple domains under management in CloudFlare?

[AlpineWhite]

Could a common thread be that we all have multiple domains under management in CloudFlare?

You are definitely onto something. I removed a site whose registration is expiring in 13 days and I was suddenly able to access my HA after going through the auth steps.

[mikehanrahan]

@maramazza @jeefm @Tekonsite - Do you all have multiple domains under management on CloudFlare?

[Tekonsite]

Yes I do but mine is fine now after I rolled back.

…

Sent from my Galaxy

-------- Original message -------- From: Mike H ***@***.***> Date: 2023-02-13 10:50 a.m. (GMT-05:00) To: brenner-tobias/addon-cloudflared ***@***.***> Cc: Tekonsite ***@***.***>, Mention ***@***.***> Subject: Re: [brenner-tobias/addon-cloudflared] Unable to connect to Home Assistant. Retrying in __ seconds - (Issue #318) @maramazza<https://github.com/maramazza> @jeefm<https://github.com/jeefm> @Tekonsite<https://github.com/Tekonsite> - Do you all have multiple domains under management on CloudFlare? — Reply to this email directly, view it on GitHub<#318 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/APUWVJRE44K7IBRA2FCKKBLWXJJ3PANCNFSM6AAAAAAUURDCGY>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[maramazza]

@mikehanrahan yes 4 domains

[jeefm]

no, I had 1 ***.tk Jean-François Millaire

…

On Mon, Feb 13, 2023 at 10:50 AM Mike H ***@***.***> wrote: @maramazza <https://github.com/maramazza> @jeefm <https://github.com/jeefm> @Tekonsite <https://github.com/Tekonsite> - Do you all have multiple domains under management on CloudFlare? — Reply to this email directly, view it on GitHub <#318 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AVBIE27OOAPUTKOMB27LWATWXJJ3RANCNFSM6AAAAAAUURDCGY> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[nicopasla]

I have exactly the same problem, I have only one domain registered on Cloudflare, My domain provider is Namecheap with a paid xyz domain

[matejlacko90]

I have also same problem, with one paid domain on the Cloudflare.
Cloudflare add-on logs are without errors. After I log in , the same screen appear (Unable to connect to Home Assistant).
Reason is that the login attempt came from my external IP address.

[mikehanrahan]

I added all of the CloudFlare IPv4's to the trusted proxy list, restarted HA. No difference - still getting the very same error after providing HA credentials. The "Login attempt Failed" notification on the HA dashboard says "Login attempt or request with invalid authentication from ..." sometimes shows an IPv4 address (not CloudFlare or Docker related), and sometimes (usually from a mobile device) shows a IPv6 address.

[johntdyer]

whats interesting is that if I refresh a few dozen times it starts working...

[Coedy]

I also have this issue purely from my work device (trying to tweak automations on my dinner break) It seems they use cloudflare too (198.41.200.XX appears in my addon's logs) I can access it via my phone (Sim data) and home computer (wifi locally but via my domain name). I just cant access it via my work devices presumably because they are using their own tunnelling?
Being a non-network guy this is a bit of a nightmare for me haha.

[Liionboy]

same problem from today at me

[chrisverros]

Some problem for me right now. It was working perfect for the last week (from the initial setup until now)
Suddenly "Unable to connect to Home Assistant."
Resetting the service or even stop it and start it didn't solve the problem.
Restarting the whole HA vm did the trick and it is working.

[Liionboy]

At me i dont know how but now is work again

[dula34]

I have same problem. Connection problem resolved after restarting add-on. After 3 hours problem appeared again

[MartinSchillinger]

I get this error too. There must be some kind of authentication problem.

The Browser is doing a POST Request to /auth/token with this payload:

client_id: https://<tunnel domain>/
code: bb48c248239f4b0ba0e58eb1b0fca6b7
grant_type: authorization_code

which gets this response:

{"error":"invalid_request","error_description":"Invalid code"}

[Raftaman34]

I'm having this issue as well, appears to work fine on a network outside my house, such as mobile network. I am also unable to access any additional hosts internally as well.

[tamorgen]

I started having this problem just today. No devices will connect externally and now I'm getting 400: bad request. I've restarted HA, restarted Cloudflared, and cleared the cache on the CloudFlare management console (not HA).

[Coedy]

Interestingly. This doesn't seem to be an add-on issue, it seems to be an issue how home-assistant plays with cloudflare.

I was fed up of this issue so tried disabling the add-on and installing the equivalent 'community add-on' on unraid (which is the same box that home assistant runs on in a VM) as I thought it would be easier for me to pass through other services as they'd be on the same 'level' as that add-on.

All other dockers get exposed perfectly (and still do). But home assistant shows the same issue after a week or two (sometimes only an hour or two) of using a new tunnel. I imagine it's the proxy system of home assistant that's fighting against cloudflare but I can't work out how to fix it. I've tried adding the cloudflare IPs to the 'http:' section of the home assistant config but still get the same.

[judahrand]

I suspect this has something to do with refresh tokens. Whenever I revoke them things resolve (for a while).

[dragonflyuk]

Can that be done remotely?

…

On Wed, Apr 5, 2023 at 6:30 PM Judah Rand ***@***.***> wrote: I suspect this has something to do with refresh tokens. Whenever I revoke them things resolve (for a while). — Reply to this email directly, view it on GitHub <#318 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AHZT3Q3GCOSE6OO67DRMOXTW7WTYHANCNFSM6AAAAAAUURDCGY> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[elcajon]

I am moving this to the Discussions section as it is not a bug in the add-on itself.

[JBarzdo]

Any updates?