Configuration blank in web administration #344
Comments
|
I reinstalled it and the configuration came back. Best regards ! |
|
Please reopen this, as I am having the same issue but reinstalling did not resolve it. |
|
I'm having the same issue. I have not tried to uninstall the add on and re-install because I don't want to risk making the problem worse. |
|
As 2 people have the same problem, I reopen this case. |
|
Same issue. It went to sqlite from mariadb. The issue is also duplicated here, with better logs. |
|
Same thing happened to me. I was surprised to see everything else (proxy hosts, SSL certs) still working as expected as creredentials are reset and the web UI is blank. Is there anything we can do to help debugging? |
|
Hello, same: |
|
The solution I found was to reinstall the nginx proxy manager addon. Your current config should still be saved in mariadb addon. as @q1t1 suggested earlier I did have to muck around with the proxy host address after reinstall to get it to find my local hostname again. I had mypihostname.local set. I was getting sent to a HA loading page with a 60sec retry timer. I set my local host back to IP number and clicked the external link from proxy hosts page. |
I tried that like 2 days ago, it works but after nginx update... Again it started to use the sqllite db. |
|
@frenck any idea what could be the issue ? Why this addon is not able to see mariadb and fallback on sqlite ? |
|
I've had to delete my 'nginxproxymanager' MySQL database and start again from scratch. Nothing else worked for me. |
|
There is a commit (NginxProxyManager/nginx-proxy-manager@ffe3db8) which could be related: it introduced the SQLite fallback and was introduced in Nginx Proxy Manager 2.9.12 (Add-on version 0.12.0). |
|
The question is: Why does it fallback (as in, what is wrong with SQL). |
|
Can you point us to the code where the SQL config is being injected/set? |
|
It doesn't use environment variables. |
|
Thx for the link. At least in the running container this information is not present. It might have been overwritten on startup though (see logs above). |
|
So, what I don't understand is: Why I don't have this issue? As in, I run this thing on multiple machines (including different test setups; I always test releases before shipping them). This sounds like something fundamentally wrong, but... I don't see it? (PS: will not respond for a bit, time enjoy my last hour of my Sunday for a bit with a beer 😉 ) |
|
I can't tell at the moment. It must be this check failing for us: Will try looking further into it in the next days. Fooling around with the proxy is always a bit messy as some of my systems rely on it working... Enjoy your beer, you deserve it! 🍻 |
|
Maybe it’s to do with Maria DB and not your add on? |
|
I had issue with an extension of mine that is using mariadb and was not able to connect to it but after restoring the extension... It works and still working until now so not sure if it's nginx or maria :/ |
|
Some additional parts of the log from my earlier post(not sure why it's renewing so much, although I probably have old certs hiding) and then a reboot log. Doesn't really provide much insight. |
Thany you very much for this! |
|
But after host reboot, I had to reinstall again nginx. Something weird is with mariadb or nginx |
|
Yes very wierd. Lost access. Reinstalling the addon doesn't completely work as then I needed to remove and re-add letsencrypt ssl config to get it to work again as it thinks it already had the certs. I just did this yesterday as well. :( reboot and lost access to NPM again. btw, it is a REBOOT and not a 'HA RESTART' that causes this I believe. |
|
well. This is how you restore access:
and then restart the addon. |
|
The one thing that still puzzles me the most: why is the proxy itself still working? Other than the login and Web UI, everything under the hood is doing its thing as always. This very page is served by Nginx Proxy Manager - if I stop the plugin, the page is no longer available.
Same for my HomeAssistant in the first browser tab at https://hassio.xxxxxxx.org/ Edit: Understanding a bit more here... The "working part" is coming from configurations injected via the volume mount |
How to get data from mariadb? I want move data to nginx in docker to another machine |
|
@calisro my apologies, i did not see that post. i only have 3 services running from hass anyway. but on a side note, until this problem arose, and after over 2.5 years of using hass in hyper-v... i did not know that i could use python from within the home assistant os thats installed in windows hyper-v. ive literally just discovered that i can run my own python scripts. and i just looked and it also runs perl scripts! so this all has been a bit of finding for me. now i have the resources to add a few more services to my domain! but if i have any more problems ill def try the fix in your post |
|
Just to keep this open... After previously having this problem and then doing the full delete/reinstall/set up from scratch a while ago Nginx proxy manager has been working fine. I was playing around with setting up access lists, and in the middle of doing that Nginx proxy manager became unresponsive. Closed the window, opened the add on again - no access using my previously working username/password. Remote access and certs still work so I'm not touching it for now, but it's very annoying that this is happening. |
@calisro - I've done this but it's not fixing the issue for me. When I run that command in SSH & Web Terminal (protection mode disabled) it appears to execute without error but there is no output from the command so I can only assume it is running correctly. Is that normal? |
Yes, copy (cp) command doesn't print anything when succeed. @timh767 In your case after restart I would check if it didn't get reverted back to bad one between your command and restart. There could be also be a possibility that for you config is fine but the database in mysql was lost, or that your account and setup was in sqlite to begin with... So check your database using phpmyadmin to confirm that your manager database is still there and is not empty. |
|
Thanks @kolorafa. Running that command gives me I can view the nginx database in phpmyadmin, and I can see my user id in the user table as well as my list of proxy hosts and certificates, and also the new access lists I was setting up when it crashed and locked me out. It appears to simply be not accepting my user password, but I know it's correct (saved in firefox password manager). Is there any way to reset a user password? |
|
OK, thanks to this link NginxProxyManager/nginx-proxy-manager#230 (comment) I reset my user password to changeme using phpmyadmin, opened NPM with my username and changeme password and was able to log in, all my proxys etc still there as before. Changed password via the NPM interface, logged out and back in again and all is good. Copied from that link by @chaptergy `For those not wanting a new admin account, but instead to just set the password of a specific account you can do the following: ` |
|
@timh767 That only works if the database is already using the correct one: mariadb. THat is a different issue. You still need to revert the production.json as mentioned many times in this thread. |
|
@calisro well my symptoms were exactly as described in the original post, and doing the above fixed it whereas none of the other methods in this thread made an difference. Thought that was relevant as others searching on this problem will arrive here and this may be a solution.
|
|
Same issue here happend after last update....got it nearly back to work with ssh custom terminal and docker exec... |
|
My Home Assistant certificate will expire in a few weeks so I thought i'd login to nginx pm to lookup the exacte date and noticed it wouldn't log me in. I use the Nginx Proxy Manager add-on with an internal certificate, so no letsencrypt. Add-on version 0.12.3 Additional information:
I can't explain what the issue is. The reverse proxy appears to be running correctly, yet the management web interface is empty. Almost as if the add-on is using two databases for the different parts. Setting the log level to debug the add-on indicates that it will be using the mariadb connection initially, then later stating that there is no environment variable provided the the database and that it will default to sqlite. Let me know if I can help to provide more information. |
this command works also for home assistant OS? |
|
There hasn't been any activity on this issue recently, so we clean up some of the older and inactive issues. |
github-actions
bot
added
the
stale
|
Also ran into this issue. See #409 Had to reinstall Nginx. Then it could connect to my MariaDB add-on again. It would however not work as reverse proxy anymore and I could not get new certificates. Therefore, also needed to reset the DB (thanks a lot for this switch, hidden in the configuration!) and had to start all over again. |
Yes |
Lovely |
|
Not stale |
github-actions
bot
removed
the
stale
Sorry for my noobish question, but where do i execute the command in home assistant os? In the terminal it doesnt work |
|
|
docker exec -it addon_a0d7b954_nginxproxymanager cp /defaults/production.json /data/manager/ works fine... however that happened, but solved :D thanks man ! |
|
This doesn't work for me:
Can only login with the default credentials (proxy is working but ui is blank). |
|
There hasn't been any activity on this issue recently, so we clean up some of the older and inactive issues. |
github-actions
bot
added
the
stale
Worked for me, also have to restart the add-on hope it won't happen again |
and others
Hi Frenck,
It's like the addon reset itself...
It tried to connect with my normal credentials, and it failed.
I the used [email protected] / changeme and it worked...
My configuration in the web administration is blanked but my old confiuration is still present in mariadb and works as intended...
Is it possible to re-link the web administration to the mariadb configuration ?
Thank you for your hard work, best regards,
The text was updated successfully, but these errors were encountered: