[GH-ISSUE #1463] 2.1.17 => 2.2.1 (After update, no longer possible via nginx-ui web to send requests to /api/config)

kerem commented

2026-03-12 23:20:00 +03:00

Owner

Originally created by @dennisoderwald on GitHub (Dec 3, 2025).
Original GitHub issue: https://github.com/0xJacky/nginx-ui/issues/1463

Describe the bug
Since updating to the latest version, I can no longer edit, enable, or disable any pages. The request is constantly pending, and nothing is visible in the debug log. A timeout does not seem to be taking effect.

When I duplicate a new page from an existing one, it works until I try to enable it.

Under “Dashboard” -> “Sites,” it says “No sites found”—but there are 39 sites. Interestingly, under the other node there, a page of 39 is displayed.

Log Output:

:29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:208        Started sitecheck refresh goroutine
Dez 03 16:43:29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:97        Found 40 indexed sites
Dez 03 16:43:29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:138        Collected 0 enabled sites for checking
Dez 03 16:43:29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:211        Sitecheck refresh goroutine completed

To Reproduce
Steps to reproduce the behavior:

Go to 'Manage sites'
Click on 'Enable'
Wait...

Screenshots
If applicable, add screenshots to help explain your problem.

Info (please complete the following information):

Server OS: NixOS
Server Arch: amd64
Nginx UI Version: 2.2.1
Your Browser: Chrome

Additional context
I use Nginx-UI with the cluster function and also synchronize another node that is on the same version.

Originally created by @dennisoderwald on GitHub (Dec 3, 2025). Original GitHub issue: https://github.com/0xJacky/nginx-ui/issues/1463 **Describe the bug** Since updating to the latest version, I can no longer edit, enable, or disable any pages. The request is constantly pending, and nothing is visible in the debug log. A timeout does not seem to be taking effect. When I duplicate a new page from an existing one, it works until I try to enable it. Under “Dashboard” -> “Sites,” it says “No sites found”—but there are 39 sites. Interestingly, under the other node there, a page of 39 is displayed. Log Output: ``` :29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:208 Started sitecheck refresh goroutine Dez 03 16:43:29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:97 Found 40 indexed sites Dez 03 16:43:29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:138 Collected 0 enabled sites for checking Dez 03 16:43:29 ingress-fsn nginx-ui[2795091]: 2025-12-03 16:43:29.524 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:211 Sitecheck refresh goroutine completed ``` **To Reproduce** Steps to reproduce the behavior: 1. Go to 'Manage sites' 2. Click on 'Enable' 3. Wait... **Screenshots** If applicable, add screenshots to help explain your problem. **Info (please complete the following information):** - Server OS: NixOS - Server Arch: amd64 - Nginx UI Version: 2.2.1 - Your Browser: Chrome **Additional context** I use Nginx-UI with the cluster function and also synchronize another node that is on the same version.

kerem

2026-03-12 23:20:00 +03:00

closed this issue
added the
bug
label

kerem commented

2026-03-12 23:20:16 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 3, 2025):

I have the feeling that, for whatever reason, the nodes are no longer 100% in sync, regardless of the error. Is there a way to synchronize them completely?

@dennisoderwald commented on GitHub (Dec 3, 2025): I have the feeling that, for whatever reason, the nodes are no longer 100% in sync, regardless of the error. Is there a way to synchronize them completely?

kerem commented

2026-03-12 23:20:22 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 3, 2025):

The “Manage Configs” section also loads indefinitely... but only if a previous request was still pending...

@dennisoderwald commented on GitHub (Dec 3, 2025): The “Manage Configs” section also loads indefinitely... but only if a previous request was still pending...

kerem commented

2026-03-12 23:20:27 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 3, 2025):

I will try to solve it in the development version tomorrow. Could you please take a screenshot of the F12 network request interface when it shows infinite loading? Also, when synchronizing nodes, use the command "journalctl -f nginx-ui" in the terminal to check the logs.

@0xJacky commented on GitHub (Dec 3, 2025): I will try to solve it in the development version tomorrow. Could you please take a screenshot of the F12 network request interface when it shows infinite loading? Also, when synchronizing nodes, use the command "journalctl -f nginx-ui" in the terminal to check the logs.

kerem commented

2026-03-12 23:20:32 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 3, 2025):

@0xJacky I just tested it, and although the number in the dashboard is still incorrect in the “dev” branch, saving is working again.

So something between version 2.2.1 and dev definitely solved the main problem with saving for me.

@dennisoderwald commented on GitHub (Dec 3, 2025): @0xJacky I just tested it, and although the number in the dashboard is still incorrect in the “dev” branch, saving is working again. So something between version 2.2.1 and dev definitely solved the main problem with saving for me.

kerem commented

2026-03-12 23:20:37 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 3, 2025):

In the other node, I always get the following error:

Dez 03 17:05:00 ingress-nbg nginx-ui[2065837]: [0.148ms] [rows:0] SELECT * FROM `nodes` WHERE `nodes`.`id` = 3 AND `nodes`.`deleted_at` IS NULL ORDER BY `nodes`.`id` LIMIT 1
Dez 03 17:03:23 ingress-nbg nginx-ui[2065837]: 2025-12-03 17:03:23.582        ERROR        github.com/0xJacky/Nginx-UI/internal/middleware/proxy.go:35        record not found

Node ID 3 is its own

@dennisoderwald commented on GitHub (Dec 3, 2025): In the other node, I always get the following error: ``` Dez 03 17:05:00 ingress-nbg nginx-ui[2065837]: [0.148ms] [rows:0] SELECT * FROM `nodes` WHERE `nodes`.`id` = 3 AND `nodes`.`deleted_at` IS NULL ORDER BY `nodes`.`id` LIMIT 1 Dez 03 17:03:23 ingress-nbg nginx-ui[2065837]: 2025-12-03 17:03:23.582 ERROR github.com/0xJacky/Nginx-UI/internal/middleware/proxy.go:35 record not found ``` Node ID 3 is its own

kerem commented

2026-03-12 23:20:42 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 3, 2025):

Thank you for the information you provided. I'll take a look at this issue.

@0xJacky commented on GitHub (Dec 3, 2025): Thank you for the information you provided. I'll take a look at this issue.

kerem commented

2026-03-12 23:20:47 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 4, 2025):

In the other node, I always get the following error:

Dez 03 17:05:00 ingress-nbg nginx-ui[2065837]: [0.148ms] [rows:0] SELECT * FROM `nodes` WHERE `nodes`.`id` = 3 AND `nodes`.`deleted_at` IS NULL ORDER BY `nodes`.`id` LIMIT 1
Dez 03 17:03:23 ingress-nbg nginx-ui[2065837]: 2025-12-03 17:03:23.582        ERROR        github.com/0xJacky/Nginx-UI/internal/middleware/proxy.go:35        record not found

Node ID 3 is its own

I'm not quite sure if this issue has been fixed in the previous version. Will it still occur after all the nodes have been upgraded to the dev version?

@0xJacky commented on GitHub (Dec 4, 2025): > In the other node, I always get the following error: > > ``` > Dez 03 17:05:00 ingress-nbg nginx-ui[2065837]: [0.148ms] [rows:0] SELECT * FROM `nodes` WHERE `nodes`.`id` = 3 AND `nodes`.`deleted_at` IS NULL ORDER BY `nodes`.`id` LIMIT 1 > Dez 03 17:03:23 ingress-nbg nginx-ui[2065837]: 2025-12-03 17:03:23.582 ERROR github.com/0xJacky/Nginx-UI/internal/middleware/proxy.go:35 record not found > ``` > > Node ID 3 is its own I'm not quite sure if this issue has been fixed in the previous version. Will it still occur after all the nodes have been upgraded to the dev version?

kerem commented

2026-03-12 23:20:52 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 4, 2025):

Are you sure that when accessing from the main node, the Dashboard of the site shows 0 records?

@0xJacky commented on GitHub (Dec 4, 2025): Are you sure that when accessing from the main node, the Dashboard of the site shows 0 records? <img width="394" height="154" alt="Image" src="https://github.com/user-attachments/assets/86a443a4-b04b-474c-b415-635e78a34792" />

kerem commented

2026-03-12 23:20:57 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 5, 2025):

@0xJacky Sorry for the late reply. Yes, I'm 100% sure.

@dennisoderwald commented on GitHub (Dec 5, 2025): @0xJacky Sorry for the late reply. Yes, I'm 100% sure.

kerem commented

2026-03-12 23:21:02 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 7, 2025):

Have you tried logging out and then logging in again, or switching to a different browser?

@0xJacky commented on GitHub (Dec 7, 2025): Have you tried logging out and then logging in again, or switching to a different browser?

kerem commented

2026-03-12 23:21:08 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 7, 2025):

Caching is disabled, but yes, I have tried different browsers and desktop/mobile devices.

@dennisoderwald commented on GitHub (Dec 7, 2025): Caching is disabled, but yes, I have tried different browsers and desktop/mobile devices.

kerem commented

2026-03-12 23:21:13 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 7, 2025):

In the other node, I always get the following error:
Dez 03 17:05:00 ingress-nbg nginx-ui[2065837]: [0.148ms] [rows:0] SELECT * FROM `nodes` WHERE `nodes`.`id` = 3 AND `nodes`.`deleted_at` IS NULL ORDER BY `nodes`.`id` LIMIT 1
Dez 03 17:03:23 ingress-nbg nginx-ui[2065837]: 2025-12-03 17:03:23.582        ERROR        github.com/0xJacky/Nginx-UI/internal/middleware/proxy.go:35        record not found
Node ID 3 is its own
I'm not quite sure if this issue has been fixed in the previous version. Will it still occur after all the nodes have been upgraded to the dev version?

I have attempted to fix this issue. You can update the main node to the latest dev version and see if there are any other problems.

@0xJacky commented on GitHub (Dec 7, 2025): > > In the other node, I always get the following error: > > ``` > > Dez 03 17:05:00 ingress-nbg nginx-ui[2065837]: [0.148ms] [rows:0] SELECT * FROM `nodes` WHERE `nodes`.`id` = 3 AND `nodes`.`deleted_at` IS NULL ORDER BY `nodes`.`id` LIMIT 1 > > Dez 03 17:03:23 ingress-nbg nginx-ui[2065837]: 2025-12-03 17:03:23.582 ERROR github.com/0xJacky/Nginx-UI/internal/middleware/proxy.go:35 record not found > > ``` > > > > > > > > > > > > > > > > > > > > > > > > Node ID 3 is its own > > I'm not quite sure if this issue has been fixed in the previous version. Will it still occur after all the nodes have been upgraded to the dev version? I have attempted to fix this issue. You can update the main node to the latest dev version and see if there are any other problems.

kerem commented

2026-03-12 23:21:18 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 7, 2025):

The nodes are set to “dev” (latest version) – the error has now been fixed on the child node. However, the following still appears on both:

No sites found
Sites will appear here once you configure nginx server blocks with valid server_name directives.

@dennisoderwald commented on GitHub (Dec 7, 2025): The nodes are set to “dev” (latest version) – the error has now been fixed on the child node. However, the following still appears on both: > No sites found > Sites will appear here once you configure nginx server blocks with valid server_name directives.

kerem commented

2026-03-12 23:21:23 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 7, 2025):

I keep stumbling over this:

Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:208        Started sitecheck refresh goroutine
Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:97        Found 40 indexed sites
Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:138        Collected 0 enabled sites for checking
Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:211        Sitecheck refresh goroutine completed

Found 40 indexed sites vs. Collected 0 enabled sites for checking

@dennisoderwald commented on GitHub (Dec 7, 2025): I keep stumbling over this: ``` Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:208 Started sitecheck refresh goroutine Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:97 Found 40 indexed sites Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:138 Collected 0 enabled sites for checking Dez 07 07:27:08 ingress-fsn nginx-ui[2915510]: 2025-12-07 07:27:08.632 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:211 Sitecheck refresh goroutine completed ``` `Found 40 indexed sites` vs. `Collected 0 enabled sites for checking`

kerem commented

2026-03-12 23:21:28 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 7, 2025):

GET /api/site_navigation:

Response:

{"data":[]}

@dennisoderwald commented on GitHub (Dec 7, 2025): `GET /api/site_navigation`: Response: ``` {"data":[]} ```

kerem commented

2026-03-12 23:21:33 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 7, 2025):

Are the sites on the main node disabled?

@0xJacky commented on GitHub (Dec 7, 2025): Are the sites on the main node disabled?

kerem commented

2026-03-12 23:21:38 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 7, 2025):

No, all sites are actively in sync on both nodes and enabled.

@dennisoderwald commented on GitHub (Dec 7, 2025): No, all sites are actively in sync on both nodes and enabled.

kerem commented

2026-03-12 23:21:43 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 7, 2025):

Please try to update to the latest dev version. I have added some debug logs. You can check the log output after the update. Thank you for your feedback.

@0xJacky commented on GitHub (Dec 7, 2025): Please try to update to the latest dev version. I have added some debug logs. You can check the log output after the update. Thank you for your feedback.

kerem commented

2026-03-12 23:21:48 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 7, 2025):

Dez 07 17:43:43 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:43:43.895        ERROR        github.com/0xJacky/Nginx-UI/internal/nginx/sandbox.go:45        Failed to create sandbox: failed to copy base configs: read /etc/nginx/modules: is a directory

Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:97        Found 40 indexed sites

Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:108        Processing enabled site: domain.com-app with 0 URLs

[...]

Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267        DEBUG        github.com/0xJacky/Nginx-UI/internal/site/status.go:23        Site domain.net-ssh considered disabled (enabledPath=/etc/nginx/sites-enabled/domain.net-ssh exists=false, maintenancePath=/etc/nginx/sites-enabled/domain.net-ssh_nginx_ui_maintenance exists=false)

Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:138        Collected 0 enabled sites for checking
Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.268        DEBUG        github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:211        Sitecheck refresh goroutine completed

And this line repeats itself per page with the same number - but server_name is definitely set in sites.

domain.net-ssh is an stream not a site

@dennisoderwald commented on GitHub (Dec 7, 2025): ``` Dez 07 17:43:43 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:43:43.895 ERROR github.com/0xJacky/Nginx-UI/internal/nginx/sandbox.go:45 Failed to create sandbox: failed to copy base configs: read /etc/nginx/modules: is a directory Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:97 Found 40 indexed sites Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:108 Processing enabled site: domain.com-app with 0 URLs [...] Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267 DEBUG github.com/0xJacky/Nginx-UI/internal/site/status.go:23 Site domain.net-ssh considered disabled (enabledPath=/etc/nginx/sites-enabled/domain.net-ssh exists=false, maintenancePath=/etc/nginx/sites-enabled/domain.net-ssh_nginx_ui_maintenance exists=false) Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.267 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/checker.go:138 Collected 0 enabled sites for checking Dez 07 17:44:37 ingress-fsn nginx-ui[2927463]: 2025-12-07 17:44:37.268 DEBUG github.com/0xJacky/Nginx-UI/internal/sitecheck/service.go:211 Sitecheck refresh goroutine completed ``` And this line repeats itself per page with the same number - but `server_name` is definitely set in sites. domain.net-ssh is an stream not a site

kerem commented

2026-03-12 23:21:54 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 8, 2025):

Could you please provide the content of the "domain.com-app" configuration file? I'd like to check if I can reproduce it locally.

@0xJacky commented on GitHub (Dec 8, 2025): Could you please provide the content of the "domain.com-app" configuration file? I'd like to check if I can reproduce it locally.

kerem commented

2026-03-12 23:21:59 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 8, 2025):

Here an example config:

upstream upstreamName {
    server ip:port;
}
server {
    http2 on;
    server_name example.com;
    add_header Strict-Transport-Security "max-age=63072000" always;
    ssl_certificate /etc/nginx/ssl/xxx/fullchain.cer;
    ssl_certificate_key /etc/nginx/ssl/xxx/private.key;
    location / {
        include /etc/nginx/snippets/base-headers.conf;
        proxy_pass http://upstreamName;
    }
    location ~ /.well-known/acme-challenge {
        proxy_set_header Host $host;
        proxy_set_header X-Real_IP $remote_addr;
        proxy_set_header X-Forwarded-For $remote_addr:$remote_port;
        proxy_pass http://127.0.0.1:9180;
    }
}

additional base-header config snippet:

proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Host $server_name;
proxy_set_header Host $http_host;

I have only anonymized the fields; the page configurations are largely identical in structure. Under “Proxy Targets,” the upstream is also correctly detected and displayed as online with latency.

@dennisoderwald commented on GitHub (Dec 8, 2025): Here an example config: ``` upstream upstreamName { server ip:port; } server { http2 on; server_name example.com; add_header Strict-Transport-Security "max-age=63072000" always; ssl_certificate /etc/nginx/ssl/xxx/fullchain.cer; ssl_certificate_key /etc/nginx/ssl/xxx/private.key; location / { include /etc/nginx/snippets/base-headers.conf; proxy_pass http://upstreamName; } location ~ /.well-known/acme-challenge { proxy_set_header Host $host; proxy_set_header X-Real_IP $remote_addr; proxy_set_header X-Forwarded-For $remote_addr:$remote_port; proxy_pass http://127.0.0.1:9180; } } ``` additional base-header config snippet: ``` proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-Proto $scheme; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header X-Forwarded-Host $server_name; proxy_set_header Host $http_host; ``` I have only anonymized the fields; the page configurations are largely identical in structure. Under “Proxy Targets,” the upstream is also correctly detected and displayed as online with latency.

kerem commented

2026-03-12 23:22:04 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 8, 2025):

It seems that there is no "listen" statement in the configuration file, so nginx-ui has ignored it.

@0xJacky commented on GitHub (Dec 8, 2025): It seems that there is no "listen" statement in the configuration file, so nginx-ui has ignored it.

kerem commented

2026-03-12 23:22:09 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 8, 2025):

It seems that there is no "listen" statement in the configuration file, so nginx-ui has ignored it.

Sorry for the confusion. I had to talk to a colleague about this first. We have included the listen part in another snippet:

listen 8443 ssl proxy_protocol;
real_ip_header proxy_protocol;
set_real_ip_from 127.0.0.1;

This is also included on every site. There is still an HAProxy in front of nginx/nginx-ui, but that shouldn't affect the problem, as according to the statement in the dashboard, it is based on the server_name.

@dennisoderwald commented on GitHub (Dec 8, 2025): > It seems that there is no "listen" statement in the configuration file, so nginx-ui has ignored it. Sorry for the confusion. I had to talk to a colleague about this first. We have included the listen part in another snippet: ``` listen 8443 ssl proxy_protocol; real_ip_header proxy_protocol; set_real_ip_from 127.0.0.1; ``` This is also included on every site. There is still an HAProxy in front of nginx/nginx-ui, but that shouldn't affect the problem, as according to the statement in the dashboard, it is based on the `server_name`.

kerem commented

2026-03-12 23:22:14 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 9, 2025):

This is the currently known limitation. If "listen" is not directly defined, it will be ignored.

@0xJacky commented on GitHub (Dec 9, 2025): This is the currently known limitation. If "listen" is not directly defined, it will be ignored.

kerem commented

2026-03-12 23:22:19 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 9, 2025):

I have added support for the inclusion of the listen directive within the contents of include files in the latest dev version.

@0xJacky commented on GitHub (Dec 9, 2025): I have added support for the inclusion of the `listen` directive within the contents of `include` files in the latest `dev` version.

kerem commented

2026-03-12 23:22:24 +03:00

Author

Owner

@dennisoderwald commented on GitHub (Dec 9, 2025):

Thank you @0xJacky -this is, of course, an edge case and does not affect everyone, but for us it leads to the next problem, which I wanted to inform you about... Whether you find a solution for this or even see a need to fix it is up to you:

The problem occurs with wildcard server names and, of course, when the listen port is not the actual public port (in an HAProxy setup like ours).

As a first step, I just wanted to let you know about this. You can decide how to deal with it. :)

@dennisoderwald commented on GitHub (Dec 9, 2025): Thank you @0xJacky -this is, of course, an edge case and does not affect everyone, but for us it leads to the next problem, which I wanted to inform you about... Whether you find a solution for this or even see a need to fix it is up to you: <img width="340" height="223" alt="Image" src="https://github.com/user-attachments/assets/6393e21d-d5fc-4c07-807b-ce518132b8b6" /> The problem occurs with wildcard server names and, of course, when the listen port is not the actual public port (in an HAProxy setup like ours). As a first step, I just wanted to let you know about this. You can decide how to deal with it. :)

kerem commented

2026-03-12 23:22:29 +03:00

Author

Owner

@0xJacky commented on GitHub (Dec 10, 2025):

Perhaps in this case, we should add an option to allow users to configure an actual Host:Port for the detection.

@0xJacky commented on GitHub (Dec 10, 2025): Perhaps in this case, we should add an option to allow users to configure an actual Host:Port for the detection.

Rows
Columns

[GH-ISSUE #1463] 2.1.17 => 2.2.1 (After update, no longer possible via nginx-ui web to send requests to /api/config) #7932