X6 update broke reverse proxy for webadmin

Support General Configuration
1

Have been using past versions without problems. Ever since updating to X6, I am no longer able to access the web admin using the same and untouched nginx reverse proxy. Every requests only ends with a 502 error. The logs that I get from Axigen aren’t any helpful either:

2024-12-10 14:39:36 +0000 08 f3d75804b6b6 WEBADMIN:00000410: [172.16.96.3:9000] connection accepted from [172.16.96.2:56718]
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000410: << GET / HTTP/1.1
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000410: << Host: axigen.xxxxx.com
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000410: << X-Forwarded-Scheme: https
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000410: << X-Forwarded-Proto: https
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000410: << X-Forwarded-For: xxx.xxx.103.1, 172.70.247.110
2024-12-10 14:39:36 +0000 08 f3d75804b6b6 WEBADMIN:00000410: connection closed with [172.16.96.2:56718]
2024-12-10 14:39:36 +0000 08 f3d75804b6b6 WEBADMIN:00000412: [172.16.96.3:9000] connection accepted from [172.16.96.2:56734]
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000412: << GET /favicon.ico HTTP/1.1
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000412: << Host: axigen.xxxxx.com
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000412: << X-Forwarded-Scheme: https
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000412: << X-Forwarded-Proto: https
2024-12-10 14:39:36 +0000 16 f3d75804b6b6 WEBADMIN:00000412: << X-Forwarded-For: xxx.xxx.103.1, 172.68.195.168
2024-12-10 14:39:36 +0000 08 f3d75804b6b6 WEBADMIN:00000412: connection closed with [172.16.96.2:56734]

And according to the Nginx’ error logs, Axigen closes the connection early for whatever reason:

2024/12/10 14:41:33 [error] 23240#23240: *19131 upstream prematurely closed connection while reading response header from upstream, client: 172.68.195.169, server: axigen.xxxxx.com request: "GET /favicon.ico HTTP/2.0", upstream: "http://172.16.96.3:9000/favicon.ico", host: "axigen.xxxxx.com", referrer: "https://axigen.xxxxx.com/"

Tried a wide variety of configurations, of which none helped. Disabled all the security in Axigen’s WebAdmin configs, and none helped. Its output level is set to max, unfortunately Aixgen’s logs still aren’t any help.

Web mail functions perfectly fine, although it’s using the exact same nginx settings

2

This may be where Nginx has too little of a timeout allowance or other similar issue.

3

Encountered the same behavior with Nginx Proxy Manager on 2 different machines, both installs with docker. In both cases the Admin interface can be accessed normally by ip:port via both http and https.

4

I’m also experiencing this exact same issue after upgrading. I tried various Nginx settings to increase wait times, increase buffer sizes, etc. to no avail. Axigen seems to immediately close the connection, so it doesn’t seem to be related to any timeout issues. Direct connection works just fine.

5

Hello all,

Thank you for your reports.

We have managed to replicate on our side and most probably we’ll have a fix soon.

Till than you may try the workaround to disable adding the ‘X-FORWARDED-FOR’ header at the HTTP proxy level (only for WebAdmin sessions).

As soon we’ll have a confirmed ETA for the fix we’ll make an update here as well.

Best regards,
Ioan

1 Like