lory995 Posted November 1, 2023 Share Posted November 1, 2023 Hi everybody, here again with another problem... I have no clue why, but i got nginx 503 problem out of nowhere, at least to my perspective. I didn't do anything strange with the server before it happened, except mess a bit up with tailscale plugin, since it was having problems with the WebUI as noted in this thread , and with wireguard as a tmp solution. It should have been solved with the latest update ( 1.50.1 not 1.52, didn't get to download it ) but there is a word in the post link that worries me.. "workaround". Anyway it seems no one else is having this problem so it's entirely possible something has gone wrong on my server alone, or it's an entirely other cause.. I have no clue to what to look for, so i leave the diagnostics to you guys hoping in your knowledge. Thanks you in advance! More info: -Server still works normally of course except for error 503 -I can access the server randomly. Like now. I just removed tailscale.plg via cli and now i can access, but yesterday i had access to the GUI for a brief period without removing it and then nothing again. tower-diagnostics-20231101-1349.zip Quote Link to comment
JorgeB Posted November 1, 2023 Share Posted November 1, 2023 Does it work if you boot in safe mode? 1 Quote Link to comment
lory995 Posted November 1, 2023 Author Share Posted November 1, 2023 At the moment the GUI is strangely accessible so i don't wanna risk breaking something. As soon as the problem arises again it's the first thing i'm gonna try and i'll update this post. Thanks jorge! Quote Link to comment
EDACerton Posted November 3, 2023 Share Posted November 3, 2023 That type of error message wouldn't be caused by the issue that the Tailscale plugin was having. I took a look at your diags though, and it appears that something is tripping the rate limiter for the WebGUI. There are lots and lots of these messages: Quote Nov 1 13:50:03 Tower nginx: 2023/11/01 13:50:03 [error] 11116#11116: *1631 limiting requests, excess: 20.440 by zone "authlimit", client: 192.168.1.4, server: , request: "GET /login HTTP/1.1", host: "192.168.1.238" Nov 1 13:50:03 Tower nginx: 2023/11/01 13:50:03 [error] 11116#11116: *1631 limiting requests, excess: 20.302 by zone "authlimit", client: 192.168.1.4, server: , request: "GET /login HTTP/1.1", host: "192.168.1.238" Nov 1 13:50:04 Tower nginx: 2023/11/01 13:50:04 [error] 11116#11116: *1631 limiting requests, excess: 20.121 by zone "authlimit", client: 192.168.1.4, server: , request: "GET /login HTTP/1.1", host: "192.168.1.238" Nov 1 13:50:05 Tower nginx: 2023/11/01 13:50:05 [error] 11116#11116: *1631 limiting requests, excess: 20.545 by zone "authlimit", client: 192.168.1.4, server: , request: "GET /login HTTP/1.1", host: "192.168.1.238" Nov 1 13:50:05 Tower nginx: 2023/11/01 13:50:05 [error] 11116#11116: *1631 limiting requests, excess: 20.419 by zone "authlimit", client: 192.168.1.4, server: , request: "GET /login HTTP/1.1", host: "192.168.1.238" Nov 1 13:50:06 Tower nginx: 2023/11/01 13:50:06 [error] 11116#11116: *1631 limiting requests, excess: 20.840 by zone "authlimit", client: 192.168.1.4, server: , request: "GET /login HTTP/1.1", host: "192.168.1.238" Nov 1 13:50:07 Tower nginx: 2023/11/01 13:50:07 [error] 11116#11116: *1631 limiting requests, excess: 20.715 by zone "authlimit", client: 192.168.1.4, server: , request: "GET /login HTTP/1.1", host: "192.168.1.238" 1 Quote Link to comment
lory995 Posted November 4, 2023 Author Share Posted November 4, 2023 That's strange... So tailscale simply let my Unraid Server (192.168.1.238) receive some kind of internal Tailscale request from my failover raspberry ( 192.168.1.4 ). That would at least give me something to work on maybe even with the problems that i'm having now with wireguard too. Note that i say that it was an Internal Tailscale request because since i removed the plugin ( via CLI which as i said i could perfectly access ) the problem officially solved itself ( i rebooted a couple times and didn't have the problem again ) and i never disconnected from the LAN the PiKVM. Any idea on how can i solve the problem? Possibly an easy way ( like a way to absolutely clean every trace of tailscale on a raspi maybe?). Usually i'm more interested in learning when possible but after 3 days of various troubleshooting between wireguard and tailscale ( not on this problem of course ) i'm quite exhausted and the time that i could allocate to this is ending. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.