Hi,

I'm facing an issue with Unraid that i'm not being able to solve.

I'm running Unraid 7 (almost since release without any issue) and had a power outage followed by a ups failure past weekend, which caused an unexpected shutdown.
Parity check went through successfully and i didn't see any issue, but yesterday i tried using the console popup to check something and it kept closing after a bit.

I investigated further and saw that for some reason nginx is constantly restarting, taking unraid-api with it (no idea for how long this has been happening  since i have not used the unraid ui for some time)

 

I've tried running in another browser or in private mode, with no browser extensions and it still happens.

 

From the logs i can see these lines, but have no idea what they mean:
 

rc.nginx: Stopping Nginx server daemon forcibly...
unraid-api[139485]: 👋 Farewell. UNRAID API shutting down!
nginx: 2025/01/30 16:48:06 [error] 137378#137378: MEMSTORE:00: tried adding WAITING chanhead 00000000009B99D0 m/
nginx: 2025/01/30 16:48:06 [error] 137378#137378: MEMSTORE:00: tried adding WAITING chanhead 00000000009B99D0 m/
rc.nginx: Nginx server daemon...  Stopped.
publish: curl to update1 failed
publish: curl to temperature failed
publish: curl to update3 failed
rc.nginx: Starting Nginx server daemon...
rc.nginx: Nginx server daemon...  Started.

I don't know if something got corrupted or anything like that.

 

Can someone help me?

valhalla-diagnostics-20250130-1639.zip

Try booting in safe mode to see if it still happens without Connect, the API is from that plugin.

I've tried booting in safe mode, and let it run for like 30 minutes.

During that time i didn't see any error and the popup i had opened stayed open for the whole time, so i guess everything works correctly in safe mode.

 

I now got back to normal mode and i'm getting the nginx crash every 2-3 minutes.

That suggests a plugin is the problem, start by uninstalling Connect, if it resolves the issue, you can try installing again, in case something is corrupt, if the issue return, there's likely a config issue.

Tried uninstalling Connect but it didn't solve the issue, still got the nginx restart every couple of minutes
 

Jan 30 18:38:41 Valhalla rc.nginx: Stopping Nginx server daemon forcibly...
Jan 30 18:38:42 Valhalla rc.nginx: Nginx server daemon...  Stopped.
Jan 30 18:38:47 Valhalla rc.nginx: Starting Nginx server daemon...
Jan 30 18:38:47 Valhalla rc.nginx: Nginx server daemon...  Started.

 

Just to add some more information (don't know if it can help), i noticed that my nginx error.log is being spammed with logs like:
 

2025/01/30 19:00:53 [alert] 267282#267282: worker process 271095 exited on signal 6
ker process: ./nchan-1.3.7/src/store/spool.c:479: spool_fetch_msg: Assertion `spool->msg_status == MSG_INVALID' failed.
2025/01/30 19:00:53 [alert] 267282#267282: worker process 271129 exited on signal 6

 

Did you reboot after uninstalling? And if it's not Connect, but it doesn't happen in safe mode, it's likely another plugin, you need to test after removing them.

Hey, so i had some time this weekend and i figured out what was happening (spoiler: i was dumb 😅).

So, before upgrading to v7, i was using Tailscale on a docker container.

After upgrading i started using the Tailscale plugin and stopped the container (just stopped, didn't remove it).

Everything was working fine, but i forgot an important detail.

The old Tailscale container had autostart turned on.

When unRAID rebooted due to the power outage the container started again, and i ended with two Tailscale instances running (plugin and container).

That was the cause of everything.

Removing the container made the issue of nginx restarting go away.

 

Thanks for the help 😅