I apologize if this is already explained somewhere, but I have been googling since last night, and nothing seems to fit my situation. I don't know which version of unraid I was on prior, but the last time I was able to see the gui it listed 7.2.2, and a day or two prior I saw there was an update available, presumably to this version.

Everything was working normally on Sunday and Monday, in the last two-three days. I was moving PC backup files from one folder to another, and it was taking forever, but they were old backups so I don't think there was any access issue or anything. I had been having issues accessing HAOS remotely, but not being very familiar, I assumed that was user error.

Yesterday, I was trying to access my Vaultwarden instance, and I couldn't connect. I assumed it was due to the server hanging, so I remoted in vial Tailscale, and rebooted it. After a few minutes I was able to connect to Vaultwarden, but in retrospect, I think it was the PC i was using and not my Vaultwarden instance. Anyway, when I got home, I couldn't login to the gui on the local network. Since I had accessed it via Tailscale, earlier, I tried that and was able to login. that's when I saw it had updated to 7.2.2.

One of my google searches indicated Tailscale plugin might be the culprit, and I should remove it. So I did, and then I lost gui access via Tailscale as well as local network access. I tried different browsers, incognito browsers, clearing cache and cookies, nothing lets me connect to the gui now. I read that possibly my network.cfg and network-rules.cfg might be corrupt, and to delete them and let new ones be created. I did that, but no change to access.

Tonight, I pulled the server, and attached it to a monitor and keyboard. I was able to boot, and got to the CLI login. I added 'unraidsafemode' to my syslinux.cfg and it seems to mention it in the diagnostics, but I it didn't help me access the gui. I removed the safemode option, and pulled another set of diagnostics. I have not been able to ssh into the server either, but again that could be operator error.

I am to the point of trying to rollback to the prior OS version, but I thought it might be good to seek the wisdom of more experienced persons, before messing things up any more than I have.

I have two versions of the diagnostics to upload, which are hopefully attached.

server-diagnostics-20260121-2242.zip server-safemode-diagnostics-20260121-2229.zip

I did save my network files before deleting from the flash drive, so if the best course for now is a rollback, I have those files to replace.

I don't see anything in the log that explains the GUI not working. What do you get when you click the link below, "not found," "connection refused," or other?

http://192.168.86.100

I will check when I get home, but I believe the error was that the address took too long to respond.

I had a similar issue, but delecting all my tailscale machines fixed it. Suggest checking out ipconfig and see what your pc is using for an ip address.

8 hours ago, demanding-chief3698 said:

I had a similar issue, but delecting all my tailscale machines fixed it. Suggest checking out ipconfig and see what your pc is using for an ip address.

The ip address assigned to the PC I am using to try to access the Unraid server, right?

I neglected to mention that I also tried to access it from my phone browser, but I likely need to check it's IP also.

I just tried accessing the server again from the local IP. I used Chrome, and then an Incognito tab in Chrome. Chrome responds with "This site can’t be reached

192.168.86.100 took too long to respond." And that is the response in both the regular and Incognito tabs. I also tried Firefox: The connection has timed out

The server at 192.168.86.100 is taking too long to respond.

I confirmed my IP, subnet, and gateway are as expected: 192.168.86.53; 255.255.255.0; 192.168.86.1. And I verified that the IP the server has is provided by my router as a reserved IP, so the server is using DHCP.

As for the Tailscale machines, I don't know how I find that data in Unraid. Is that stored on the flash drive?

While I would like to fix this instance of Unraid, I am leaning more towards going back to the prior version. I

You can boot with a different flash drive using a stock Unraid install, no key needed, then see if that works; that will help confirm if there's a config problem with the current install or an external LAN issue.

14 hours ago, JorgeB said:

You can boot with a different flash drive using a stock Unraid install, no key needed, then see if that works; that will help confirm if there's a config problem with the current install or an external LAN issue.

Thanks for that suggestion. I will try that when I get home.

While I haven't made any recent network changes, it always feels like my browsers and my router are plotting against me.

I tried creating a new USB, but either it isn't compatible, or I am doing something wrong. But then I remembered that I have another USB I created for another server that's not in use, and I plugged it in. It's running 7.1.4, and I can login from the local IP just fine. I don't want to mess around like this too much, for fear of messing up drive assignments on that USB, or worse messing up my existing shares on my main server.

Any suggestions for next steps? Do I back up this extra USB, and try to create a new UNRAID install on it to see if the version is the issue? Should I attempt a rollback after saving all the files off the original USB? As much as it's always a work in progress, I really don't want to have to recreate everything from scratch if there is a way to avoid it.

All your suggestions, and any additional feedback, are greatly appreciated.

In case I didn't mention it, the router has the IP address for the server reserved, so it connected on the same IP as was used previously. Only difference is that the original config had my two onboard ethernet ports bonded, and this config didn't. Would that have anything to do with any of the issues I encountered?

I found another newer USB, installed 7.2.3, and it boots and I can login from a browser using the same router assigned IP address. I don't want to install any key, so I stopped there.

Is there a way to unbind my ethernet ports on the original USB drive, or is that really not likely part of the problem?

Thinking out loud, since one of the first troubleshooting steps was deleting Network.cfg and Network-rules.cfg and letting them be recreated, would the bond between eth0 and eth1 be in place still, or would that have been reset with the new network files? Just trying to rule out specific issues. I am also going to create another USB with the same OS version I am having issues with to make sure that OS version isn't the cause. Not sure why I didn't do that the first time.

With the USB setup for 7.2.2 I am able to get to the same new UNRAID install screen, at the same local IP address. So I am thinking it's a corruption of some kind in my USB drive.

Edited January 24Jan 24 by spikeekips

OK, last update and I will leave this thread alone for a bit, and stop blowing it up. I copied my original /config folder to the new USB running 7.2.2 and I am presented with the same inability to login via the GUI. So that means there is a problem with my /config folder then, correct? I didn't move anything else over, only the /config folder.

The config folder contains your configuration. Unless you intentionally put something else on flash, the only thing on it are the OS archives and your config.

37 minutes ago, trurl said:

The config folder contains your configuration. Unless you intentionally put something else on flash, the only thing on it are the OS archives and your config.

I'm not sure what you mean, unless you are just clarifying the purpose of the flash drive.

I had seen other tutorials that indicated the main thing needed to get your system back up, assuming it was a drive failure, was the /config folder. But it seems like that folder from my flash drive is the issue, as I can boot into unRAID, at least to the point of requesting a trial key, with a stock install of 7.1.4, 7.2.2, and 7.2.3, so evidently it isn't the OS version. And since I used a different USB flash drive, and copied over the config folder, it doesn't seem to be the flash drive either, unless it somehow compromised the folder contents.

But I booted my other server using the questionable flash drive, and it pulls the IP assigned to it, but it's GUI isn't accessible.

To add to the symptoms list: can't access GUI; no SMB shares are visible in Windows; Plex, Jellyfin, Vaultwarden, Home Assistant, etc are all inaccessible. The IP address doesn't respond to ping either, when installed on my main server, or my other server.

The only thing that I can say for sure is that I can access the Server itself when I connect a keyboard and monitor, but I don't know what if anything I can check from there. And others have said the diagnostics didn't indicate a problem, at least as far as the GUI issue is concerned.

For testings sake, on the new flash drive I went into /config and removed the contents of /wireguard, and then I changed the name of /plugins, and now it boots and the GUI is accessible.

I'm going to try this on the original flash drive, then re-add plugins, then try wireguard, and see what if anything seems to be the cause.

To eliminate plugins as a problem, you can boot in SAFE mode, which you already tried, but might still be useful in combination with other things.

Many files in config are simple text and can be edited by putting flash in PC.

There is a startup script, the go file, at config/go. Unless you edit it, the only thing it does is start the webUI (emhttp). And that is how it will be with a new install.

To eliminate Dockers/VMs as a problem, you can disable Docker (config/docker.cfg) and VM Manager (config/domains.cfg).

Arguably, the most important part of your configuration is your drive assignments. These are in config/super.dat and config/pools. If you have your drives you have your data, and your shares are just the combined top level folders of your drives.

Your share settings are in config/shares, each share has a corresponding .cfg file. The webUI settings made for a share are all stored in its .cfg file. If a share doesn't have a .cfg file, the share will still exist if its data exists, but it will have default settings. A share is not accessible on the network by default, but there are entries for that in its .cfg file.

Other .cfg files in config contain other settings from the webUI, often you can figure those out since they are text.

Tha sugegsts it's a config issue, you can try redoing the flash drive, backup the current one first and then recreate it using the USB tool and just restore the bare minimum, like the key, super.dat and the pools folder for the assignments, also copy the docker user templates folder (\config\plugins\dockerMan\templates-user), if all works you can then reconfigure the server or try restoring a few config files at a time from the backup to see if you can find the culprit.

Thank you for all the information and suggestions. I was able to get the system running with GUI by removing the files in the Wireguard folder, and the Tailscale plugin folder.

I am not patient enough, or brave enough, to try to add it back it, at least not the original files. I am going to install the Tailscale plugin again sometime soon, as well as add some WG tunnels in, but not until I get my new NIC installed in the server.

Thanks again