Admin account unable to reach server

I have seen some similar problems related to managed accounts however, I am having the opposite problem and am unable to address the issue as I cannot work from the admin account in the plex GUI.

I had not made any changes in a long time prior to this issue. This just happened quite literally overnight. I noticed on my TV first, and it had happened across all devices. On my Admin Account, my server just disappeared, my cousin’s remote plex server is showing up as though it is my local plex main server. If I switch over to my managed account, my local server appears normally. However, I cannot access server settings in the GUI as it is a low-level managed account for guest access.

Having no access to any server settings in the GUI to make any changes limits where/how I can approach and troubleshoot this. I disabled VPN, port forwarding, and any routing-related settings. Took down my firewall, nothing is changing.

I tried changing the IP address on the Truenas jail. as in the past with a new IP this has initiated a kind of start-up wizard with the libraries, but not this time, it’s still defaulting to my cousin’s server.

It almost seems as though Plex at layer 7 believes that my cousin’s server is my local server. I’ve seen a few other folks with similar issues, mostly losing their servers on their managed accounts. But they have admin access to try fixes. Examples of folks I’ve found with loss on admin accounts have not posted their solutions to these problems.. I’ve seen one referring to DNS, but I don’t know to what capacity, so again, I don’t know how or if they resolved the issue.

Server Version#:11.3-RELEASE-p14
Player Version#: 4.145.1
<If providing server logs please do NOT turn on verbose logging, only debug logging should be enabled>

In Plex Web, navigate to Settings -> Plex Home. Does your admin user have a crown and the label “Home Admin?”

It does, I’ve only got 2 users on the account, the admin account, which I normally watch on, and a separate guest account.

I’ve actually moved a bit forward. I stumbled upon some interesting steps. I guess for de-claiming and re-claiming, which has put me in an interesting place right now.

I removed the preferences,xml file, and removed the PlexOnlineHome, PlexOnlineMail, PlexOnlineToken, and PlexOnlineUsername values. And replaced the modified copy. Of course, I saved the one I removed from the server.

This has now reopened the Admin user. I haven’t had much time to work on this. The problem started a couple of weeks back, and I’ve had to live with it until now. Too much in the works. I’ve been able to watch shows, but I found some time today.

I have a new problem right now, and I may need to put the un-modified preferences.xml file back. Because now it is not letting me reclaim the server… I’m not going to have much more time today I can spare a little more time tomorrow to investigate the new changes. I was briefly reading that I may be able to insert a new token into the file, but I need to figure wrap my mind around where that token is going to come from exactly.

It took some time, but for anyone else with similar issues. I feel inclined to share how I finally solved this. I tried countless times to solve this issue with different combinations of preferences.xml settings. I was about ready to break down the server and migrate to a fresh install when I tried a few more things.

I was able to discover that the Plex Media Server iocage jail, in Truenas was connecting to the Internet but unable to resolve DNS. I began to wonder if there were problems with perhaps Unbound or Pi-hole interfering. I made allowances for plex.direct in the whitelists and domains for OPNsense, and PI-hole respectively from finding others whom have had similar issues with these.

I decided to look closer ensure Proxmox was resolving correctly, ping 8.8.8.8 and google.com for tests, this revealed that Truenas itself was resolving. Then I went down a level, I came into the SSH session within the jail itself, I was able to ping out 8.8.8.8 however, I couldn’t resolve domains like google.com, plex.tv, etc.

I was a little stumped, continued to check logs and find any clues that something was blocking it. I decided to try ChatGPT. Going over troubleshooting steps with the LLM led me to the same conclusions, until the resolv.conf file came up. This was the issue.

It has somehow been wiped clean, I now understand this is not unheard of in BSD settings with plugins/jails. After testing /etc/resolv.conf with 8.8.8.8 and 1.1.1.1. It began resolving in the CLI. I reconfigured resolv.conf one last time with the correct DNS server. And it is now resolving successfully.

Going into the :32400/web/index.html#!/ and finally, after a good long time living with the problem, the server claimed successfully.

I’ve seen a lot of solutions revolving around trying to reclaim, by forcing a claim token into the preferences.xml file. The times I tried this, plugging a token in would lock me out afterwards. I tried no preferences.xml, replacing with an older one. Nothing worked, and I’ve not seen anything with this resolv.conf file.

1 Like

I’ve been having this problem for months, and it’s driving me crazy. Unfortunately I don’t understand this solution - I’m handy with software, but I’m not a computer scientist. Can someone explain this in a step-by-step way so those of us without a depth of knowledge can escape this hell?

Hi I’m sorry that my explanation wasn’t very clear. I also noticed that I started with one problem. Then encountered a new one which my solution mainly covers. However, the second problem may have been caused by my first solution. Or it could have been the root cause the whole time. That issue was the DNS resolution. I would note that I’m virtualizing Truenas and running Plex out of a jail in that instance, my case may be a little unique. But could happen many different ways.

The first problem was that I locked myself out and Plex was acting glichy I solved this issue by modifying the prefrences.xml config file, in the root directory (main folder) of the Plex Media Server. Refer to this from the plex support page. In doing this, I highly recommend you keep a save copy of your original before you go rooting around in config files.

This gave me access to the server via “ipaddres:32400/web/…” browser address, but I was still unable to “claim” the server again. This is another issue I tried a number of fixes with, many of them suggesting you can use code generated by going to https://account.plex.tv/en/claim. And copying this into the preferences.xml>token “” field. This is an example of ways to modify with token. However this did not work because my container was not resolving DNS. Which just for anyone who may read this, is how computers translate IPs into human readable addresses. IF they cannot do this they will not be able to connect and do services. Alot of solutions on googling this revolve around modifying this file in one way or another.

once in command line of the plexmediaserver, I suggest googling your specific setup so you don’t connect to the wrong thing by accident, or fail to connect. Generally it would be something like:
Take caution when entering to the root level, changes will be permanent
root@PMSipaddress maybe like: root@192.168.1.1 which ever it’s address is.
- sign in using password
- you’re now remote access into your server via SSH
- try to do pings from here to confirm connection
ping 8.8.8.8
ping google.com (or any other web address)
- check your resolv.conf file to confirm missing nameserver.
cat /etc/resolv.conf
- If you only see one line “search local domain”
- modify could be:
ee /etc/resolv.conf
- this will bring up a simply text editor, add the
needed lines, use escape button to bring up menu,
and 1, 1 to save exit
**** This could work for you may need different commands*** These are basic linux commands should work in something like powershell but not guaranteed*****

So how did I confirm DNS was the issue? I Had connection locally, it was clear that I didn’t have proper internet connection, new videos we not getting their meta data, for example. To troubleshoot and find out why. I needed to go command line of “each level” Proxmox, Truenas, then the Plex container and use a command “ping”. SSH would be the easier way to do this, but in my case for example I could also just use the Truenas webgui and go straight to a shell from there. Were ever your Plex Media Server is that is were you need to go.
Each setup will be different, maybe you’re on a bare-metal machine, and not a virtualized environment like mine. I would suggest taking steps to isolate the issue like I did then you will truly know what the problem is. Check firewall rules, connections, everything that is part of the chain.

Using Ping and confirming DNS? I know my computer and other devices resolve because I’m on the internet right now. On containers and other machines, vms you may need to use command line because that’s the only way to speak to them. I simply used the ping command on the address 8.8.8.8 and “ping google.com” (which is a public Google DNS server.) I did this from the command line of each instance down, So Proxmox could ping both IP and google.com. Then Truenas it could do the same, then I got to the container, it could only ping 8.8.8.8 not google.com. This made the issue very clear.

I found, with the help of ChatGPT, a file called resolve.conf and this step is similar to the above steps in the links with modifying the prefrences.xml file. It’s just a different file; /etc/resolv.conf, with differen’t layout. It is were some servers/clients store their DNS configuration, it holds the names/addresses of DNS servers to use. I it did not have any address to DNS servers. I simply needed to add the line(s) nameserver . Example using google’s like many folks do. line would be: namesever 8.8.8.8 .This is the same address we used to test ping. If you use a different address just make sure it’s named in this file. Simple google search will explain further about DNS, here is a wiki page about resolve.conf.

But in a nutshell, my Plex Media Server could not connect to the internet. Therefore, it couldn’t be claimed because the service couldn’t reach it. This problem was caused by the resolve.conf file being corrupted, wiped whatever the reason, it was blank. The file can be volatile. Simply adding the address of a DNS server for it to use fixed this issue.

I don’t mean to discourage, but to do this, may not be the easiest for anyone with no ability to use the command line; there may be a learning curve required because this will likely require command line knowledge. I did use ChatGPT to help me locate the specific file, but commands like ping and being able to address my servers via the terminal were a must for me to implement this. There are various resources around for SSH, and ping there will be slightly different ways depending on the setup, so I don’t just want to blindly say this is the way. It could be possible to manually locate this file and edit with a text editor. Again, set-ups may or may not allow this. And if you change the format, you will nerf the file.

If there is something specific I can be more clear on, I would like to help, but just pinging the two addresses one numeric and one worded, clearly showed the problem. Then, just locating and modifying the file resolv.conf located in the /etc folder of the plex server install, this fixed the problem at the end.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.