I am new to TrueNAS and have been loving it so far, but just ran into an issue where I can’t access the web UI anymore. The system is still up and running as I can access my apps via IP:Port and can still access my SMB shares over the network… the TrueNAS UI just won’t load. Typically, I can just access it via either truenas.local or my 192.168.50.50, but neither work anymore. I just get nothing, it never loads.
When this happened the other day I rebooted the system directly with a monitor and keyboard plugged in and then it was accessable again. But the next day I couldn’t access it again. I randomly rebooted my router and then I was able to access it again… but again today I’m getting the same issue.
Just some extra context, I’m running the latest stable version of TrueNAS Scale. This is all just over my local network. TrueNAS has a DHCP reservation on my router so the IP never changes. TrueNAS had been running perfectly for about a month and I’d never had an issue.
I guess start by listing your detailed setup. Expand the TrueNAS Scale section under Protopia’s post to get an idea of what he provides.
Do you have a keyboard, mouse and monitor always attached to your TrueNAS or do you have SSH set up so you can access the TrueNAS console when the GUI isn’t responding?
Your description is a bit confusing as you can reach apps and SMB but not the GUI. Rebooting TrueNAS fixed it but you also fixed it with a router reboot. A router reboot shouldn’t fix the GUI if it hung or crashed but might help if you have a networking setup issue or conflict.
What is your network setup? Do you have only one NIC or multiple networks? Bridging involved, etc?
Is the system ‘broken’ right now? It’s helpful to diag when it is.
Thanks for the reply. I agree, it’s confusing to me as well (Apps/SMB working, but not web GUI).
Here is more info about my system:
Lenovo ThinkCentre M93
CPU: Intel Core i7-4770
RAM: 16GB DDR3-1600
SSD: System Drive Orico 128GB, Apps drive SanDisk 240GB
HDD: 2 x 4TB HGST Ultrastar (mirrored vdev)
TrueNAS Scale 25.04.0
Apps: immich, FileBrowser, Tailscale
Network Setup (1gbps) all ethernet connected:
ISP Gateway — ASUS Router (ASUSWRT) — unmanaged gigabit switch — TrueNAS (1 NIC)
I don’t usually have a monitor/keyboard hooked up, but do right now. I don’t have SSH enabled.
I just rebooted the system again just now (via monitor/keyboard) and can now get back into the Web GUI just fine. (The router reboot fix yesterday could have just been a coincidence).
Ok after doing some more thinking and poking around, here are a couple other pieces of info…
First, I received a message on June 22 saying…
“Boot pool status is ONLINE: One or more devices has experienced an error resulting in data corruption. Applications may be affected…”
I couldn’t really figure out what the issue was and the system seemed to still be working. I rebooted it a few days later and it wouldn’t boot… I was getting some kind of message that it failed to mount the file system. The only thing I could do was reinstall TrueNAS and restore from a config backup (which was super easy), everything was back up and running!
I just went to look at reports to see if that gave any insight and noticed that for the past month my free memory had been hovering around 5gb (see screenshot below)… You can then see between June 22 and June 29 my system was offline for a bit which is when it wouldn’t boot after getting the above error. Since then it seems like my system memory free continues to decline until there is only about 1gb left. I’m wondering if this is when the web GUI becomes unavailable. You can see at the very end of the screenshot my free memory was really low, then after a reboot it was back up to the top.
Check for App updates. There may be an update to fix a memory issue. You can also check you App configuration for RAM settings under each app. Keep an eye on the Memory and try to see if you can associate it to any activity. It might be backups, Immich scanning or doing image things? If you get SSH set up or just try logging in at the console when you have problems, you can try looking at TOP and HTOP to see what is going on.
Thanks for the info, I’m also new to Linux so I didn’t know about TOP and HTOP, those look really helpful. So far I haven’t lost access to the GUI the rest of the day today after the reboot and the free memory has stayed relatively high. I’ll definitely try to check TOP and/or HTOP (directly from the machine via monitor/keyboard) if this happens again.
I used to have a DHCP setup similar to this, but it’s really better to assign a static IP to your server, especially if your router acts up and chokes on DHCP or DNS or whatever it’s doing to break things.
I recommend making the server static. It will solve a few problems. Also, if your hardware is dodgy, look into @joeschmuck’s report thread. Really cool script and it will send you 2 emails daily, checking on the health of your drives, so you can see failure coming a mile away and don’t get surprised by a failing disk.
Will the script be unknown territory to install and schedule in TrueScale? Of course, but it is so well documented and written that it is very, very easy to get started.
Thanks @afrosheen I had gone back and forth about setting a static IP or just a DHCP reservation and had read that it really didn’t matter either way and some people seemed to prefer DHCP reservations because it was easier to manage them all in one place so that’s why I landed on that.
I’ll give a Fixed IP on my TrueNAS box a try though, couldn’t hurt!
Also, I haven’t experienced the web GUI becoming inaccessible again since my reboot yesterday (24+ hours) so that’s good. I also realized my memory isn’t really getting all used up… I did more reading and learned about ZFS Cache intentionally using almost all unused memory until it’s needed. So I don’t think free memory was my issue.
Yeah free ram is considered wasteful in the Unix world, because it could be doing something. So that’s intentional, I would be more concerned if my server wasn’t eating all the ram.
You’re right about dhcp vs static. People have varying opinions on it, I was dhcp for ages until I started rebuilding my network and couldn’t always count on dhcp/dns. Static ip’s “just work” like hand written host files, in your case it sidesteps a sketchy router. We’ll get you on a homemade opnsense box next lol.
Thanks, totally makes sense about static IPs, definitely makes one less variable in the equation when troubleshooting things.
Haha, I’m definitely on a homelab journey! I switched to Home Assistant running bare metal on a mini pc a while back, have another box where I’m experimenting with/learning proxmox, and then this TrueNAS machine for backups/media storage. You’re right, i’m sure it’s only a matter of time before jump into the router/firewall world as well!
I haven’t experienced this issue again in about a day and a half, still not sure what it was, could have been some fluke thing. I’m going to call this resolved even though we never nailed down the cause.
This happened to me as well, will try to change it to a static IP. The containers & VM still work but can’t access the ui. I’m using a first generation Ryzen so once in a while the whole server gets unresponsive (but that always included the apps and vm) but this is something new since I’ve upgraded to the latest version. The only other thing I’ve changed is I moved away from using USB sticks for the boot pool… I now have an optane m2… so maybe something in the latest version that causes this… or a new way of reacting to ryzen instability.
