Bob-omb

Thanks @JorgeB As you said, cache and delta failed mounting earlier today. Since the cache disk holds the docker image the system could not find it and must have made a new docker image. But where? For future readers here is the process I did: I deleted the docker image at /mnt/user/system/docker/docker.img I started the array. The system could not mount disk 1 and downloads so I tried to start the array again. This time all disks mounted successfully. Now I see that the locations of appdata-share, the system-share and the vm-share is suddenly on a different disk, probably because the cache disk failed to mount previously. I changed the location of these shares from downloads to cache in the share settings. I see that I still have a 161 GB docker.img file in /mnt/downloads/system/docker). So I make a backup of it —just in case. I also have a 161 GB docker.img file in /mnt/cache/system/docker/ So I make a backup of it —just in case. Then I add /mnt/cache/system/docker/docker.img to the "Docker vDisk location". I delete the appdata, system and vm in the downloads pool. I enabled the Docker service again. I go to the Docker-tab and all my dockers are back again 😃👍 I really wish Unraid would not automatically make system shares at random disks whenever it cannot find the cache.

This is the second time all my docker apps have disappeared in the last year. I was in the process of changing SMB settings in Unraid (enabling "hide "dot" files") so I was forced to restart the array. The first time I tried to restart the array I got the usual error saying "Unmountable: Unsupported or no file system" on some of the drives. This has occurred before and I have never found a solution to it other than attempting to restart the array a couple of times. So I did a restart of the array through the GUI. After restarting the array all my docker apps were gone: WTF Unraid... This is the second time this happened. 😵 Last time was after I upgraded from 7.0.1 to 7.0.2. I had to restore everything and spend two days on configuring it correctly. I have attached the diagnostics. Then I try to restart my server with the reboot option. I wait for 8 minutes for the server to restart. Nothing happens. When I try to SSH I can't get access. KVM is a black screen but the power icon indicates the server is on. I do an unclean power down on the KVM. I then start the server again in Unraid OS mode. I start the array with no issues (for once). I go to the docker tab and the same error "No docker containers installed". I know I can re-install all the docker apps but to what end? I did that previously and here we are again. This is probably going to continue happening. What should I do? nova-diagnostics-20250607-0933.zip

Same issue in Unraid 7.0.1. upgrading to 7.1.2 corrupted the docker image and all my docker apps were gone. I tried to run /etc/rc.d/rc.nginx restart I get: rc.nginx: Restarting Nginx server daemon... rc.nginx: Checking configuration for correct syntax and then trying to open files referenced in configuration... rc.nginx: Stopping Nginx server daemon gracefully...Seems stuck. Meanwhile the system log shows: May 29 16:04:33 nova rc.nginx: Restarting Nginx server daemon... May 29 16:04:33 nova rc.nginx: Checking configuration for correct syntax and then trying to open files referenced in configuration... May 29 16:04:33 nova rc.nginx: /usr/sbin/nginx -t -c /etc/nginx/nginx.conf May 29 16:04:33 nova rc.nginx: Stopping Nginx server daemon gracefully... May 29 16:04:52 nova rc.nginx: Stopping Nginx server daemon forcibly... May 29 16:04:52 nova rc.nginx: Nginx server daemon... Stopped. May 29 16:04:57 nova rc.nginx: Starting Nginx server daemon... May 29 16:04:58 nova rc.nginx: Nginx server daemon... Started.And then nothing happens... it's just stuck on the operation...

What makes you say that?

Thanks but nothing happens when I click on the Unit ID dropdown...

I have had this problem for months in Unraid 7 and the last time I asked for help here I was told it was a hardware problem because I have an Intel 14900K. Might be, but it felt like a defensive conclusion. I ran a memtest which came out negative for errors. The issue occurs from time to time and I have a feeling this has nothing to do with hardware since I have no other issues. Everything works fine except I suddenly can't login anymore. For example all docker containers works while this issue persists. I tried to run /etc/rc.d/rc.nginx restart which triggers rc.nginx: Stopping Nginx server daemon gracefully... but it get's stuck there and nothing happens. I have a feeling this has to do with nginx. I really love Unraid but this issue is so frustrating.

OK doesn't work for me. But then again my whole server went FUBAR once i upgraded to 7.1.2.

Is there anything new in Unraid 7.1.2 which changes how container networks work? I am used to route containers through another container with VPN. I usually add --net=container:binhex-qbittorrentvpn to the extra parameters of the docker app i want to route through VPN. However this does not seem to work after upgrading...

Help me setup Baserow with Tailscale in Unraid Can I get some help with setting up Baserow for Tailscale please? I have been at this for days 🫣😢 I can successfully turn on Tailscale with SSH in the Baserow docker configuration in Unraid like this: Everything seems to work fine... The app is authenticated and routed through the Tailnet... But when I try to access the Tailscale address for the app I get this message: It seems like something is not properly handled within the application settings. These are my docker settings: Am I doing something wrong here? Update: If I change the BASEROW_PUBLIC_URL to https://baserow.MY_TAILSCALE_DNS.net/ I can successfully access Baserow through the Tailscale URL, but not using the local URL of http://192.168.1.110:5555

Is this plugin deprecated? I am asking because it is archived on Github: https://github.com/dlandon/recycle.bin

Sure I can run a memtest. Is it possible to do within Unraid now? I remember reading something about it but I can't find it again. I thought segfaults was related memory, not the CPU. The CPU was timely patched with the microcode updates from Intel and it has been verified without errors with the "Intel Processor Diagnostic Tool". Although it might have been damaged I am not sure how else I am supposed to check it.

The web interface is inaccessible. I can access the server with SSH. Trying to shut down the system with (shutdown -h now) but it never does. The system is going down for system halt NOW! sys log says: Mar 10 04:29:36 nova kernel: ata7.00: failed command: READ FPDMA QUEUED Mar 10 04:29:36 nova kernel: ata7.00: cmd 60/40:a0:c0:fb:1f/40:00:e4:00:00/40 tag 20 ncq dma 688128 in Mar 10 04:29:36 nova kernel: res 43/84:f0:0f:01:20/00:3a:e4:00:00/40 Emask 0x412 (ATA bus error) <F> Mar 10 04:29:36 nova kernel: ata7.00: status: { DRDY SENSE ERR } Mar 10 04:29:36 nova kernel: ata7.00: error: { ICRC ABRT } Mar 10 04:29:36 nova kernel: ata7: hard resetting link Mar 10 04:29:37 nova kernel: ata7: SATA link up 6.0 Gbps (SStatus 133 SControl 300) Mar 10 04:29:37 nova kernel: ata7.00: configured for UDMA/133 Mar 10 04:29:37 nova kernel: ata7: EH complete Mar 10 04:35:58 nova emhttpd: spinning down /dev/sde

All my 12 drives are encrypted. The only one not encrypted is the Samsung SSD 850 EVO 1TB and that one is not part of an array. It's just an unassigned device. Are you seeing any other drives not encrypted?

I see. I am completely open to the suggestion that I might have bad hardware and I will happily look into it. It just seems like a hasty and convenient conclusion, but what do I know. I have used the Intel Processor Diagnostic Tool to check the CPU for errors. It did not find any. I also ran a memtest a while back without any errors. Perhaps you have some experience on where I should start looking? You seem kind of frustrated btw. I really appreciate the help, just making sure I fully get it.

OK. I guess I will have to narrow it down then. What do you mean by "detecting a wrong key"? If I had entered the wrong key it would not have unlocked the other drives, right? Sorry if I am asking too much, I am just curious about how this works. Not trying to argue or anything.

You mean the encryption key? If I enter the wrong encryption key I should get an error message, no? This doesn't make sense because all my drives are encrypted so why are some available when others are not? I restarted the server and everything seems fine now, but I have had this issue before so I wanted to report it either way. The 14900K is microcode patched by the way. It is also validated for any errors so it should theoretically not be an issue.

Just updated from 7.0.0 to 7.0.1. Rebooted, now it says four of the disks cannot be mounted with the error "Unmountable: wrong file system". What should I do? I have attached the diagnostics.

And... It's back! 😊👍 Fully restored using the config folder. Thanks a whole bunch @JorgeB and @trurl. Without your help I would have given up or bought a new computer. Lime should probably mention this issue somewhere though. I am not exactly using an obscure NIC or anything. Although this was technically a kernel issue most users are not going to care. They just want to plug and pay. I am happy it is solved now after two full days of troubleshooting.

Thanks so much @JorgeB Unfortunately because of all this mess I decided to make a new flash drive. I have not lost my config because I have a backup of the old flash drive. I really want to avoid setting up the config (users, passwords, shares, user scripts etc...) all over again. How can I restore the old flash drive and apply this workaround? I have never restored an Unraid flash disk. Also, how should I go about applying this workaround? Should I just copy the "70-persistent-net.rules" and "cdc_ether.conf" to their respective folder?

Since I had no luck so far I entered this in the Terminal: echo "blacklist cdc_ether" > /boot/config/modprobe.d/cdc_ether.conf I restarted. I still get the "Interface Ethernet Port 1 is down. Check Cable." message, but at least now I can access the web interface with hostname or IP. Can I ask some questions out of curiosity? Why is this not patched by Lime if it is a known bug? I spent two days on this. If it had not been for @JorgeB I would have given up on Unraid. Where is this kernel bug tracked? Why is it still saying "Interface Ethernet Port 1 is down. Check Cable." Is it possible to add a configuration change to the flash disk instead of having to type these commands after boot? Btw here is the new diagnostics file after I added the blacklist command.

I installed Windows and without drivers the network interface was not detected. After installing the "DRV_LAN_Intel_I226_UWD_TP_W10_64_V11443_20240710R.zip" driver it was recognized. This leads me to believe that the NIC is buggy after some BIOS update or something. I update the BIOS from 4001 to 4001. In Unraid GUI I used Terminal to enter: mkdir -p /boot/config/udev Then: echo 'SUBSYSTEM=="net", DRIVER=="cdc_ether", NAME:="usb0"' >>/boot/config/udev/70-persistent-net.rules I shutdown Unraid. Started it again. Still no IP for the network. Did a diagnostics. Attached it here

I see. But I am not using the IPMI card for Unraid. Actually the IPMI card works fine, but the NIC on the MB doesn't. Neither of the two LAN pors on the backplate of the MB works it seems. Also, this has been working for a year. I even ran Unraid 7 since it was released. So I am kind of confused 😕 I am installing Windows right now, but I will test your solution after that.

Thanks. I will try that, but what do you mean by issue with the NIC? Issue? Like in a known bug?

Btw the MB is an Asus Pro WS W680-ACE IPMI