Stable Releases

Note, at time of writing, 6.10.3 is current and this happened on all 6.10.X releases thus far. I previously had crashing issues at least once every 72 hours with 6.9 using docker containers w/ static IPs.   To combat that at the time I created a docker-specific VLAN for the few containers that needed static IPs. Fast forward to 6.10.X and the release of the ipvlan mode which was supposed to help people with this issue.   I went ahead and removed my VLAN and swapped

Unfortunately, with 6.12 it is not possible anymore to replace a drive while keeping parity intact. The corresponding script or the manual command "dd bs=1M if=/dev/zero of=/dev/mdX status=progress" does nothing in regard of writing zeros to the drive. The script doesn't recognize the error and "thinks" it has finished correctly.   corresponding docs entry: https://docs.unraid.net/de/legacy/FAQ/shrink-array#the-clear-drive-then-remove-drive-method   Log of script: *** Clea

I upgraded from 6.11.5 to 6.12.0 and immediately noticed that a Windows 11 VM I have no longer works - it sits there with the unraid logo constantly and doesn't seem to boot at all.   I tried recreating the VM from scratch as it's a pretty simple VM and even that failed. Checked /config/vfio-pci.cfg and its empty, as is /var/log/vfio-pci-errors   Rolled back to 6.11.5 and the VM started working again. Diagnostics attached. ubernas-diagnostics-20230616-1331.zi

I attempted to upgrade to 6.12.0 on two servers.  One worked great without any problems as far as I can tell, the other will not boot into 6.12.0.  I manually rolled it back to 6.11.5 and all was well.   What I have seen with the problematic server. The boot process starts and just seems to end at some point and video output is lost It never boots completely and I cannot access the GUI (with one exception - I got diagnostics then) The one time I was able to get in

On the dashboard, the "system information" main panel you can not remove it from being shown. The info it shows is somewhat redundant as its shown in the top banner, but more importantly there is icons to stop/restart/shutdown array that one could easily accidentally click.   Please allow users to remove this panel   Also, I'm shocked you did not put the 'mover' button on this or the cache panel..   feature request?  

I have a NVIDIA GeForce GT 710, which is not supported in the 500 series drivers, so I'm forced to use the legacy driver (v470.182.03). This is set with the Nvidia Driver plugin.   While on 6.11.5 and doing the update os to 6.12, it fires off the plugin helper which upgrades nvidia driver to latest v535.54.03. This was suboptimal as this driver does not support my video card. It would be nice if the plugin helper would ignore upgrading if the user is using the legacy 470.x dr

The fisrt picture is docker menu tab and only have one docker info, but in the second picture at dashboard can be seen that is missing many containers

On 6.12 I have to configure my network adapters in Settings/NetworkSettings.   I have added the wg2 and the tun5 interface of my wireguard and openvpn instance to the extra interfaces. I can reach the WebUI via both of them, including SSH. In addition, I also have added my zerotier interface. The ZT instance is running inside of docker, whereas OpenVPN and wireguard run on bare metal.   However, I cannot access any SMB shares.   If I inspect the smb.conf a

The VM fails to boot exactly after the graceful reboot to update to 6.12! Can it be related to KVM version change?   A Windows based VM works fine. HELP!?  

Following Spaceinvader One's absolutely fantastic video of the upcoming (at the time) release of 6.12 I added my existing ZFS pool into the pools after upgrading.   After adding it and rebooting - one of the three datasets did not add in properly. I noticed that shares were created to manage the data sets. The one that didn't add properly "rsnapshot" came in empty. I checked the mountpoints both where I had it mounted and where the share put it and they were both empty so I removed the

On 6.11.5 I had a pool created with steini84's plugin with the following configuration: root@Tower:~# zpool list -v NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT tank 43.6T 31.4T 12.3T - - 49% 71% 1.01x ONLINE - raidz1-0 43.6T 31.4T 12.3T - - 49% 71.9% - ONLINE ata-ST12000NM001G-2MV103_ZL274PFX 10.

Screenshot attached. Same issue in all themes. Downgrading to 6.11.5 restores. Upgraded, downgraded and upgraded again to check while trying to get the Tailscale interface setup. Changing the order in 11.5 and then upgrading again held.

Posting this on behalf of multiple Tailscale plugin users. Related Discussions https://forums.unraid.net/topic/136889-plugin-tailscale/page/7/#comment-1272479 (contains diagnostics) https://github.com/dkaser/unraid-tailscale/issues/57 (further discussion, includes the relevant configuration file) Symptoms SSH will fail to start when: Normal/Wireguard interfaces only have IPv4 addresses, and Extra (e.g., tailscale) interfaces have IPv4 + IPv6 addresses.

I had 3 VMs, and the are not visible anymore. The vdisks in the domains shade do still exist. The zfs mount -a did not work. I also tried disabling and enabling VMs in the VM Manager. I rebooted multiple times.  downgraded to  6.12.0-RC8 but there they were gone to. I upgraded again, still gone.   I do not know exactly yet were to look for more details.   The diagnostics file is generated right after the first upgrade when I first discovered.   Any idea's? 

Setting display preferences to Black or White allows the change which is preserved when changing back to Azure or Grey theme. LMK if you want a diagnostics pull.

Never quite understand how the maths works on this... What does RAM, FLASH, LOG & DOCKER denote?   Please can we have it simplified for 6.12?   Total Memory - Current Usage % and in GB   Then breakdown in % & GB - Unraid System - Docker - VMs - ETC

Unraid 6.11, but as reported in the post mentioned below, it looks it's happening in 6.12 RC too.   The /run partition is full. The reason is that Plex container (using --runtime=nvidia) is logging every 5 minutes the following in /run/docker/containerd/daemon/io.containerd.runtime.v2.task/moby/<identifier>/log.json:   {"level":"info","msg":"Running with config:\n{\n \"AcceptEnvvarUnprivileged\": true,\n \"NVIDIAContainerCLIConfig\": {\n \"Root\": \"\"\n },\n \"NVI

After having a few usb sticks failing I have a fancy industrial one but is only 2GB.    Unraid on my stick is using 1.6GB so no data left to build a zipped version of the stick to upload... for the automatic cloud backups.   Please can we add an option to allow the backup to be complied on the CACHE drive say appdata or somewhere and then upload it.   Also...    https://docs.unraid.net/unraid-os/manual/getting-started

查询日志后出现这个问题，不知道机器死机是否因为这个，因为安装的虚拟机作为家里的主路由，机器频繁死机，导致全家断网， 请问有什么解决的办法么。请大神们给予帮助，不胜感激，

To replicate the bug, if you have two bridges in your unRAID the Docker Network Type dropdown will only offer up the first one like so:     When instead it should offer all the bridges. In my system I have both br0 and br1. I'm able to select either for Virtual Machines but not with Docker. This is I believe how it should look:     Thank you.

if I use SMB Security Settings -》 Export: -》 Yes(hidden), then I can not login the SMB folder after using account name and password from WIN10. But if I use SMB Security Settings -》 Export: -》 Yes, without hidden, then I can login the SMB folder after using account name and password. So I can not to visit hidden folder with account name and password? or neet some other settings? tower-diagnostics-20230605-2130.zip

First, Unraid VM crashed. When operate VM actions, show "Hook script execution failed: internal error: Child process (LC_ALL=C PATH=/bin:/sbin:/usr/bin:/usr/sbin HOME=/ /etc/libvirt/hooks/qemu Kation-VM prepare begin -) unexpected exit status 126: libvirt: error : cannot execute binary /etc/libvirt/hooks/qemu: Input/output error" See 1535 diagnostics. Then reboot unraid. VM worked. But Docker start failed. Run ./rc.docker start then ./rc.docker status show stopped S

Using docker compose it is possible to create a stack that contains more then one network, however the Docker Containers screen will only show one at a time   To reproduce: Run Unraid 6.11.5 Install Docker Compose.Manager from apps Create a new stack with the setting below (note: you will need to update the ip4_address to match your network) Save and Run   version: "3" services: nginx: container_name: Nginx-Proxy-Manager-Official image:

At the moment, I'm not too happy with UnRAID; this is my first time using the software, and the lack of Web UI when the software boots is totally unacceptable.    The only way I've managed to get this working is by connecting a keyboard and screen to the server, logging into the terminal and then typing the following commands   /etc/rc.d/rc.nginx reset /etc/rc.d/rc.nginx reload   At this point, the Web UI sometimes starts working. But this sometimes must be don

I have been using hddtemp in a docker to read drive temperatures and have noticed that when I spin down drives and then let the hddtemp try to read it's temperature it knocks a drive out of the array (it will show up in assigned devices). I have replicated it this a few times, and to stop it from occuring now I check if the drive is on standby (with smartmontools) before using hddtemp to read the temp. Drive is a 10TB Seagate, but note it won't happen with all Seagate drives, just appears t