prune

You may find this one interesting : https://product.unraid.net/p/7-2-3-intel-arc-pro-b50-discrete-gpu-works-slower-throttles

Hello Amagician, I'm interested by your topic, cause I plan to add an Intel Arc GPU to my setup (to run LLMs). I actually transcode using the integrated UHD 750 gpu of my Core i9 CPU. (It works very well) Just a few comments as I can't really compare: You are using i915 driver that don't looks optimal with Arc cards, XE should be preferable, Unraid users are waiting for next Unraid 7.3 release which includes a new kernel with hopefully full support for Arc cards You didn't specified a temporary folder for transcoded files I saw a post on this forum saying that he had to use the :plexpass docker image instead of :latest to get hardware transcoding support (I really doubt here cause I'm not using this image and hw transcoding is working on my setup) You are talking about an Arc B card, support for it was just added in latest Plex release 1.43.1.10576 Keep us updated. Do you finally found a solution ?

Good to know, thanks Niklas. I see that there is even an rc19 release. But all the story should be pinned at the forum homepage, so that users stop upgrading until 6.12.4 is out and the one using 6.12.x can give the RC a try.

Well said, I fully agree. We are many Unraid users suffering a serious bug in this 6.12.x release, making our servers unusable. Please handle this case seriously, danioj has proposed several ways to handle and keep us informed.

I fully agree. Looks like this "fix" works for some people but not for others, it seems for me that it is not reliable and shouldn't be exposed as a solution. I, for example, have switched containers to a dedicated secondary nic with ipvlan with no luck. Also, being forced to use a secondary nic is not a solution, this is a workaround. In the past, I never had any problems with Docker networking on RedHat or Windows, and I always had only 1 nic. So Unraid will be the only platform in history where 2 nic's are mandatory in order to use Docker ?? Really ? 🤔

Hi, I too didn't had any issue with 6.11.5. A few days ago I uograded to 6.12.3, since then I experience server crashes every day. WebUI is inaccessible after few minutes following a restart. My favorite ssh command actually is /etc/rc.d/rc.docker stop So that I can get the webui back... and tune the configuration. Documentation and feedback from users talks about switching from macvlan to ipvlan but so far the situation is the same for me. Still digging a bit before switching back to 6.11.5. 🤷‍♂️

Well, I think I have the same problem has several people here, the problem is described here : So I followed the guidelines regarding docker and the network with version 6.12 : https://docs.unraid.net/unraid-os/release-notes/6.12.0/#call-traces-related-to-macvlan Even after switching Docker to ipvlan, I still have troubles. So I will try to follow this documentation as I got 2 network cards on my server :

Well finally, I loose access to the webui again, I really don"t know what's the matter. I've attached the log, if someone can dig into it and point me to any relevant info. That would be cool. big-diagnostics-20230726-2104.zip

Thank you very much jmztaylor for your answer. I discovered after reading your message that even SSH access was inaccessible this evening, whereas it was working in the morning. Had to restart the router, and switches, but that did not solve the problem, saw that the server was like down so I restarted it. And now I can access the webui and SSH. Now, due to the upgrade I think there is still something to figure out as the "Dashboard" doesn't displays any metrics for CPU, memory, network, air flow, parity, array, cache... and more surprisingly, the "Main" page is empty, I only see tables with empty lines for Array, Pool, Boot and Unassigned devices 🤷‍♀️ What is even more surprising is that all my docker containers and VM's are working like if nothing happened. I think I still have plenty of tuning/diagnosis/reading to do on my side before bothering you guys with my problems. And I would open a new thread if I need more help. Anyway, it's good to know that there is a place where when you ask, someone will answer 🙏

Hello guys, I'm in trouble since a few days, I'm unable to reach the webui. My web browser Firefox displays a PR_CONNECT_RESET_ERROR or a PR_END_OF_FILE_ERROR error message when I try to access my server at https://MyServerName.local I also tried with the IP address https://192.168.4.98. No way. I think this behaviors follows a "docker network prune" command launched by myself (I was just trying to fix a strange behavior with my docker network). Following the prune, a message told me that network wg0 have been removed.... Since then, I rebooted several times, network wg0 is back, inspected the log files, tried to stop and restart nginx, and even upgraded to version 6.12.3 in the hope that it could fix things. But no. Sincerely, I don't know where to start, so could you point me in a way so that I can diagnose what's the matter. For example, how would I get logs from the webui webserver ? So far I have been unable to get any relevant message from syslog. Best regards, Mathieu

+1 I would like to be able to see somewhere all failed login attempts on Unraid services

+1 I want this too !

Just to bump this thread. I too would find very interesting to have some visibility about where this product is going.

Hi, For sure something is wrong. Looks like you choose the wrong sensor.

Hi, I'm new to Unraid so it is hard for me to help you but: I never had such error messages so your problem is not the same as mine, in fact I had no error but only CPU temperature. Looking at the error message, it looks like AutoFan plugin search for an nct6687 driver in folder /sys/devices/platform and for a file in that folder does not exists. Even with the Nuvoton NCT6687 driver application plugin installed, I don't have this folder and files on my installation. You could try to force installation of the driver by issuing the following commands in a terminal: depmod -a modprobe nct6687 Note that I'm not an expert and you should better get in touch with the plugin developper to get an advice.

Hi, Looks like there is a missing step and a missing field in the Unraid UI in order to fully configure the Telegram agent. I think we need to put the telegram chatid somewhere. When you read the /boot/config/plugins/dynamix/notifications/agents/Telegram.sh script, you see that it references a /boot/config/plugins/dynamix/telegram/chatid file containing the chatid number. But in the Unraid telegram agent configuration UI, there is no chatid field (I'm using 6.9.2) In order for the Telegram agent to work, I had to create the file with my bot chatid: echo "MyChatIDnumber" > /boot/config/plugins/dynamix/telegram/chatid (Just replace MyChatIDnumber by your chat ID number including the minus (-) sign if any) Then clicked the test button in the Telegram agent configuration page and it worked. How to retrieve chatid : How to obtain Telegram chat_id for a specific user? - Stack Overflow How to retrieve token : https://core.telegram.org/bots#generating-an-authorization-token

Hi, I had the same problem with an Asrock H570M-itx motherboard, got only CPU temperature but no fan / fan speed. In order for the Dynamix system temperature and fan control plugin to see the fans, I had to modify syslinux.cfg file and add a "acpi_enforce_resources=lax" kernel option on boot. So edit the file with nano editor, using the following command line: nano /boot/syslinux/syslinux.cfg Here is an extract where you will make the modification: label Unraid OS menu default kernel /bzimage append initrd=/bzroot Change the content to : label Unraid OS menu default kernel /bzimage append initrd=/bzroot acpi_enforce_resources=lax Exit and save the modification by using Ctrl + X keys. And reboot in order for this kernel option to be taken into account. I also installed the plugin named "Nuvoton NCT6687 Driver" from the Community Applications, but I don't know if it is mandatory, because my chip is detected as NCT6798. With this modifications, all my chassis and CPU fans are detected and I can control them using the Fan Auto Control plugin.