Jaburges

Heads up just raised an issue with the dev: https://github.com/fmartinou/whats-up-docker/issues/324https://github.com/fmartinou/whats-up-docker/issues/324 However its the UNRAID Docker. Some ENV VARs use `HA` as the {trigger_name} and some use `mqtt` They all need to be consistent per docs here: https://fmartinou.github.io/whats-up-docker/#/configuration/triggers/mqtt/https://fmartinou.github.io/whats-up-docker/#/configuration/triggers/mqtt/ `WUD_TRIGGER_MQTT_HA_HASS_PREFIX` `WUD_TRIGGER_MQTT_HA_HASS_ENABLED` `WUD_TRIGGER_MQTT_HA_URL` `WUD_TRIGGER_MQTT_HA_USER` `WUD_TRIGGER_MQTT_HA_PASSWORD` Where `HA` is the {trigger_name}. Can you update @AlexGreenUK? Also FYI @tk40 in case you havent fixed. Confirmed working nicely with the above changes

UGH - fixed it by simply restarting bluetooth on the host /etc/rc.d/rc.bluetooth start SO Bluetooth dongle is being passed through, and the integration works - range isnt great though I've tried increasing the range on the Lock side, but just ordered a USB extension cable

@L0rdRaidenI've updated the firmware files, and the boot config - so the firmware and config files load. Added the Nerd-pack to get BlueZ and i'm able to run `hciconfig` on the host AND the docker. I'm passing the DBus & the device through /dev/bus/usb/XXX/XXX the error I get now is: Failed to start Bluetooth: [org.freedesktop.DBus.Error.ServiceUnknown] The name org.bluez was not provided by any .service files

Hi there, on edge the Web UI link from Unraid pulls up http://IP:5572 which redirects to http://IP:5572/#/dashboard However - the Edge UI auth pop up continually requests auth. each time i try and do anything a new auth pop up. On Chrome however, this happened once, it took me to the Web UI (not the Chrome UI auth box) and i could sign in to the site. no more issues

Ah it seems the magic is with a CNAME not an A record. I used to use a shell script that allowed for A records, but this CNAME set up with docker seems easier

are people still able to use this?? It seems Godaddy is now forcing the use of an IP address x.x.x.x in the 'points to' for the A record 😞

Hey folks, Not really sure what happened - all of a sudden getting errors: Get Docker Details for ip: 192.168.x.x Failed Request failed with status code 503 Get Main Details for ip: 192.168.x.x Failed Request failed with status code 503 Get VM Details for ip: 192.168.x.x Failed Request failed with status code 503 Get Docker Details for ip: 192.168.x.x Failed Request failed with status code 503 Get Dashboard Details for ip: 192.168.x.x Failed with status code: [object Object] Tried adding privileged to the docker. Didn't work Tested I can access the docker.sock remotely ok (from node-red) Bridge network was working great until I did something - any thoughts

weird - no clue why its looking in sys/ what is the full error - what file is pointing to that?

you downloaded 5.10.1 as above? Have you tried the make command? does it fail on missing files? if so goto /usr/src/linux then run `find -name XXXXX` where XXXX is the file that is missing from the make.

Ok so got it working on 6.9.0-rc2 (Huge thanks to @pgbtech for the initial steps helping me solve some issues I had) Rough workflow is: 1. Download Coral gasket and apex sourcefiles 2. Download and extract Linux kernel 5.10.1 (6.9.0-rc2), and copy over Unraid patches/mods 3. Install dependencies and add symlinks 4. make gasket and apex 5. add to drivers and boot config 6. Enjoy I also enabled all of the DevTools plugin as well (make for example is needed) [Please note at time of writing, the symlinks didnt bring all the required files into the right location, and I didnt' know enough about 'WHY' they were separate. I did notice for example `types.h` was different in different locations, so took me some time to work out exactly which files to copy over] cd /root mkdir packages cd packages/ wget https://mirrors.edge.kernel.org/pub/linux/kernel/v5.x/linux-5.10.1.tar.gz mv linux-5.10.1.tar.gz linux-5.10.1.tgz cd /usr/src/ mv linux-5.10.1-Unraid/ linux-5.10.1-Unraid_orig tar zxvf ~/packages/linux-5.10.1.tgz ln -s linux-5.10.1 linux cp -rf linux-5.10.1-Unraid_orig/* linux cp -f linux-5.10.1-Unraid_orig/.config linux ln -sf /usr/src/linux/include/uapi/linux /usr/include/linux ln -sf /usr/src/linux/include/uapi/asm /usr/include/asm ln -sf /usr/src/linux/include/uapi/asm-generic /usr/include/asm-generic cd /root/ wget http://slackware.cs.utah.edu/pub/slackware/slackware64-14.2/slackware64/d/bison-3.0.4-x86_64-1.txz wget http://slackware.cs.utah.edu/pub/slackware/slackware64-14.2/slackware64/l/ncurses-5.9-x86_64-4.txz installpkg bison-3.0.4-x86_64-1.txz installpkg ncurses-5.9-x86_64-4.txz cd /usr/src/linux make modules_prepare Note at this point the make might fail due to missing files cp /usr/src/linux/arch/x86/include/asm/types.h /usr/include/asm/types.h cp /usr/src/linux/arch/x86/include/asm/posix_types.h /usr/include/asm/posix_types.h cp /usr/src/linux/arch/x86/include/asm/posix_types_64.h /usr/include/asm/posix_types_64.h cp /usr/src/linux/arch/x86/include/generated/uapi/asm/ioctls.h /usr/include/asm/ cp /usr/src/linux/arch/x86/include/generated/uapi/asm/ioctl.h /usr/include/asm/ cp /usr/src/linux/arch/x86/include/generated/uapi/asm/sockios.h /usr/include/asm/ cp /usr/src/linux/arch/x86/include/generated/uapi/asm/socket.h /usr/include/asm/ cp /usr/src/linux/include/linux/compiler_types.h /usr/include/asm/ Now try and make make modules_prepare export CONFIG_STAGING_APEX_DRIVER=m export CONFIG_STAGING_GASKET_FRAMEWORK=m make M=/root/apex/ insmod gasket.ko insmod apex.ko lsmod|grep apex mkdir /boot/config/drivers cp /root/apex/gasket.ko /boot/config/drivers/ cp /root/apex/apex.ko /boot/config/drivers/

little out of my depth but the top of the file in question: `errno.h` #include <uapi/linux/errno.h> there are a bunch of *.h (limits.h, errno.h) that look for `uapi/linux/file.h` Not sure if i'm doing something wrong, but the files are all over the place and thus not found by the make command. So either need editing, moving, or another symlink created I created another symlink ln -s /usr/src/linux/include/uapi /usr/include/uapi But i still get errors with other files not located where they are expected. @DiabloXP did you copy or edit limits.h types.h etc?

Thanks for sharing It seems the makefile is looking for a file uapi/linux/error.h /usr/src/linux/arch/x86/include/uapi only contains `asm` directory are there any other symlinks that you added that may resolve this?

i must be doing something wrong. Running Unraid 6.9.0-rc2 Bison 3.0.4 (as per guide above) Linux kernel (according to 6.9.0-rc2) is 5.10.1 getting an error on make modules_prepare SYNC include/config/auto.conf.cmd HOSTCC scripts/basic/fixdep HOSTCC scripts/kconfig/conf.o In file included from /usr/include/bits/local_lim.h:38, from /usr/include/bits/posix1_lim.h:161, from /usr/include/limits.h:183, from /usr/lib64/gcc/x86_64-slackware-linux/9.2.0/include-fixed/limits.h:194, from /usr/lib64/gcc/x86_64-slackware-linux/9.2.0/include-fixed/syslimits.h:7, from /usr/lib64/gcc/x86_64-slackware-linux/9.2.0/include-fixed/limits.h:34, from scripts/kconfig/conf.c:7: /usr/include/linux/limits.h:5:10: fatal error: uapi/linux/limits.h: No such file or directory 5 | #include <uapi/linux/limits.h> | ^~~~~~~~~~~~~~~~~~~~~ compilation terminated. make[2]: *** [scripts/Makefile.host:112: scripts/kconfig/conf.o] Error 1 make[1]: *** [Makefile:602: syncconfig] Error 2 Makefile:688: include/config/auto.conf.cmd: No such file or directory make: *** [Makefile:710: include/config/auto.conf.cmd] Error 2 Is this a dependancy thing? I've enabled all the Dev Pack, but not sure what else is needed from the NerdPack? Any thoughts?

Did you use the exact steps or edit for Unraid kernel 5.1.0 as you are on 6.9.x ?

OMFG - no I didn’t - this is gold! Thanks for sharing! I’ll give this a try for 6.9.x

Moved from the wrong area: Hi Folks, So this wasn't as easy as I thought as the host OS needs the drivers. According to the instructions: Get started with the M.2 or Mini PCIe Accelerator | Coral I need to install these packages: sudo apt-get install gasket-dkms libedgetpu1-std However apt-get isn't available on Unraid - bugger The odd thing is the device shows up in lspci - so i'm wondering if I can still use it? I'm not sure if I can find the device on the host and map it to the docker. Is there a way to run the drivers in another docker and use that on the host? or can someone help me figure out how to map it? I'm going to be using Frigate which is docker. mapping -v /dev:/dev the device shows up inside the docker, but it doesn’t appear usable, so I ‘think’ the issue is the drivers on the host - any suggestions? I’m wondering if I can take a backup/snapshot of my Unraid Host OS, and then try and install the packages? Although not sure how to install without a package manager?

@dlandonJust pointed a friend to the Unraid template and its missing the ENV variables for hook processing and Yolo etc. Is it still actually there and they can add it manually or will they need an older dockerfile?

I guess there a few options: 1. community maintain the zoneminder docker 2. Consider separating the ML components to a separate docker and Zoneminder docker 3. move to another NVR docker like Shinobi, motioneye, blue iris (Shinobi has tensorflow built in, but there is deepstack or frigate for object detection docker components). happy to help where I can

Sorry it’s been a burden, it is a huge complicated docker, and you’ve put a lot of great work into creating this. We and Zoneminder owe you a lot of thanks! in terms of the best way to approach Zoneminder moving forward - could a micro services approach work? It will add a little more overhead but would it make sense to have Zoneminder + SQL + Eventserver (hooks) and then could be docker composed together or upgraded individually? thinking a no-hooks ES, and a hooks ES? I could be over simplifying? my knowledge isn’t super deep, but If I can help LMK

@dlandon FYI in the latest build for GPU, as the file is mv not cp the .pub is not there when opencv_ok has 'yes' present. So fails on rebuilding Jan 7 07:47:39 3aa4a63b1d0a EventServer: Compiling opencv with GPU Support Jan 7 07:47:39 3aa4a63b1d0a EventServer: Installing cuda toolkit... Warning: apt-key output should not be parsed (stdout is not a terminal) gpg: can't open '/var/cuda-repo-10-2-local-10.2.89-440.33.01/7fa2af80.pub': No such file or directory Warning: apt-key output should not be parsed (stdout is not a terminal) gpg: can't open '/var/cuda-repo-10-2-local-10.2.89-440.33.01/7fa2af80.pub': No such file or directory W: GPG error: file:/var/cuda-repo-ubuntu1804-11-1-local Release: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY F60F4B3D7FA2AF80 EDIT: Also need to update to 20.04 Nvidia components

AMAZING! and ffmpeg shows cuvid and cuda for `-hwaccels` Heroic work! Happy New Year

Thanks for the quick response - that was exactly what i was looking for. I've been asking as many folks to create unraid templates where possible, but sometimes it doesnt make sense for them. Thanks for maintaining the CA - its a real game changer for Unraid!

Possible stupid question incoming.... Is it possible to create an empty Docker template. The user can then add variables, paths and the dockerhub info in the GUI (assuming you pull the update from dockerhub and not github?) This would be handy for: 1. Testing Docker images that arent in the CA 2. Docker images maintained by someone who doesnt want to build the Unraid App and still benefit from the update mechanics (and Unraid-API)

Yeah - still no dice on this. FFMPEG on 18.04 with cuvid is a right cluster!

This is the thread on Zoneminder forum: https://forums.zoneminder.com/viewtopic.php?t=30153 @sivart