Azreal

posted this in general support but no real luck so hopefully find some here, added some additional details also :-). Been having some networking issues off / on since before the 7.0 upgrade, but seems to be worse now. Originally i was having issues connecting to unraid connect, the CA wasn't working properly to detect updates. doing the unraid-api restart command seemed to resolve it sometimes or maybe for 5 minutes then back... my containers all still seemed to work fine and i didn't notice any problems with their connectivity. For the last week or so i've been having issues where the containers are unable to connect to internet. For the ones that have ping command available they show a name resolution error. When this came up last week i found some articles pointing to turn off enable host access to custom network, do a force update, try again. Some variation of that did work for me eventually. I believe the last thing i did was to set them to host networking, spin them up, they connected, set them back to my custom network IP and then start and they all just worked. until last night.... I've done similar steps today, turned off the enable host access to custom network and tried again. no change. set to host network for the container and they work fine, when i go back to my custom network and the designated ip they start up fine but don't connect. i've also wiped out and rebuilt the docker image file with just a handful of the applications restored while i try and work through this, no improvement. I've run this ipvlan network for 2-3 years now mostly without issue once i understood how it worked, not sure what's going on now and how to troubleshoot next. I do run an unbound container to provide dns but as part of the troubleshooting here i pointed all the vlans to just use automatic dns from my net provider and bypass my dns. also have tailscale running on this in case that matters - my primary container i access via tailscale is still accessible remotely so it seems fine high level overview of my network is i have containers on different vlan's - most on the same subnet as the host server (.10.x), i have my omada container on the same vlan as the networking gear (.1.x) I do have 2 nics in the server and one goes to each subnet. Containers on both subnets are having problems. but not all of the containers have this problem... If i set omada to use host network, i can get to internet and it all works, as soon as i set back to custom network it doesn't. diagnostics attached. suggestions appreciated. rlyeh-diagnostics-20250204-1631.zip

yes docker pull works, it's the application in that container that isn't getting out to the internet for some reason. if i set the container to host networking it works, when i set it to my customer ipvlan network ip it won't. ip's for each app on ipvlan network has been my setup for couple of years without any headaches or issues like i'm having now.

Hello from Docker! This message shows that your installation appears to be working correctly. To generate this message, Docker took the following steps: 1. The Docker client contacted the Docker daemon. 2. The Docker daemon pulled the "hello-world" image from the Docker Hub. (amd64) 3. The Docker daemon created a new container from that image which runs the executable that produces the output you are currently reading. 4. The Docker daemon streamed that output to the Docker client, which sent it to your terminal. To try something more ambitious, you can run an Ubuntu container with: $ docker run -it ubuntu bash Share images, automate workflows, and more with a free Docker ID: https://hub.docker.com/ For more examples and ideas, visit: https://docs.docker.com/get-started/

Been having some networking issues off / on since before the 7.0 upgrade, but seems to be worse now. Originally i was having issues connecting to unraid connect, the CA wasn't working properly to detect updates. doing the unraid-api restart command seemed to resolve it sometimes or maybe for 5 minutes then back... my containers all still seemed to work fine and i didn't notice any problems with their connectivity. For the last week or so i've been having issues where the containers are unable to connect to internet. For the ones that have ping command available they show a name resolution error. When this came up last week i found some articles pointing to turn off enable host access to custom network, do a force update, try again. Some variation of that did work for me eventually. I believe the last thing i did was to set them to host networking, sping them up, they connected, set them back to my custom network IP and then start and they all just worked. until last night.... I've done similar steps today, turned off the enable host access to custom network and tried again. no change. set to host network for the container and they work fine, when i go back to my custom network and the designated ip they start up fine but don't connect. I've run this ipvlan network for 2-3 years now mostly without issue once i understood how ti worked, not sure what's going on now and how to troubleshoot next. I do run an unbound container to provide dns to my vlans but as part of the troubleshooting here i pointed all the vlans to just use automatic dns from my net provider and bypass my dns. diagnostics attached. suggestions appreciated. rlyeh-diagnostics-20250204-0742.zip

yeah was hoping to understand why it's happening when it didn't for months. however, where would i define it? is it under the go2rtc -> streams? or is it in the camera setup under output_args or inputs? i see it in the reference under the detect stream, but that doesn't sound right to me or i don't think it aligns with what i'm doing anyways. these particular camera are just continuous recording, no object detection or other triggered type events for these.

Wierd live view resizing on the fly. running frigate for a few months w/out issues. couple of weeks ago the live view started acting weird. i have 8 tapo c210's. they always showed the view in a nice widescreen looking sizing, now they randomly resize themselves looks like similar to bouncing from 16:9 to 4:3 - but i never changed any settings. they will just adjust on their own with no visible reason why. screenshots included for reference. A and C are the same cams, B and D are the same cams. You can see the size difference on B and D. they'll just resize seemingly randomly. what is going on here? i think i saw in the config i can specify a size but never had to previously.

installing my newly aqcuired pcie card this week, any specific recommendations / example links re: active cooling?

the lasted frigate UI hits this squarely on the head in my opinion. the scrubbing / review interface is really helpful. bars in the timeline to show if there's movement in those frames, more movemen, bigger bars type of thing. you can see all the cameras in sync while you scrub so if you're tracking something cam to cam you can see all of those feeds at the same time. Curious why you care about the filesize though, that's all managed by the system. Depending on the camera settings and what is going on in the images the file size can vary. I have 8 cams of the same model and each one produces vastly different file sizes, some have very active images some are predominantly static 90%+ of the time. but again, except for managing my storage i don't really care too much about the individual files. you can also configure some pretty simple or pretty complex storage rules depending on your needs. my doorbell cam generates the most storage per day so i keep fewer days of that in long term storage. the rest of my cams i have a longer storage period for. there's a lot to love in the latest frigate even if you're not using or not interested in the various AI type stuff they're investing a lot of work in.

didn't think about the heat sink at all and hadn't seen comments for that elsewhere so thank you!

Mini PCIe Coral Device? https://coral.ai/products/pcie-accelerator 2 Questions: 1. Has anyone used this and how did it perform for them? 2. My machine currently is too old to suppport IOMMU so i can't use the USB version, but assume since that's totally different tech stack i shouldn't need to worry about accessing a mini pcie device right? (if you're wondering my very long in the tooth i7-4770k w/16GB DDR3 RAM) runs well enough but obviously hammering the system using CPU and 9 cameras and 20+ other containers going as well. don't mind investing 25$ for the coral if the feedback on the question above is positive

thank you thank you. don't know my way around the file structures so well yet so didn't know where to find it. appreciate your response.

trying to run the reprocess command after i edited the config.yml to exlude some categories and it errors. did this other day no problem but honestly wouldn't be surprised if i'm doing something wrong. I open console from docker page for the container and shouldn't it just be "bitmagnet reprocess" and enter? I receive "bash: bitmagnet: command not found" i get similar if i try from the unraid console with "docker exec -it binhex-bitmagnet bitmagnet reprocess" returns the following: OCI runtime exec failed: exec failed: unable to start container process: exec: "bitmagnet": executable file not found in $PATH: unknown am i doing it wrong or is there something up wtih the latest image?

Solved it myself.... suddenly dawned on me that i setup Tailscale the other week with spaceinvaderone's video, turned it off and it works now. i got what i was after with that, but apparently needs some additional config to play nice with qbittorrent. fun times incoming. if anyone knows right offhand what to change to allow containers to ignore tailscale or select them individually that would be very helpful. for anyone else's benefit - there's a very simple toggle for using or not using tailscale dns, it even says it will impact server and containers, i'm not sure how it get turned on but turning it off and restarting was all i had to do.

logs below. currently it shows me trying 8.8.8.8,1.1.1.1 but my normal setup is my own isntance of unbound running internally. same errors across the board for all dns i've tried. this has been pretty damn stable for me >1 year. i even tried isntalling another instance side by side under another IP and redoing the config from scratch, same issues. docker run -d --name='binhex-qbittorrentvpn' --net='br0' --ip='192.165.10.8' --pids-limit 2048 --privileged=true -e TZ="xxxxx" -e HOST_OS="Unraid" -e HOST_HOSTNAME="Rlyeh" -e HOST_CONTAINERNAME="binhex-qbittorrentvpn" -e 'TCP_PORT_6881'='6881' -e 'UDP_PORT_6881'='6881' -e 'TCP_PORT_8080'='8080' -e 'TCP_PORT_8118'='8118' -e 'VPN_ENABLED'='yes' -e 'VPN_USER'='xxxxxx' -e 'VPN_PASS'='xxxxxx' -e 'VPN_PROV'='pia' -e 'VPN_CLIENT'='wireguard' -e 'VPN_OPTIONS'='' -e 'STRICT_PORT_FORWARD'='yes' -e 'ENABLE_PRIVOXY'='no' -e 'WEBUI_PORT'='8080' -e 'LAN_NETWORK'='192.165.10.0/24' -e 'NAME_SERVERS'='8.8.8.8,1.1.1.1' -e 'VPN_INPUT_PORTS'='' -e 'VPN_OUTPUT_PORTS'='' -e 'DEBUG'='true' -e 'UMASK'='000' -e 'PUID'='99' -e 'PGID'='100' -l net.unraid.docker.managed=dockerman -l net.unraid.docker.webui='http://[IP]:[PORT:8080]/' -l net.unraid.docker.icon='https://raw.githubusercontent.com/binhex/docker-templates/master/binhex/images/qbittorrent-icon.png' -v '/mnt/user/media/_torrents/':'/data':'rw' -v '/mnt/user/appdata/binhex-qbittorrentvpn':'/config':'rw' --sysctl="net.ipv4.conf.all.src_valid_mark=1" 'binhex/arch-qbittorrentvpn' 8b828fa260c27e3c01189c8d3ad72f98e49c7e07ff488205255d1b04ce2706c8 The command finished successfully! 2024-09-03 12:16:46.030435 [info] Host is running unRAID 2024-09-03 12:16:46.058144 [info] System information: Linux 8b828fa260c2 6.1.106-Unraid #1 SMP PREEMPT_DYNAMIC Wed Aug 21 23:36:07 PDT 2024 x86_64 GNU/Linux 2024-09-03 12:16:46.083214 [info] Image tags: INT_RELEASE_TAG=2024073001,IMAGE_RELEASE_TAG=4.6.6-1-01 2024-09-03 12:16:46.124578 [info] PUID defined as '99' 2024-09-03 12:16:46.154655 [info] PGID defined as '100' 2024-09-03 12:16:46.201928 [info] UMASK defined as '000' 2024-09-03 12:16:46.230146 [info] Permissions already set for '/config' 2024-09-03 12:16:46.255551 [info] Deleting files in /tmp (non recursive)... 2024-09-03 12:16:46.283534 [info] VPN_ENABLED defined as 'yes' 2024-09-03 12:16:46.317664 [info] VPN_CLIENT defined as 'wireguard' 2024-09-03 12:16:46.343950 [info] VPN_PROV defined as 'pia' 2024-09-03 12:16:46.376452 [info] WireGuard config file (conf extension) is located at /config/wireguard/wg0.conf 2024-09-03 12:16:46.416228 [info] VPN_REMOTE_SERVER defined as 'ca-vancouver.privacy.network' 2024-09-03 12:16:46.450141 [info] VPN_REMOTE_PORT defined as '1337' 2024-09-03 12:16:46.472329 [info] VPN_DEVICE_TYPE defined as 'wg0' 2024-09-03 12:16:46.497840 [info] VPN_REMOTE_PROTOCOL defined as 'udp' 2024-09-03 12:16:46.528381 [info] USERSPACE_WIREGUARD not defined (via -e USERSPACE_WIREGUARD), defaulting to 'no' 2024-09-03 12:16:46.558654 [info] NAME_SERVERS defined as '8.8.8.8,1.1.1.1' 2024-09-03 12:16:46.604374 [debug] iptables default policies available, setting policy to drop... 2024-09-03 12:16:46.658186 [debug] ip6tables default policies available, setting policy to drop... 2024-09-03 12:16:50.719670 [debug] Having issues resolving name 'ca-vancouver.privacy.network', sleeping before retry... 2024-09-03 12:16:59.767254 [debug] Having issues resolving name 'ca-vancouver.privacy.network', sleeping before retry...

been running this container >1 year pretty much no problem. as of last week's appdata backup i noticed on restart it wasn't connecting anymore. I see the following in the debug logs. 2024-09-03 10:47:21.224651 [debug] Having issues resolving name 'ca-vancouver.privacy.network', sleeping before retry. I've tried several endpoints and it just changes the location it says it can't reach. I can hit these end points with the app on my windows 10 vm. i'm using wireguard with port forwarding required. - same type of problem with openvpn too. account is still valid, again i've been able to connect just fine with the application. any suggestions on how to resolve? - i've already removed / reinstalled the container. i setup another instance from scratch just to get same errors. i've changed my name servers from my private ones to public ones. none of these changed things. anyone else having similar issue or have these end points working?