0.12.0 Beta 2 #4948
Replies: 29 comments 152 replies
-
|
I have created a short survey for those that are using the new OpenVINO and TensorRT detectors. It will aid in understanding the expected performance for different devices and help identify other issues that may be occurring. Thanks! Answers can bee seen here |
Beta Was this translation helpful? Give feedback.
-
|
Not really a performance comment so I'll put it here, and I realise this isn't a Frigate problem as such - but OpenVINO nearly melted my mini-PC, a GMK NucBox, Intel Celeron J4125 with UHD600 iGPU. While it's actively cooled, and does throttle the CPU when it gets too hot, this form factor really wasn't built for continuous high GPU activity - temps hit 100C and kernel spammed a load of hardware errors before it powered down. My own fault really, and I am overdue moving Frigate onto one of my "proper" server - but given you get quite a few people here from HomeAssistant backgrounds, and many of them are running on lower power PCs (many even passively cooled), it might be worth a warning somewhere in the docs of the CPU/GPU-stretching capabilities of the model. |
Beta Was this translation helpful? Give feedback.
-
|
I tested on a passively cooled cheap celeron 10th gen laptop and it ran openvino without issues. Definitely worth keeping in mind but will need to keep getting data about which devices it does and doesn't work well with so more specific recommendations can be made |
Beta Was this translation helpful? Give feedback.
-
|
Interesting, I wonder if it's down to variations in the iGPU models. I will do some more experimenting! |
Beta Was this translation helpful? Give feedback.
-
|
About tensort container, is there any specific reason to use 22.07-py3? Is better to use the latest avaible tag in this case 22.12-py3? Thanks |
Beta Was this translation helpful? Give feedback.
-
|
cc @NateMeyer |
Beta Was this translation helpful? Give feedback.
-
|
Just asking, as there is no "latest" tag manifest I'll have to update the unraid template manually and it's good no know if I have to maintain it in the latest avaiable tag or I can keep 22.07 as default and just add a note to users that they can change the tag manually if they need to. |
Beta Was this translation helpful? Give feedback.
-
|
That tag was chosen to line up (approximately) with the versions of the libraries used in the image. TensorRT will complain if you generate the model with newer libraries. |
Beta Was this translation helpful? Give feedback.
-
|
Ok Understood |
Beta Was this translation helpful? Give feedback.
-
|
I wanted to react on the Beta1 discussion, but it seems to be closed. I know that my VAAPI is working, but I am wondering if I can get this to work under
|
Beta Was this translation helpful? Give feedback.
-
|
What system are you running on? It needs to have privileged enabled for intel_gpu_top to work |
Beta Was this translation helpful? Give feedback.
-
|
Copied from the other discussion: I am running it succesfully on a Also just tested I have mounted |
Beta Was this translation helpful? Give feedback.
-
|
I see, everything you are showing there is working as intended. It is a synology limitation that intel_gpu_top will not work in containers unfortunately. |
Beta Was this translation helpful? Give feedback.
-
|
UnRaid betatesters you need to uninstall and install again the beta apps from CA. I tested both nvidia and classic and both are running fine in my system. |
Beta Was this translation helpful? Give feedback.
-
|
UnRaid betatesters. I have pushed a modification that will be available within 2 hours. I changed the extra parameter
I also changed the TensorRT template with similar modifications but with only I tested both changes in my system with no issues. You must delete and reinstall to test this. These changes are made following the recomendation of the nvidia-driver pluging maintainer on UnRaid to have a better reliability and to try to avoid the issues found by some users like: |
Beta Was this translation helpful? Give feedback.
-
|
I just modified old template to suit 0.12 needs but I'm struggling with webrtc - everything else works as intended. Do I need to install go2rtc along and if so - it uses 1954 port which your template uses too... |
Beta Was this translation helpful? Give feedback.
-
|
Ok so jsmpeg works, mse works but is slower, webrtc never loads up. and frigate-go2rtc.yaml and part of logs: |
Beta Was this translation helpful? Give feedback.
-
|
Does the CA template set network mode as host? You will need to do that OR add a local IP address candidate to the frigate-go2rtc.yaml |
Beta Was this translation helpful? Give feedback.
-
|
The network is bridged and the port is mapped. The frigate-go2rtc.yaml has to be the one proposed on the docs and it works fine. |
Beta Was this translation helpful? Give feedback.
-
|
Ok it's in bridge mode so I had to add local ip and it started to work - thanks for that seems this image works as intended - nothing bad in logs, detection is all good - happy days. |
Beta Was this translation helpful? Give feedback.
-
|
Glad it's working 👍 |
Beta Was this translation helpful? Give feedback.
-
|
when setting up the models in the config file for tensorRT how do we know what model to use and its width and height? |
Beta Was this translation helpful? Give feedback.
-
|
awesome, thanks for your time. im building the models now. |
Beta Was this translation helpful? Give feedback.
-
|
@NickM-27 the TensorRT Performance Questionare. do we give you the information to add to that? |
Beta Was this translation helpful? Give feedback.
-
|
It's a survey you should be a able to fill it out yourself. |
Beta Was this translation helpful? Give feedback.
-
|
is there a link to the survey? i found the google sheets link that i have to request write access to. |
Beta Was this translation helpful? Give feedback.
-
|
There's two links in my comment it's the first one https://forms.gle/cnCR8DFt8qJix5ws5 |
Beta Was this translation helpful? Give feedback.
-
|
Alright guys I'm back and getting the same error "Address Already in use" detectors:
cpu1:
type: cpu
num_threads: 3
cpu2:
type: cpu
num_threads: 3
mqtt:
enabled: True
host: 10.0.0.26
database:
path: /media/frigate/frigate.db
logger:
default: info
logs:
frigate.event: debug
#######Global Camera Settings#######
detect:
width: 1920
height: 1080
fps: 15
record:
enabled: True
events:
pre_capture: 15
post_capture: 15
retain:
default: 5
mode: active_objects
objects:
dog: 2
cat: 2
restream:
enabled: True
############ Cameras!!!##############
######Driveway Camera!!##########
cameras:
driveway:
ffmpeg:
output_args:
record:
-f segment
-segment_time 45
-segment_format mp4
-reset_timestamps 1
-strftime 1
-codec:v h264_vaapi
-global_quality 25
-acodec aac
-max_interleave_delta 0
-fps_mode passthrough
hwaccel_args:
-hwaccel vaapi
-hwaccel_device /dev/dri/renderD128
-hwaccel_output_format yuv420p
inputs:
- path: rtsp://10.0.0.35:8554/driveway
roles:
- detect
- record
- path: rtsp://[email protected]:2600/media/ch0/stream0
roles:
- restream
objects:
track:
- person
- cat
- dog
filters:
person:
min_area: 5000
max_area: 100000
min_score: 0.5
threshold: 0.75
mask:
- 523,1080,897,0,0,0,0,1080
cat:
min_area: 5000
max_area: 100000
min_score: 0.5
threshold: 0.8
mask:
- 523,1080,897,0,0,0,0,1080
dog:
min_area: 5000
max_area: 100000
min_score: 0.5
threshold: 0.8
mask:
- 523,1080,897,0,0,0,0,1080
############Backroom Camera#######
backroom:
ffmpeg:
output_args:
record:
-f segment
-segment_time 45
-segment_format mp4
-reset_timestamps 1
-strftime 1
-codec:v h264_vaapi
-global_quality 25
-acodec aac
-fps_mode passthrough
hwaccel_args:
-hwaccel vaapi
-hwaccel_device /dev/dri/renderD128
-hwaccel_output_format vaapi
inputs:
- path: rtsp://10.0.0.36:8554/backroom
roles:
- detect
- record
- path: rtsp://[email protected]:88/videoMain
roles:
- restream
objects:
track:
- person
- cat
- dog
filters:
person:
min_area: 5000
max_area: 100000
min_score: 0.5
threshold: 0.75
mask:
- 1920,330,1920,0,0,0,0,343
cat:
min_area: 5000
max_area: 100000
min_score: 0.5
threshold: 0.8
mask:
- 1920,330,1920,0,0,0,0,343
dog:
min_area: 5000
max_area: 100000
min_score: 0.5
threshold: 0.8
mask:
- 1920,330,1920,0,0,0,0,343I tried changing ports ie 5000 to 5006 or whatever which didn't change anything. |
Beta Was this translation helpful? Give feedback.
-
|
I notice 1935 is still there, have you removed that from your docker compose? If you are going to use the restream no reason to leave that in there |
Beta Was this translation helpful? Give feedback.
-
|
Yeah, Im going to get that removed. |
Beta Was this translation helpful? Give feedback.
-
Do you have any other suggestions on this issue? Thanks in advance. |
Beta Was this translation helpful? Give feedback.
-
|
If you remove all the ports from the mapping to the container does the problem go away? If so then add one port back at a time till you find the culprit. |
Beta Was this translation helpful? Give feedback.
-
|
@Dvalin21 is that output from netstat from when frigate was running? I don't see why it would be a problem suddenly, but Frigate does also use port 5002 inside the container. I wouldn't expect that to cause an issue unless you were running in host mode. |
Beta Was this translation helpful? Give feedback.
-
|
Stream issue over vpn/cellular I am testing Frigate 0.12.0Beta2 with Home Assistant using the webrtc pointing to go2rtc api included in Frigate. On local lan everything works without any error. When I use cellular connection on my phone and access HA over vpn, I see following errors in HA. 2023-01-07 15:52:09.734 ERROR (MainThread) [frontend.js.latest.202301040] http://192.168.0.10:8123/webrtc/video-rtc.js:563:33 NotFoundError: The object can not be found here 2023-01-07 16:35:06.875 ERROR (MainThread) [frontend.js.latest.202301040] http://srv.home.com:8123/webrtc/video-rtc.js:402:31 Uncaught InvalidStateError: Failed to read the 'buffered' property from 'SourceBuffer': This SourceBuffer has been removed from the parent media source. type: custom:webrtc-camera Anyone else seen this issue? I wonder if this is due to slow network or something else. If I use Frigate card in HA it does show the camera feed over vpn but there is about 18sec latency. Additional update: |
Beta Was this translation helpful? Give feedback.
-
|
Just upgraded from 0.11.1 and unfortunately having some issues.
mqtt:
host: 172.16.16.1
user: username
password: password
cameras:
front:
ffmpeg:
inputs:
# - path: rtsp://localhost:8854/front
# roles:
# - detect
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=2
roles:
- detect
- restream
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=0
roles:
- record
objects:
track:
- person
- vehicle
- animal
record:
events:
required_zones:
- sidewalk
- driveway
snapshots:
required_zones:
- sidewalk
- driveway
zones:
sidewalk:
coordinates: 0,741,1533,339,1359,286,921,350,0,561
driveway:
coordinates: 468,1080,1920,1080,1920,583,1785,519,1620,646,1121,484,304,701
motion:
mask:
- 1510,354,1920,285,1920,0,1324,0
- 0,1080,250,1080,147,675,0,703
side:
ffmpeg:
inputs:
# - path: rtsp://localhost:8854/side
# roles:
# - detect
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=2
roles:
- detect
- restream
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=0
roles:
- record
objects:
track:
- person
- animal
motion:
mask:
- 244,0,254,285,435,273,430,79,1641,1080,1920,1080,1920,0
improve_contrast: True
threshold: 15
back:
ffmpeg:
inputs:
# - path: rtsp://localhost:8854/back
# roles:
# - detect
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=2
roles:
- detect
- restream
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=0
roles:
- record
objects:
track:
- person
- animal
motion:
mask:
- 1357,48,1666,101,1920,201,1920,0,0,0,0,79
inside:
ffmpeg:
inputs:
- path: rtsp://localhost:8854/inside
roles:
- detect
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=2
roles:
- restream
- path: rtsp://username:[email protected]:554/cam/realmonitor?channel=1&subtype=0
roles:
- record
motion:
mask:
- 1920,0,1920,74,633,72,637,0
record:
enabled: True
retain:
days: 7
mode: all
events:
retain:
default: 14
mode: active_objects
detect:
width: 1920
height: 1080
model:
labelmap:
2: vehicle
3: vehicle
5: vehicle
7: vehicle
15: animal
16: animal
17: animal
ffmpeg:
# hwaccel_args: -hwaccel vaapi -hwaccel_device /dev/dri/renderD128 -hwaccel_output_format yuv420p
hwaccel_args: preset-vaapi
output_args:
# record: -f segment -segment_time 10 -segment_format mp4 -reset_timestamps 1 -strftime 1 -c copy
record: preset-record-generic-audio
detectors:
coral:
type: edgetpu
device: pci |
Beta Was this translation helpful? Give feedback.
-
|
Haha, thanks! Man, I even looked at that multiple times... 🤦 Edit: Everything appears to be working but the only thing I'm seeing in the logs now is: |
Beta Was this translation helpful? Give feedback.
-
|
@shred86 so is this all solved? |
Beta Was this translation helpful? Give feedback.
-
|
@NickM-27 Everything appears to be working fine now! However, I do have the "Failed to initialize PMU!" entry in my Frigate logs. Not sure if that's a nuisance error or something I need to address? |
Beta Was this translation helpful? Give feedback.
-
|
That's going to be because intel GPU top isn't working, may need to run as privileged for that to work. |
Beta Was this translation helpful? Give feedback.
-
|
Just updated my docker compose with privileged set to true and that fixed it, thanks! |
Beta Was this translation helpful? Give feedback.
-
|
Incredible work on this. Thank you very much. Beta 2 running like an absolute champ. I do have one question- I am on an intel 12700k cpu, which comes with the openvino 3.0 compatible Intel GNA; I set Frigate to use openvino AUTO and by golly the cpu is hardly being touched. Is it using the Intel GNA? If so, Frigate and an Alderlake processor make a pretty cool combo. 😎 |
Beta Was this translation helpful? Give feedback.
-
|
Ohh, the GPU took the wheel. At any rate, Frigate & 12700k a winning combo. 👌👌
|
Beta Was this translation helpful? Give feedback.
-
|
I'm not sure if it would be using the GNA, you could try setting it to GNA instead of AUTO to check. But my guess is it's running on the GPU. |
Beta Was this translation helpful? Give feedback.
-
|
It's definitely running on the GPU, (detector & ffmpeg) and doing a fine job. I'll try GNA instead of AUTO and see what happens. Thanks. |
Beta Was this translation helpful? Give feedback.
-
|
No go with GNA - maybe I need to add its path to the container, but I am not sure what the path even is.. .System sees it, though. |
Beta Was this translation helpful? Give feedback.
-
|
I'm using unraid and me thinks there should be a /dev/gna or /dev/intel_gna and there isn't... so maybe needs drivers installed. |
Beta Was this translation helpful? Give feedback.
-
|
I am helping a friend set his up and he is using the Annke C800 cameras. I noticed in the new config; https://deploy-preview-4055--frigate-docs.netlify.app/configuration/camera_specific It still refers to rtmp and not restream in a few places, is that still correct or does it need to be changed to restream? The link also has H264 when the camera doesn't support that? |
Beta Was this translation helpful? Give feedback.
-
|
Rtmp is still in frigate just deprecated so there are some places it hasn't been removed. Restream is recommended in all cases instead of rtmp. |
Beta Was this translation helpful? Give feedback.
-
|
https://deploy-preview-4055--frigate-docs.netlify.app/configuration/camera_specific/ Looking at the Annke C800 on this site is restream in place of rtmp also when viewing from mse no image shows Only see image with jsmpeg with no sound When in mse,webrtc i have image in debug only. |
Beta Was this translation helpful? Give feedback.
-
|
I am facing the same issue. I purposely don't want to restream (extra cpu usage), so MSE is mij go to solution. In the 'old' frigate I had a working live view. Now with this beta the same as here. Only debug is giving me a live preview. |
Beta Was this translation helpful? Give feedback.
-
What you're saying doesn't make sense, MSE requires the restream. If you want the old live view then just choose jsmpeg |
Beta Was this translation helpful? Give feedback.
-
|
Just realized there is a bug where mse is shown as the only option instead of jsmpeg when restream is disabled, will fix for beta3 |
Beta Was this translation helpful? Give feedback.
-
|
That's quick, thanks :) |
Beta Was this translation helpful? Give feedback.
-
What issues are you having? I have AMD GPU running on the 11.0 (went back to it since I can't figure out why 12.0beta2 keeps telling me address already used) |
Beta Was this translation helpful? Give feedback.
-
|
Any idea why 2 of my cameras are showing as 5fps and are using 7-8% cpu for ffmpeg when I have detect disabled on them? They are both set at 15fps and are only used to record 24/7, detection not neccessary. I didn't think ffmpeg did any converting when only recording?
|
Beta Was this translation helpful? Give feedback.
-
|
Detect disabled only turns off object detection, the stream has always been decided and used 24/7 when detect is disabled. For example if you had recordings but only wanted to save recording segments with motion in them. Also required for birdseye. There are existing feature requests to be able to fully disable the detect stream and the ffmpeg process when not in use. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for clarifying, I've set the resolution and fps to match each camera's output now to hopefully stop any re-encoding :) |
Beta Was this translation helpful? Give feedback.
-
|
That's not how that works. The stream will always be decoded for the detect stream no matter what, and the detect stream runs even when object detection and motion detection is disabled. Setting higher fps is most likely just going to create more work for the decoder. If you're not using the camera for any object or motion detection then you'd be better off setting the fps to 1 |
Beta Was this translation helpful? Give feedback.
-
|
Tried Beta 2 and I'm so impressed - great work! One issue, in the log I see: Any idea why? Running on Ubuntu Linux and using Docker Compose. |
Beta Was this translation helpful? Give feedback.
-
|
I fixed it by adding: |
Beta Was this translation helpful? Give feedback.
-
|
The documentation is a little sparse on Restream at the moment, which is to be expected, but I'd like a little more clarification on maintaining a single connection to the cameras. From what I've read it seems this is the only way to maintain a single connection to the camera: So would it be correct to assume this will create two connections? And if so, surely there's a way to simplify it or handle it in the background so a new path doesn't have to be manually defined for each one. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, what you have laid out is correct. However, it is also on purpose given that there needs to be a way for a user to connect directly to the camera if they choose. It could certainly be made more clean but I believe that's a job for a future version. For now the restream is mainly a way to replace rtmp and provide live view options. I also don't see a problem with defining multiple paths, that way it is explicit exactly what is happening. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, as long as it's intentional I have no issues doing it manually for each camera. I just wanted to make sure the documentation and my interpretation of it was correct. |
Beta Was this translation helpful? Give feedback.
-
|
anyone have a live stream url working via actionable HA notification? what ive been using is this: |
Beta Was this translation helpful? Give feedback.
-
|
Using https://github.com/AlexxIT/WebRTC 3.0.0 beta I have it pointed right at go2rtc inside frigate (frigate_ip:1984) and use the included service to create a link to the stream |
Beta Was this translation helpful? Give feedback.
-
I've tried a few ways but still can't figure it out. Can you post your config @NickM-27 ? Looks like the way I tried above is wrong |
Beta Was this translation helpful? Give feedback.
-
|
The url should just be the camera name (this is with frigate card but process is the same) - camera_entity: camera.front_doorbell_cam
live_provider: webrtc-card
webrtc_card:
url: front_doorbell_cam
|
Beta Was this translation helpful? Give feedback.
-
|
I'm using the addon and i'm seeing BTW |
Beta Was this translation helpful? Give feedback.
-
|
Those errors are due to Intel gpu stats not working, this is expected to not work in the addon install. I'll look into the copy button. |
Beta Was this translation helpful? Give feedback.
-
|
I get this error when trying to save the config file.. I am sure its an easy fix.. does something need to be added to the docker compose file? |
Beta Was this translation helpful? Give feedback.
-
|
You probably have |
Beta Was this translation helpful? Give feedback.
-
|
You are correct, there was an I did not read the release notes that close, but it is in there to remove it also. Good Job!! Thanks!! |
Beta Was this translation helpful? Give feedback.
-
|
@yayitazale do you get these errors in your logs? I am getting 30-40 per day |
Beta Was this translation helpful? Give feedback.
-
|
You can see if you are running low on space on the /storage page on the UI: |
Beta Was this translation helpful? Give feedback.
-
|
I pumped the cache folder to 1 gig and still get the errors, so it must be something else |
Beta Was this translation helpful? Give feedback.
-
|
|
Beta Was this translation helpful? Give feedback.
-
|
Theorically, if you have 4 cameras with ~= 3000 MB/hr 3000MB/hr / 3600s/hr = 0.833MB/s per camera > *4 cameras * every 10 sec cache is emptied = 33,33 MB So with 100MB yo should be fine if I'm understanding it correctly @NickM-27 can you confirm? |
Beta Was this translation helpful? Give feedback.
-
|
I have 7 cameras. I've increased the shm to 512 meg & cache to 1000 meg. Doesn't seem to have made a difference, still getting the same cache errors: |
Beta Was this translation helpful? Give feedback.
-
|
Has anyone else noticed increased instances of getting the "No frames have been received, check error logs" error on the 0.12 betas? Seems to happen every night (sometimes randomly during the day) on different camera(s) but can't seem to figure out the cause. Cameras aren't set to restart or anything and it's odd they don't recover like pre-v0.12.0. Restart usually fixes it though. I can't tell if the logs are helpful in this but just turned on debugging to see if that shows anything useful for tonight. |
Beta Was this translation helpful? Give feedback.
-
|
You have python bus error which is covered in the docs https://docs.frigate.video/troubleshooting/faqs#fatal-python-error-bus-error |
Beta Was this translation helpful? Give feedback.
-
|
Ugh! How did I miss that?? My eyes were drawn to the pretty green on the page! I'll check that out. Thanks! |
Beta Was this translation helpful? Give feedback.
-
|
after the latest rc.7 release of webrtc in beta2 getting weird stream of reolink camera its a h264 5mp source cam but webrtc seems to be streaming it out at 240FPS. Direct stats from cam: Direct stats from webrtc The FFMPEG cmds for this cam is: |
Beta Was this translation helpful? Give feedback.
-
|
Interesting, I tried it with all 3 of my reolink cams and I am seeing the restream as the same fps as the original. Will see if others see the same |
Beta Was this translation helpful? Give feedback.
-
|
I restarted the container and the issue hasnt come back - though I notice the ffprobes seem to offer transient responses on the sub stream for the same camera i noticed: Same command again a short time later perhaps ffmpeg set the wrong FR initially due to initially not having data from the stream?? 🤷 |
Beta Was this translation helpful? Give feedback.
-
|
Hi: I have been running the Frigate 12 beta 1 and 2 for last week and I see a substantial increase in CPU usage. My average CPU usage has gone from about 15% to 35% using the same hardware and almost the same "config" just adapted to the new requirements: You can see in the next picture the impact of the beta (blue line indicates the beta test start date, first test were done with a different container name). I se almost the same CPU usage with Nvidia tensort versions: |
Beta Was this translation helpful? Give feedback.
-
|
When having the |
Beta Was this translation helpful? Give feedback.
-
|
I don't see any difference with |
Beta Was this translation helpful? Give feedback.
-
|
Okay, I'd say to just wait and see after my fix above is merged and beta3 is released. There's going to be a few more CPU metrics in the system page which may shed some light on things |
Beta Was this translation helpful? Give feedback.
-
|
ok! |
Beta Was this translation helpful? Give feedback.
-
|
on my system the go2rtc causes cpu usage bit higher as more clients use the camera streams. |
Beta Was this translation helpful? Give feedback.
-
|
Hello, Running 0.12b2 as HA addon. To enable restream I add to overwrite default network config restream port to 8554 I can't get Webrtc to work. When i hit the button in the Nginx log I have this
but nothing in the go2rtc log
I created a frigate-go2rtc.yaml as described in docs. I port forwarded 8555 in my router. Am I missing something ? A way to "open" API port ? |
Beta Was this translation helpful? Give feedback.
-
|
thanks for hint about browser log didn't knew about that, it shows :
I am gonna dig about this |
Beta Was this translation helpful? Give feedback.
-
|
I had that error initially. I'll have to check my notes on how I solved it, but off the top of my head I think I had both ports 1984 and 1985 mapped, but I then removed 1985; but don't quote me on that. I will do some checking and see if I can jog my memory. Disregard...I was thinking of something completely different; it's been a long day! |
Beta Was this translation helpful? Give feedback.
-
|
I activated debug in go2rtc file :
chrome debugger gives me this : any idea ?
|
Beta Was this translation helpful? Give feedback.
-
|
I digged a bit, with standard go2rtc (frigate shutt off during tests) webrtc works OK on chrome and companion app. I can telnet 8555. When I try with frigate, I cannot and it doesn't work. I guess it has something to do with my first message I had to manually overwrite ports to enable restream. I just tested again - if "default" I can't telnet 8554, if overwritten with 8554... it works. I guess it is the same problem for 8555 but not available in gui !? |
Beta Was this translation helpful? Give feedback.
-
|
Oh thank you, that is most likely the issue, I will create a PR to fix that for the addon |
Beta Was this translation helpful? Give feedback.
-
|
Ok looks like I have most of my cameras running, I just need help with using my GPU AMD PCIe card to help. |
Beta Was this translation helpful? Give feedback.
-
|
Ok we can test it with beta3, I'll start preparing the changes on the beta template. |
Beta Was this translation helpful? Give feedback.
-
Why do you want to make it longer? These segments are internal metadata and not meant to be interacted with by the user directly. There are a lot of advantages to the 10 second timeframe. |
Beta Was this translation helpful? Give feedback.
-
I agree, I think it is worth a try. Less apps in the store is good, we just need to make sure everyone removes what they don't need. Although I am not sure what happens if people leave variables that do not apply to their system? |
Beta Was this translation helpful? Give feedback.
-
|
@yayitazale The default value for LIBVA_DRIVER_NAME is |
Beta Was this translation helpful? Give feedback.
-
|
I pushed the change to "frigate-beta" app, it will be aviable soon. https://gitlab.com/yayitazale/unraid-templates/-/commit/074f6c1e667085c1d71948bb761c991abbfe2a8b |
Beta Was this translation helpful? Give feedback.
-
|
Amazing progress! So much new, great stuff. Would it be possible, at some point, to get an option to set the timezone of the Frigate location? When managing homes on multiple continents or simply traveling, it can get annoying and confusing to have to do a bit of mental arithmetic whenever you're looking at events or recordings. The system timezone is probably not the right option either, since that tends to be set to UTC on most servers. Please let me know if this is the wrong place to discuss this and if I should create a feature request instead. |
Beta Was this translation helpful? Give feedback.
-
|
With this update recordings are stored in utc and times in the frontend are the timezone of your browser With the new HA integration times in the media browser match the time set in HA settings. A feature request issue would be the right place to go into more detail 👍 |
Beta Was this translation helpful? Give feedback.
-
|
New issue: #5018 HA has the same issue. ;) |
Beta Was this translation helpful? Give feedback.
-
|
Very nice. Those acceleration presets are nice, and something got fixed either in Ubuntu/Linux firmware or ffmpeg or something in the latest few releases acceleration finally works on my Alder lake CPU now. I tried everything before to get it to work but had weird driver issues, it's now cleared up. |
Beta Was this translation helpful? Give feedback.
-
|
We updated the internal driver in the container to fix that. |
Beta Was this translation helpful? Give feedback.
-
|
I have not seen this error before... everything has been running great lately. It has not come up again. |
Beta Was this translation helpful? Give feedback.
-
|
never seen that before, looks like an error in the dependency. Seems like nothing to worry about unless it pops up again |
Beta Was this translation helpful? Give feedback.
-
|
The only thing I did before that time frame was update the HA integration to beta 2.. unless that had something to do with it. But everything seems to working fine. |
Beta Was this translation helpful? Give feedback.
-
|
It has to do with the frontend websocket dependency |
Beta Was this translation helpful? Give feedback.
-
|
@yayitazale do i need to select the gpu in the device setup.
do i need to change PCIe ACS override?? |
Beta Was this translation helpful? Give feedback.
-
|
I don't understand this. If you are launching frigate docker you don't need to change none of those. |
Beta Was this translation helpful? Give feedback.
-
|
I'm a bit confused, and before I open an issue, I was hoping to get some clarification here. My setup is HassOS with the HassOS addon. Clearly in the minority because all I see is Docker Compose and I have absolutely no clue about Docker. I have 0.12.0 Beta 2 and 4.0.0 beta 2 and WebRTC 3.0-beta 4. Any way, here are my questions:
Is this something that needs to be done for addon users as well? There still are Docker components?
So as an addon user I have to create a go2rtc.yaml somewhere and have frigate-go2rtc.yaml as well?
This seems to concern addon users as it refers to frigate-go2rtc.yaml. I need a nudge in the right direction. I assume I don't have Docker Compose as it's not installed. Docker CLI does, so is this done with docker run? |
Beta Was this translation helpful? Give feedback.
-
This is an addon limitations so addons don't have this option. This also won't work in beta2 you'll need to wait for beta3
All you need is the frigate-go2rtc.yaml file which goes next to frigate.yml |
Beta Was this translation helpful? Give feedback.
and others
-
Major Changes for 0.12.0
❗❗Notice❗❗
I want to recognize @NickM-27 for all the contributions he made on this release and all the support he helps provide in the issues. If you have been considering sponsoring this project with either a one time contribution or a recurring contribution, I would request that you do so at his sponsors page.
Link to updated docs: https://deploy-preview-4055--frigate-docs.netlify.app/
Changes Since Beta 1
preset-intel-vaapiandpreset-amd-vaapihave been renamed topreset-vaapidue to there being no differences between themrestream -> video_encodingoption which can be used to re-encode the restreamNoneoptionBREAKING CHANGES
./run: line 7: exec: nginx: not found, then you need to stop overriding your PATH. This happens with both Proxmox LXC and Portainer ([Support]: Beta11 Can't Find FFMPEG #3223 the database will need to be deleted to roll-back successfully.liveconfiguration has been moved torestream -> jsmpegRetention limited by available storage
Frigate now limits its recordings to available storage, if the storage for recordings gets below 1 hour left then Frigate will delete the oldest recording segments to make room for newer ones. Frigate will also not fail if there is no space to save recordings.
RTSP restreaming
Due to many issues with RTMP based restream, RTMP is now deprecated in favor of a new RTSP restream. This also comes with a new
restreamrole which is enabled by default. The restream of a camera can be accessed atrtsp://frigate_ip:8554/config_camera_name. The restream can also be used inside Frigate to reduce connections to the camera, see the restream docs for more info.Birdseye restreaming is also now supported at
rtsp://frigate_ip:8554/birdseyeNOTE: Port
8554will need to be mapped in the docker run or docker compose file for the restream to be accessed outside the container.New live stream options
The new RTSP restream also enables new
MSEandWebRTClive view options in the frontend which enable low-latency full-framerate live views that support audio.NOTES:
The live view options are set in the Frigate WebUI for each camera individually.
New detector types
Frigate now supports new detector types along with the Google Coral TPU.
OpenVINO
The OpenVINO detector type runs an OpenVINO IR model on Intel CPU, GPU and VPU hardware. OpenVINO is supported on 6th Gen Intel platforms (Skylake) and newer. A supported Intel platform is required to use the GPU device with OpenVINO. The MYRIAD device may be run on any platform, including Arm devices. For detailed system requirements, see OpenVINO System Requirements
TensorRT
NVidia GPUs may be used for object detection using the TensorRT libraries. Due to the size of the additional libraries, this detector is only provided in images with the -tensorrt tag suffix. The TensortRT detector is able to run on x86 hosts that have an Nvidia GPU which supports the 11.x series of CUDA libraries. The minimum driver version on the host system must be >=450.80.02. Also the GPU must support a Compute Capability of 5.0 or greater. This generally correlates to a Maxwell-era GPU or newer, check the TensorRT docs for more info.
NOTE: The link in the docs is for the final release location, for the beta the script is located at https://raw.githubusercontent.com/blakeblackshear/frigate/dev/docker/tensorrt_models.sh
New system page
The
Debugpage has been renamed toSystem. It now includes much more information and links to easily getvainfoandffprobedata.The GPUs section will show each GPU based on the hwaccel args, depending on the type of GPU different information is supported. For example Nvidia GPUs will have the name while others will be generic, also intel does not support memory usage. There is also a
VAINFObutton which will make it easier to diagnose hwaccel issues and verify that the hwaccel driver is being used correctly.Each camera will have their own process CPU & memory usage so it will be easier to see which process is using those resources. There is also an
FFPROBEbutton which will be helpful to understand what each stream is presenting and make it easier to include that in support issues.The debug config has also been removed, the config can be copied using the config page in the WebUI.
New storage page
Frigate WebUI now has a built in storage page which shows the general usage for both storage and memory. It also shows a per-camera storage usage and stream bandwidth. This should make it a lot easier to have a general idea how much storage is being used by each camera.
New logs page
Frigate WebUI now has built in logs, this will make copying and viewing logs much easier. Currently the logs are static meaning the page must be refreshed to view new logs.
New config editor
Frigate WebUI now has a built in config editor with syntax highlighting and schema validation. This means that the config will be validated before it can be saved which will make yaml config much easier to manage.
NOTE: The previous docker-compose or docker CLI may have had the /config.yml set to
:roor read-only. That will need to be removed for the config editor to be able to save the new file.MQTT is now optional
Frigate no longer requires MQTT to function, MQTT is also setup asynchronously so errors will be more clear and frigate won't stop when mqtt is enabled but not setup correctly.
NOTE: MQTT is still required for the Frigate-HomeAssistant Integration
No changes are needed for existing users.
FFMpeg presets
FFMPEG presets for common configurations are now added, making the config cleaner and allowing the underlying args to be changed between releases without being a breaking change. It is highly recommended to update the configuration to use these presets.
See https://deploy-preview-4055--frigate-docs.netlify.app/configuration/ffmpeg_presets for more info on how these should be used.
Example: Enabling Audio In Recordings
before:
after:
This also enables presets for hardware acceleration which will be used to further reduce CPU usage by using the GPU to optimize scaling and other workloads.
Improved events filters
Recordings updates
Recording segments are now stored in UTC to avoid issues with DST. In order to avoid a breaking change, the path of the stored segments has changed from
/media/frigate/recordings/%Y-%m/%d/%H-%s.mp4to/media/frigate/recordings/%Y-%m-%d/%H-%s.mp4.Recordings in the frontend are now shown in the timezone of the device viewing frigate, so the timezone set on the server running frigate is irrelevant.
No changes are needed for existing users.
What's Changed
-user_agentforrtmpstreams by @felipecrs in Do not apply-user_agentforrtmpstreams #4596New Contributors
Full Changelog: v0.11.1...v0.12.0-beta1
Images:
blakeblackshear/frigate:0.12.0-beta2ghcr.io/blakeblackshear/frigate:0.12.0-beta2blakeblackshear/frigate:0.12.0-beta2-tensorrtghcr.io/blakeblackshear/frigate:0.12.0-beta2-tensorrtThis discussion was created from the release 0.12.0 Beta 2.
Beta Was this translation helpful? Give feedback.
All reactions