[SUPPORT] blakeblackshear - Frigate

[yayitazale]

2 hours ago, mikey6283 said:

I aming have problem starting Frigate NV after installing Tapo cameras. The streams work on VLC , however when i add a rtsp  i receive errors in config.yml.

Could some advise where the error is in config. please see logs and current logs;

 

config.ymlconfig.yml 2.06 kB · 2 downloads _frigate_logs (1).txt.crdownload 26.97 kB · 1 download

Any help would be appreciated as i have looking for days and i must be missing something.

 

 

Not sure about your problem but you have something that you can clean in the config file:

 

koipond:
        - ffmpeg:rtsp://admin1:[email protected]:554/stream1#video=copy#audio=copy#audio=copy#audio=aac

 

In this case you can do this as you are only transcoding audio:
 

koipond:
        - rtsp://admin1:[email protected]:554/stream1
        - ffmpeg:koipond#audio=aac

 

Then here:

 

koipond:
      ffmpeg:
        output_args:
          record: preset-record-generic-audio-copy
        inputs:
          - path: rtsp://127.0.0.1:8554/koipond
            input_args: preset-rtsp-restream
            roles:
          - path: rtsp://127.0.0.1:8554/koipond_sub
              - record
            input_args: preset-rtsp-restream
            roles:
              - detect

 

If you are using the same input/ouput args for all the cameras you can add the ffmpeg part as a general block instead of defining it in every camera:

 

ffmpeg:
  input_args: preset-rtsp-restream
  output_args:
    record: preset-record-generic-audio-copy

 

The logs error just means frigate is not online yet and the integration requested the stats, you should have some other error somewhere.

[mikey6283]

15 hours ago, yayitazale said:

Not sure about your problem but you have something that you can clean in the config file:

 

koipond:
        - ffmpeg:rtsp://admin1:[email protected]:554/stream1#video=copy#audio=copy#audio=copy#audio=aac

 

In this case you can do this as you are only transcoding audio:
 

koipond:
        - rtsp://admin1:[email protected]:554/stream1
        - ffmpeg:koipond#audio=aac

 

Then here:

 

koipond:
      ffmpeg:
        output_args:
          record: preset-record-generic-audio-copy
        inputs:
          - path: rtsp://127.0.0.1:8554/koipond
            input_args: preset-rtsp-restream
            roles:
          - path: rtsp://127.0.0.1:8554/koipond_sub
              - record
            input_args: preset-rtsp-restream
            roles:
              - detect

 

If you are using the same input/ouput args for all the cameras you can add the ffmpeg part as a general block instead of defining it in every camera:

 

ffmpeg:
  input_args: preset-rtsp-restream
  output_args:
    record: preset-record-generic-audio-copy

 

The logs error just means frigate is not online yet and the integration requested the stats, you should have some other error somewhere.

I have 4 cameras a reolink and 3 Tapo RTSP , The Reolink  is  displayed in HA overview however the 1 tapo isn't  . all cameras are viewable VLC . the ffmeg block , do you add this before  go2RTC ?

 

 In the errors I do not understand  how to rectify the issue to get  Frigate working? Any advice ?

config error.15.06.24 config.yml

[yayitazale]

On 6/15/2024 at 12:20 PM, mikey6283 said:

I have 4 cameras a reolink and 3 Tapo RTSP , The Reolink  is  displayed in HA overview however the 1 tapo isn't  . all cameras are viewable VLC . the ffmeg block , do you add this before  go2RTC ?

 

 In the errors I do not understand  how to rectify the issue to get  Frigate working? Any advice ?

config error.15.06.24 7.38 kB · 0 downloads config.yml 1.98 kB · 0 downloads

What I think you should use.

config(4).yml

[NCM]

Is it possible to use the same PCI TPU between 2 Frigate containers or am I going to need to get a second TPU? Trying to run a second container with the beta release (eventually I'd like to run a Frigate container for a second premises) but the logs make it look like the TPU isn't detected although it's set up identically as the stable container.

[yayitazale]

2 hours ago, NCM said:

Is it possible to use the same PCI TPU between 2 Frigate containers or am I going to need to get a second TPU? Trying to run a second container with the beta release (eventually I'd like to run a Frigate container for a second premises) but the logs make it look like the TPU isn't detected although it's set up identically as the stable container.

No, it is not possible to use it at the same time. What I do is just stop the old one during the testing of the beta and if anything goes wrong I just go back.

[mikey6283]

On 6/16/2024 at 4:52 PM, yayitazale said:

What I think you should use.

config(4).yml 1.8 kB · 7 downloads

 Thanks for advice, the problem was HA/FVR addon so in intailed  HA again, and new addon, this solved the problem. 

[dynamo666]

I am noticing a problem every time I export a recording from the webUI. After I select a time, let's say from camera:livingroom time:10pm to 11pm on a camera, the recording is stored with a file name that is 3 hours behind, i.e. the livingroom_7pm-8pm.mp4. Any way to resolve this issue? Noticed this for every file name. Thanks.

[yayitazale]

On 6/29/2024 at 7:48 AM, dynamo666 said:

I am noticing a problem every time I export a recording from the webUI. After I select a time, let's say from camera:livingroom time:10pm to 11pm on a camera, the recording is stored with a file name that is 3 hours behind, i.e. the livingroom_7pm-8pm.mp4. Any way to resolve this issue? Noticed this for every file name. Thanks.

Did you passthrought /etc/localtime to the container?

[TurboStreetCar]

Hello, i seem to be having a problem with scheduled trim.

When the trim runs, i get a bunch of errors about the drive having insufficient space, and it sometimes causes the container to lock up. 

Ive tried turning off auto trim in the drive settings, but it seems to be when the scheduled trim runs in the scheduler settings.

Has anyone had an issue with this? I searched for "trim" and found zero results in this thread.

[yayitazale]

7 minutes ago, TurboStreetCar said:

Hello, i seem to be having a problem with scheduled trim.

When the trim runs, i get a bunch of errors about the drive having insufficient space, and it sometimes causes the container to lock up. 

Ive tried turning off auto trim in the drive settings, but it seems to be when the scheduled trim runs in the scheduler settings.

Has anyone had an issue with this? I searched for "trim" and found zero results in this thread.

I have autotrim enabled and I have been using Frigate for many years now and I didn't ever experience that.

[TurboStreetCar]

16 minutes ago, yayitazale said:

I have autotrim enabled and I have been using Frigate for many years now and I didn't ever experience that.

This log is from last week, But i had Trim in the scheduler set for monday 4AM, and every monday morning, i would have these errors. This week, i changed the trim in scheduler to tuesday, and have no errors.

In the frigate docker log, you can see it shut down for appdata-backup at 3:00, and then it starts having drive space errors at 4:57, which i belive is when the trim is acting upon the CCTV drive frigate is recording to.

In the unraid syslog, you can see the trim completes at 6:41, which is when all of the disk space errors in the frigate log abruptly stop. 

Im not sure if i can somehow prevent the scheduled trim from trimming the CCTV drive?

My CCTV drive is set up as a pool and is not on the array. Maybe im missing something in its configuration?

frigate.log Unraid syslog.log

[yayitazale]

22 hours ago, TurboStreetCar said:

This log is from last week, But i had Trim in the scheduler set for monday 4AM, and every monday morning, i would have these errors. This week, i changed the trim in scheduler to tuesday, and have no errors.

In the frigate docker log, you can see it shut down for appdata-backup at 3:00, and then it starts having drive space errors at 4:57, which i belive is when the trim is acting upon the CCTV drive frigate is recording to.

In the unraid syslog, you can see the trim completes at 6:41, which is when all of the disk space errors in the frigate log abruptly stop. 

Im not sure if i can somehow prevent the scheduled trim from trimming the CCTV drive?

My CCTV drive is set up as a pool and is not on the array. Maybe im missing something in its configuration?

frigate.log 769.55 kB · 1 download Unraid syslog.log 564 B · 0 downloads

Not sure but this is out of the scope of this topic, you should open a new topic in https://forums.unraid.net/forum/55-general-support/ as this is not a error of the container deployment neither the template. I don't have the skills to help you with this.

[TurboStreetCar]

2 hours ago, yayitazale said:

Not sure but this is out of the scope of this topic, you should open a new topic in https://forums.unraid.net/forum/55-general-support/ as this is not a error of the container deployment neither the template. I don't have the skills to help you with this.

Gotcha, just made a post.

Figured id post here as i would assume the scheduled trim would be something most people use and seems to cause a conflict. 

Are you running a scheduled trim?

[awediohead]

re-writing original question to express myself better:

 

So having banged my head against a brick wall for quite a few weeks (time and health allowing) trying to figure out why I was getting "Playback cannot continue. No available working or supported playlists." errors - when the live streams were viewable and the saved recordings were viewable via VLC once downloaded off my server - I finally tried a config that was not working in Firefox or Waterfox (in terms of my being able to watch recordings) with Safari and then Brave browsers - where it worked perfectly.

 

I would really like to understand why? Is this a known problem?

 

Secondly, having tried to find answers to this for weeks now with various search engines, reading the Frigate docs and threads here and elsewhere, I think it's understandable that I've assumed all along that there was something very wrong with my config for such a basic functionality to be broken.  We're talking a HUGE amount of time totally wasted trying to fix something that wasn't broken in the first place. At least now I can proceed with setting up my config - but I am concerned that others maybe in the same boat and just give up on Frigate entirely - maybe there should be some sort of pinned warning not to use Mozilla browsers? Or am I missing something?

Thanks

 

Edited July 2 by awediohead
re-written

[Bruceflix]

15 hours ago, TurboStreetCar said:

Gotcha, just made a post.

Figured id post here as i would assume the scheduled trim would be something most people use and seems to cause a conflict. 

Are you running a scheduled trim?

Most people wouldn't be using a SSD for CCTV recordings

[yayitazale]

17 hours ago, awediohead said:

re-writing original question to express myself better:

 

So having banged my head against a brick wall for quite a few weeks (time and health allowing) trying to figure out why I was getting "Playback cannot continue. No available working or supported playlists." errors - when the live streams were viewable and the saved recordings were viewable via VLC once downloaded off my server - I finally tried a config that was not working in Firefox or Waterfox (in terms of my being able to watch recordings) with Safari and then Brave browsers - where it worked perfectly.

 

I would really like to understand why? Is this a known problem?

 

Secondly, having tried to find answers to this for weeks now with various search engines, reading the Frigate docs and threads here and elsewhere, I think it's understandable that I've assumed all along that there was something very wrong with my config for such a basic functionality to be broken.  We're talking a HUGE amount of time totally wasted trying to fix something that wasn't broken in the first place. At least now I can proceed with setting up my config - but I am concerned that others maybe in the same boat and just give up on Frigate entirely - maybe there should be some sort of pinned warning not to use Mozilla browsers? Or am I missing something?

Thanks

 

Hi this is not the forum to blame on any development bug. In any case, just with a little search in the github you can see that the problem is in the Browser side, having issues with h.265:

 

https://github.com/blakeblackshear/frigate/issues/9696

https://github.com/blakeblackshear/frigate/issues/9101

 

 

[TurboStreetCar]

9 hours ago, Bruceflix said:

Most people wouldn't be using a SSD for CCTV recordings

Im using a WD Purple HDD.

It apparently supports trim, but this may be a bug in the scheduled trim function of unraid, as the script checks for "spinning disks", but scheduled trim seems to operate differently then the script in the dynamix folder.

[awediohead]

6 hours ago, yayitazale said:

Hi this is not the forum to blame on any development bug. In any case, just with a little search in the github you can see that the problem is in the Browser side, having issues with h.265:

 

https://github.com/blakeblackshear/frigate/issues/9696

https://github.com/blakeblackshear/frigate/issues/9101

 

 

The problem can happen with h264 or h265 or any combination of the two in Firefox based browsers. I tested this with multiple changes and reversions of my config switching between h264 and h265.

In all cases when it wouldn't work in Firefox when the same Frigate WEBUI is accessed via Brave (and presumably Chrome) or Safari  - it "just works", indicating that there is a fault with Firefox.

 

It was the threads you linked to - plus others - that first made me consider that it might be a browser problem and not a Frigate config or camera problem - both of which were reasonable assumptions I would have thought.

 

I doubt very much that it would occur to many regular Unraid users (as opposed to devs and professional IT guys) that the browser they've been using to administrate Unraid for years without issue, is the reason their Frigate docker appears not to work!

 

Obviously this is not the place to report bugs or inadequacies with Firefox !!

 

I just hoped my sharing my experience might prevent other Firefox users trying to set up Frigate on Unraid from wasting as much time as I have. IOW just trying to help!

[yayitazale]

On 7/3/2024 at 4:48 PM, awediohead said:

The problem can happen with h264 or h265 or any combination of the two in Firefox based browsers. I tested this with multiple changes and reversions of my config switching between h264 and h265.

In all cases when it wouldn't work in Firefox when the same Frigate WEBUI is accessed via Brave (and presumably Chrome) or Safari  - it "just works", indicating that there is a fault with Firefox.

 

It was the threads you linked to - plus others - that first made me consider that it might be a browser problem and not a Frigate config or camera problem - both of which were reasonable assumptions I would have thought.

 

I doubt very much that it would occur to many regular Unraid users (as opposed to devs and professional IT guys) that the browser they've been using to administrate Unraid for years without issue, is the reason their Frigate docker appears not to work!

 

Obviously this is not the place to report bugs or inadequacies with Firefox !!

 

I just hoped my sharing my experience might prevent other Firefox users trying to set up Frigate on Unraid from wasting as much time as I have. IOW just trying to help!

I use Firefox and I didn't experience any issue at all...

[awediohead]

15 hours ago, yayitazale said:

I use Firefox and I didn't experience any issue at all...

Ok - fair enough - so any idea how to explain my experience? Sorry but having just spent literally weeks trying to 'fix' my config only to discover that all I had to do to make it work reliably was to change my browser, I'm extremely confused - but at the same time would really like to understand?

[dynamo666]

On 7/1/2024 at 1:53 AM, yayitazale said:

Did you passthrought /etc/localtime to the container?

Yes. Did not change the container setting under Unraid frigate docker setting. Is there another place where I have to pass it? I can select the correct time from Export screen, but the output .mp4 filename is always 3 hours behind.

[mikey6283]

Hi I hav ebeen struggling for a while to  Frigate up and running.  I have a new Frigate , and new coral tPU module. Config is good,  The issue is that  i am not getting any images.

 

**

2024-07-16 13:15:29.901495671  [2024-07-16 13:15:29] ffmpeg.front.detect            ERROR   : [rtsp @ 0x556432b99c40] method DESCRIBE failed: 404 Not Found
2024-07-16 13:15:29.901580458  [2024-07-16 13:15:29] ffmpeg.front.detect            ERROR   : rtsp://127.0.0.1:8554/front_sub?video=copy: Server returned 404 Not Found
2024-07-16 13:15:29.906455098  [2024-07-16 13:15:29] watchdog.koipond               ERROR   : Ffmpeg process crashed unexpectedly for koipond.
2024-07-16 13:15:29.906537091  [2024-07-16 13:15:29] watchdog.garage                ERROR   : Ffmpeg process crashed unexpectedly for garage.
2024-07-16 13:15:29.906595758  [2024-07-16 13:15:29] watchdog.koipond               ERROR   : The following ffmpeg logs include the last 100 lines prior to exit.
2024-07-16 13:15:29.906647091  [2024-07-16 13:15:29] watchdog.garage                ERROR   : The following ffmpeg logs include the last 100 lines prior to exit.
2024-07-16 13:15:29.906700240  [2024-07-16 13:15:29] ffmpeg.koipond.detect          ERROR   : [rtsp @ 0x5625c2cbcc40] method DESCRIBE failed: 404 Not Found
2024-07-16 13:15:29.906748012  [2024-07-16 13:15:29] ffmpeg.garage.detect           ERROR   : [rtsp @ 0x55aababe4c40] method DESCRIBE failed: 404 Not Found
2024-07-16 13:15:29.906793269  [2024-07-16 13:15:29] ffmpeg.koipond.detect          ERROR   : rtsp://127.0.0.1:8554/koipond_sub?video=copy: Server returned 404 Not Found
2024-07-16 13:15:29.906840551  [2024-07-16 13:15:29] ffmpeg.garage.detect           ERROR   : rtsp://127.0.0.1:8554/BackGarden_sub?video=copy: Server returned 404 Not Found
2024-07-16 13:15:29.911296841  [2024-07-16 13:15:29] watchdog.garage                ERROR   : No new recording segments were created for garage in the last 120s. restarting the ffmpeg record process...
2024-07-16 13:15:29.911394410  [2024-07-16 13:15:29] watchdog.garage                INFO    : Terminating the existing ffmpeg process...
2024-07-16 13:15:29.911438060  [2024-07-16 13:15:29] watchdog.garage                INFO    : Waiting for ffmpeg to exit gracefully...
2024-07-16 13:15:29.911494422  [2024-07-16 13:15:29] watchdog.koipond               ERROR   : No new recording segments were created for koipond in the last 120s. restarting the ffmpeg record process...
2024-07-16 13:15:29.911543032  [2024-07-16 13:15:29] watchdog.koipond               INFO    : Terminating the existing ffmpeg process...
2024-07-16 13:15:29.911598276  [2024-07-16 13:15:29] watchdog.koipond               INFO    : Waiting for ffmpeg to exit gracefully...
2024-07-16 13:15:29.914677646  [2024-07-16 13:15:29] watchdog.BackGarden            ERROR   : Ffmpeg process crashed unexpectedly for BackGarden.
2024-07-16 13:15:29.914754402  [2024-07-16 13:15:29] watchdog.BackGarden            ERROR   : The following ffmpeg logs i**

 I noticed that the  Coral TPU is found by unRaid but the the Figate HA addon. I have also tried to upgrade the go2rtc without any luck . New 1.9.2 upgrade is in HA config ( as go2rtc.zip) i hvave tried to change the name to go2trc(without the zip but it just revertsbag zip. as suggested.

***************

rames from ffmpeg process. 2024-07-16 12:55:37.385177591 [2024-07-16 12:55:37] frigate.video ERROR : garage: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:37.391101789 [2024-07-16 12:55:37] frigate.detectors.plugins.edgetpu_tfl INFO : Attempting to load TPU as pci:0 2024-07-16 12:55:37.397651347 [2024-07-16 12:55:37] frigate.detectors.plugins.edgetpu_tfl INFO : TPU found 2024-07-16 12:55:37.441445452 [2024-07-16 12:55:37] frigate.video ERROR : BackGarden: Unable to read frames from ffmpeg process. 2024-07-16 12:55:37.441511320 [2024-07-16 12:55:37] frigate.video ERROR : BackGarden: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:37.596006068 [2024-07-16 12:55:37] frigate.video ERROR : koipond: Unable to read frames from ffmpeg process. 2024-07-16 12:55:37.596091494 [2024-07-16 12:55:37] frigate.video ERROR : koipond: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:42.402977920 [2024-07-16 12:55:42] frigate.video ERROR : front: Unable to read frames from ffmpeg process. 2024-07-16 12:55:42.403032052 [2024-07-16 12:55:42] frigate.video ERROR : front: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:57.319134966 [2024-07-16 12:55:57] watchdog.front ERROR : Ffmpeg process crashed unexpectedly for front. 2024-07-16 12:55:57.319244204 [2024-07-16 12:55:57] watchdog.front ERROR : The following ffmpeg logs include the last 100 lines prior to exit. 2024-07-16 12:55:57.319387807 [2024-07-16 12:55:57] ffmpeg.front.detect ERROR : [rtsp @ 0x55fdc3112c40]

*******

 

I very much appreciated some help.

[mikey6283]

 Hi

 

 I have m2 Pci Coral TPU installed on unraid . which is detected , however it is not detected in the Frigate addon.

 

error mesaaage

024-07-17 09:06:46.470367261 [2024-07-17 10:06:46] frigate.detectors.plugins.edgetpu_tfl INFO : Attempting to load TPU as pci:0 2024-07-17 09:06:46.474550830 [2024-07-17 10:06:46] frigate.app INFO : Output process started: 260 2024-07-17 09:06:46.488883343 [2024-07-17 10:06:46] frigate.detectors.plugins.edgetpu_tfl ERROR : No EdgeTPU was detected. If you do not have a Coral device yet, you must configure CPU detectors. 2024-07-17 09:06:46.489352421 Process detector:cora1:

 

Any way to solve this , as this could be reason why i am not able to get Frigate running.

[yayitazale]

21 hours ago, mikey6283 said:

Hi I hav ebeen struggling for a while to  Frigate up and running.  I have a new Frigate , and new coral tPU module. Config is good,  The issue is that  i am not getting any images.

 

**

2024-07-16 13:15:29.901495671  [2024-07-16 13:15:29] ffmpeg.front.detect            ERROR   : [rtsp @ 0x556432b99c40] method DESCRIBE failed: 404 Not Found
2024-07-16 13:15:29.901580458  [2024-07-16 13:15:29] ffmpeg.front.detect            ERROR   : rtsp://127.0.0.1:8554/front_sub?video=copy: Server returned 404 Not Found
2024-07-16 13:15:29.906455098  [2024-07-16 13:15:29] watchdog.koipond               ERROR   : Ffmpeg process crashed unexpectedly for koipond.
2024-07-16 13:15:29.906537091  [2024-07-16 13:15:29] watchdog.garage                ERROR   : Ffmpeg process crashed unexpectedly for garage.
2024-07-16 13:15:29.906595758  [2024-07-16 13:15:29] watchdog.koipond               ERROR   : The following ffmpeg logs include the last 100 lines prior to exit.
2024-07-16 13:15:29.906647091  [2024-07-16 13:15:29] watchdog.garage                ERROR   : The following ffmpeg logs include the last 100 lines prior to exit.
2024-07-16 13:15:29.906700240  [2024-07-16 13:15:29] ffmpeg.koipond.detect          ERROR   : [rtsp @ 0x5625c2cbcc40] method DESCRIBE failed: 404 Not Found
2024-07-16 13:15:29.906748012  [2024-07-16 13:15:29] ffmpeg.garage.detect           ERROR   : [rtsp @ 0x55aababe4c40] method DESCRIBE failed: 404 Not Found
2024-07-16 13:15:29.906793269  [2024-07-16 13:15:29] ffmpeg.koipond.detect          ERROR   : rtsp://127.0.0.1:8554/koipond_sub?video=copy: Server returned 404 Not Found
2024-07-16 13:15:29.906840551  [2024-07-16 13:15:29] ffmpeg.garage.detect           ERROR   : rtsp://127.0.0.1:8554/BackGarden_sub?video=copy: Server returned 404 Not Found
2024-07-16 13:15:29.911296841  [2024-07-16 13:15:29] watchdog.garage                ERROR   : No new recording segments were created for garage in the last 120s. restarting the ffmpeg record process...
2024-07-16 13:15:29.911394410  [2024-07-16 13:15:29] watchdog.garage                INFO    : Terminating the existing ffmpeg process...
2024-07-16 13:15:29.911438060  [2024-07-16 13:15:29] watchdog.garage                INFO    : Waiting for ffmpeg to exit gracefully...
2024-07-16 13:15:29.911494422  [2024-07-16 13:15:29] watchdog.koipond               ERROR   : No new recording segments were created for koipond in the last 120s. restarting the ffmpeg record process...
2024-07-16 13:15:29.911543032  [2024-07-16 13:15:29] watchdog.koipond               INFO    : Terminating the existing ffmpeg process...
2024-07-16 13:15:29.911598276  [2024-07-16 13:15:29] watchdog.koipond               INFO    : Waiting for ffmpeg to exit gracefully...
2024-07-16 13:15:29.914677646  [2024-07-16 13:15:29] watchdog.BackGarden            ERROR   : Ffmpeg process crashed unexpectedly for BackGarden.
2024-07-16 13:15:29.914754402  [2024-07-16 13:15:29] watchdog.BackGarden            ERROR   : The following ffmpeg logs i**

 I noticed that the  Coral TPU is found by unRaid but the the Figate HA addon. I have also tried to upgrade the go2rtc without any luck . New 1.9.2 upgrade is in HA config ( as go2rtc.zip) i hvave tried to change the name to go2trc(without the zip but it just revertsbag zip. as suggested.

***************

rames from ffmpeg process. 2024-07-16 12:55:37.385177591 [2024-07-16 12:55:37] frigate.video ERROR : garage: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:37.391101789 [2024-07-16 12:55:37] frigate.detectors.plugins.edgetpu_tfl INFO : Attempting to load TPU as pci:0 2024-07-16 12:55:37.397651347 [2024-07-16 12:55:37] frigate.detectors.plugins.edgetpu_tfl INFO : TPU found 2024-07-16 12:55:37.441445452 [2024-07-16 12:55:37] frigate.video ERROR : BackGarden: Unable to read frames from ffmpeg process. 2024-07-16 12:55:37.441511320 [2024-07-16 12:55:37] frigate.video ERROR : BackGarden: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:37.596006068 [2024-07-16 12:55:37] frigate.video ERROR : koipond: Unable to read frames from ffmpeg process. 2024-07-16 12:55:37.596091494 [2024-07-16 12:55:37] frigate.video ERROR : koipond: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:42.402977920 [2024-07-16 12:55:42] frigate.video ERROR : front: Unable to read frames from ffmpeg process. 2024-07-16 12:55:42.403032052 [2024-07-16 12:55:42] frigate.video ERROR : front: ffmpeg process is not running. exiting capture thread... 2024-07-16 12:55:57.319134966 [2024-07-16 12:55:57] watchdog.front ERROR : Ffmpeg process crashed unexpectedly for front. 2024-07-16 12:55:57.319244204 [2024-07-16 12:55:57] watchdog.front ERROR : The following ffmpeg logs include the last 100 lines prior to exit. 2024-07-16 12:55:57.319387807 [2024-07-16 12:55:57] ffmpeg.front.detect ERROR : [rtsp @ 0x55fdc3112c40]

*******

 

I very much appreciated some help.

You have something wrong in the config file, you are using restreamed links but is not finding ffmpeg is not finding them

[yayitazale]

22 minutes ago, mikey6283 said:

 Hi

 

 I have m2 Pci Coral TPU installed on unraid . which is detected , however it is not detected in the Frigate addon.

 

error mesaaage

024-07-17 09:06:46.470367261 [2024-07-17 10:06:46] frigate.detectors.plugins.edgetpu_tfl INFO : Attempting to load TPU as pci:0 2024-07-17 09:06:46.474550830 [2024-07-17 10:06:46] frigate.app INFO : Output process started: 260 2024-07-17 09:06:46.488883343 [2024-07-17 10:06:46] frigate.detectors.plugins.edgetpu_tfl ERROR : No EdgeTPU was detected. If you do not have a Coral device yet, you must configure CPU detectors. 2024-07-17 09:06:46.489352421 Process detector:cora1:

 

Any way to solve this , as this could be reason why i am not able to get Frigate running.

I don't know how are you launching the container, but if it is found in unraid, you just need to pass-through it using the proper entry in the template.