Skip to content
View in the app

A better way to browse. Learn more.

Unraid

A full-screen app on your home screen with push notifications, badges and more.

To install this app on iOS and iPadOS
  1. Tap the Share icon in Safari
  2. Scroll the menu and tap Add to Home Screen.
  3. Tap Add in the top-right corner.
To install this app on Android
  1. Tap the 3-dot menu (⋮) in the top-right corner of the browser.
  2. Tap Add to Home screen or Install app.
  3. Confirm by tapping Install.

[SUPPORT] blakeblackshear - Frigate

Featured Replies

Good morning,

It turns out I'm having trouble getting my Coral M.2 Accelerator A+E key to work. It used to work perfectly through an M.2 adapter, but now that I’ve changed the motherboard (due to a warranty issue), I can’t get it to work anymore.

Here are the details of the change:

Previously, I had this motherboard:

cwwk
No image preview

CWWK N100/i3-N305 six-bay NAS monster board/4x 2.5G/6x SA...

CWWK is committed to developing a reliable, secure, and customizable firewall or VPN appliance for network professionals and enthusiasts.

And now I’ve switched to this one:

https://cwwk.net/collections/nas/products/cwwk-eight-slot-10g-nas-motherboard-n150-n305-n355-dual-2-5g-network-card-10g-10g-port-single-ddr5-dual-nvme-nas-motherboard

The new board has an M.2 E-Key PCIe & CNVi Wi-Fi socket (I’m not sure if it’s supposed to work there).

I’ve tried both that connector and the M.2 slot with the adapter I already had, but neither works — I get the error:

“No PCI-based Coral TPU found!”

I’m attaching pictures of everything.

Thanks!

20251029_164802.jpg

IMG_20251029_221116_887.jpg

IMG_20251030_054142_111.jpg

Edited by Lerk

  • Replies 1.6k
  • Views 571.9k
  • Created
  • Last Reply

Top Posters In This Topic

Most Popular Posts

  • Glodenox
    Glodenox

    I ran into an issue where the Docker container didn't want to start any more after the upgrade to Unraid 7.3.0, with the following error: Error response from daemon: error while creating mount source

  • I released now a Plugin for Unraid 6.9.0rc2 and up so that the installation of the Drivers is as easy as going to the CA App and search for 'Coral-Driver'.   It will take about an hour to sh

  • Hi @mathgoy, I've been using Frigate with an NVIDIA GPU.   to set it up, go the the docker template and add: --rm --runtime=nvidia to the "Extra Parameters".   Y

Posted Images

  • Author
5 hours ago, Lerk said:

Good morning,

It turns out I'm having trouble getting my Coral M.2 Accelerator A+E key to work. It used to work perfectly through an M.2 adapter, but now that I’ve changed the motherboard (due to a warranty issue), I can’t get it to work anymore.

Here are the details of the change:

Previously, I had this motherboard:

cwwk
No image preview

CWWK N100/i3-N305 six-bay NAS monster board/4x 2.5G/6x SA...

CWWK is committed to developing a reliable, secure, and customizable firewall or VPN appliance for network professionals and enthusiasts.

And now I’ve switched to this one:

https://cwwk.net/collections/nas/products/cwwk-eight-slot-10g-nas-motherboard-n150-n305-n355-dual-2-5g-network-card-10g-10g-port-single-ddr5-dual-nvme-nas-motherboard

The new board has an M.2 E-Key PCIe & CNVi Wi-Fi socket (I’m not sure if it’s supposed to work there).

I’ve tried both that connector and the M.2 slot with the adapter I already had, but neither works — I get the error:

“No PCI-based Coral TPU found!”

I’m attaching pictures of everything.

Thanks!

20251029_164802.jpg

IMG_20251029_221116_887.jpg

IMG_20251030_054142_111.jpg

Did you try to unisntall the plugin and install it again?

33 minutes ago, yayitazale said:

Did you try to unisntall the plugin and install it again?

Yes.

The only difference (not sure if this is expected) is that the repository is from Limetech. As far as I understand, it’s the official one and the same I’ve successfully used before.

I’ll need to check if there’s an option in the BIOS to enable PCIe/CVNI.

Recently replaced my Coral A+E m.2 TPU with a Hailo-8L B+M m.2 TPU, which has made Frigate stable, with the TPU itself never going offline in the weeks since. However, today I noticed the following error keeps popping up associated with the kernel, such that when it does it happens numerous times that day, but then it won't happen for at least a week or so. Is this a concern?

Log excerpt here

Nov  2 09:47:26 server kernel: ------------[ cut here ]------------
Nov  2 09:47:26 server kernel: WARNING: CPU: 2 PID: 2683488 at include/linux/rwsem.h:80 find_vma+0x29/0x60
Nov  2 09:47:26 server kernel: Modules linked in: xt_connmark iptable_mangle xt_comment iptable_raw wireguard curve25519_x86_64 libcurve25519_generic libchacha20poly1305 chacha_x86_64 poly1305_x86_64 ip6_udp_tunnel udp_tunnel libchacha udp_diag nft_masq nft_chain_nat nf_conntrack_netlink xt_nat veth xt_conntrack xfrm_user xfrm_algo xt_addrtype dm_crypt dm_mod hailo_pci(O) md_mod zfs(PO) spl(O) xt_MASQUERADE xt_tcpudp xt_mark tun nf_tables nfnetlink ip6table_nat iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ntfs3 tcp_diag inet_diag nct6775 nct6775_core hwmon_vid apex(O) gasket(O) ip6table_filter ip6_tables iptable_filter ip_tables x_tables efivarfs bridge stp llc bonding tls intel_rapl_common x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel i915 kvm iosf_mbi drm_buddy ttm i2c_algo_bit drm_display_helper drm_kms_helper crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel sha512_ssse3 sha256_ssse3 drm sha1_ssse3 aesni_intel crypto_simd cryptd rapl mei_hdcp mei_pxp wmi_bmof intel_wmi_thunderbolt nvme
Nov  2 09:47:26 server kernel: i2c_i801 intel_gtt intel_cstate intel_uncore e1000e i2c_smbus agpgart nvme_core mei_me mei ahci i2c_core libahci intel_pch_thermal video wmi backlight acpi_tad acpi_pad button
Nov  2 09:47:26 server kernel: CPU: 2 UID: 0 PID: 2683488 Comm: frigate.detecto Tainted: P        W  O       6.12.24-Unraid #1
Nov  2 09:47:26 server kernel: Tainted: [P]=PROPRIETARY_MODULE, [W]=WARN, [O]=OOT_MODULE
Nov  2 09:47:26 server kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./Z390 Pro4, BIOS P4.70 01/20/2021
Nov  2 09:47:26 server kernel: RIP: 0010:find_vma+0x29/0x60
Nov  2 09:47:26 server kernel: Code: 90 0f 1f 44 00 00 48 83 ec 10 65 48 8b 04 25 28 00 00 00 48 89 44 24 08 31 c0 48 89 34 24 48 8b 87 a8 00 00 00 48 85 c0 75 02 <0f> 0b 48 83 c7 40 48 83 ca ff 48 89 e6 e8 b5 a6 88 00 48 8b 54 24
Nov  2 09:47:26 server kernel: RSP: 0018:ffffc90024547da8 EFLAGS: 00010246
Nov  2 09:47:26 server kernel: RAX: 0000000000000000 RBX: ffff8881ac335600 RCX: 0000000000000080
Nov  2 09:47:26 server kernel: RDX: 0000000000000070 RSI: 0000148f839e0000 RDI: ffff8881f8bc7900
Nov  2 09:47:26 server kernel: RBP: ffffc90024547e38 R08: 0000000000000dc0 R09: 00000000ffffffff
Nov  2 09:47:26 server kernel: R10: ffff88817ba392f8 R11: 0000148fcc9fb5c0 R12: 0000000000000000
Nov  2 09:47:26 server kernel: R13: 0000148fcc9fb5c0 R14: ffff8881bcdf1100 R15: 0000000000000000
Nov  2 09:47:26 server kernel: FS:  0000148fcc9fd6c0(0000) GS:ffff88884e080000(0000) knlGS:0000000000000000
Nov  2 09:47:26 server kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Nov  2 09:47:26 server kernel: CR2: 0000148fa8153028 CR3: 000000045a59e005 CR4: 00000000003726f0
Nov  2 09:47:26 server kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Nov  2 09:47:26 server kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Nov  2 09:47:26 server kernel: Call Trace:
Nov  2 09:47:26 server kernel: <TASK>
Nov  2 09:47:26 server kernel: hailo_vdma_buffer_map+0xa0/0x590 [hailo_pci]
Nov  2 09:47:26 server kernel: hailo_vdma_buffer_map_ioctl+0x10b/0x210 [hailo_pci]
Nov  2 09:47:26 server kernel: hailo_pcie_fops_unlockedioctl+0x1ed/0x5b0 [hailo_pci]
Nov  2 09:47:26 server kernel: vfs_ioctl+0x1b/0x30
Nov  2 09:47:26 server kernel: __do_sys_ioctl+0x52/0x80
Nov  2 09:47:26 server kernel: do_syscall_64+0x68/0xe0
Nov  2 09:47:26 server kernel: entry_SYSCALL_64_after_hwframe+0x76/0x7e
Nov  2 09:47:26 server kernel: RIP: 0033:0x148fd1fc3d5b
Nov  2 09:47:26 server kernel: Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00
Nov  2 09:47:26 server kernel: RSP: 002b:0000148fcc9fb510 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
Nov  2 09:47:26 server kernel: RAX: ffffffffffffffda RBX: 0000148fcc9fb810 RCX: 0000148fd1fc3d5b
Nov  2 09:47:26 server kernel: RDX: 0000148fcc9fb5c0 RSI: 00000000c0287604 RDI: 0000000000000057
Nov  2 09:47:26 server kernel: RBP: 0000000004e98380 R08: 0000000000000001 R09: 0000000000000000
Nov  2 09:47:26 server kernel: R10: 0000000000001900 R11: 0000000000000246 R12: 0000000004e98410
Nov  2 09:47:26 server kernel: R13: 0000000004e98438 R14: 0000148f839e0000 R15: 0000000004e98380
Nov  2 09:47:26 server kernel: </TASK>

Edited by john_smith

  • Author
On 11/2/2025 at 6:58 PM, john_smith said:

Recently replaced my Coral A+E m.2 TPU with a Hailo-8L B+M m.2 TPU, which has made Frigate stable, with the TPU itself never going offline in the weeks since. However, today I noticed the following error keeps popping up associated with the kernel, such that when it does it happens numerous times that day, but then it won't happen for at least a week or so. Is this a concern?

Log excerpt here

Nov  2 09:47:26 server kernel: ------------[ cut here ]------------
Nov  2 09:47:26 server kernel: WARNING: CPU: 2 PID: 2683488 at include/linux/rwsem.h:80 find_vma+0x29/0x60
Nov  2 09:47:26 server kernel: Modules linked in: xt_connmark iptable_mangle xt_comment iptable_raw wireguard curve25519_x86_64 libcurve25519_generic libchacha20poly1305 chacha_x86_64 poly1305_x86_64 ip6_udp_tunnel udp_tunnel libchacha udp_diag nft_masq nft_chain_nat nf_conntrack_netlink xt_nat veth xt_conntrack xfrm_user xfrm_algo xt_addrtype dm_crypt dm_mod hailo_pci(O) md_mod zfs(PO) spl(O) xt_MASQUERADE xt_tcpudp xt_mark tun nf_tables nfnetlink ip6table_nat iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 ntfs3 tcp_diag inet_diag nct6775 nct6775_core hwmon_vid apex(O) gasket(O) ip6table_filter ip6_tables iptable_filter ip_tables x_tables efivarfs bridge stp llc bonding tls intel_rapl_common x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel i915 kvm iosf_mbi drm_buddy ttm i2c_algo_bit drm_display_helper drm_kms_helper crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel sha512_ssse3 sha256_ssse3 drm sha1_ssse3 aesni_intel crypto_simd cryptd rapl mei_hdcp mei_pxp wmi_bmof intel_wmi_thunderbolt nvme
Nov  2 09:47:26 server kernel: i2c_i801 intel_gtt intel_cstate intel_uncore e1000e i2c_smbus agpgart nvme_core mei_me mei ahci i2c_core libahci intel_pch_thermal video wmi backlight acpi_tad acpi_pad button
Nov  2 09:47:26 server kernel: CPU: 2 UID: 0 PID: 2683488 Comm: frigate.detecto Tainted: P        W  O       6.12.24-Unraid #1
Nov  2 09:47:26 server kernel: Tainted: [P]=PROPRIETARY_MODULE, [W]=WARN, [O]=OOT_MODULE
Nov  2 09:47:26 server kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./Z390 Pro4, BIOS P4.70 01/20/2021
Nov  2 09:47:26 server kernel: RIP: 0010:find_vma+0x29/0x60
Nov  2 09:47:26 server kernel: Code: 90 0f 1f 44 00 00 48 83 ec 10 65 48 8b 04 25 28 00 00 00 48 89 44 24 08 31 c0 48 89 34 24 48 8b 87 a8 00 00 00 48 85 c0 75 02 <0f> 0b 48 83 c7 40 48 83 ca ff 48 89 e6 e8 b5 a6 88 00 48 8b 54 24
Nov  2 09:47:26 server kernel: RSP: 0018:ffffc90024547da8 EFLAGS: 00010246
Nov  2 09:47:26 server kernel: RAX: 0000000000000000 RBX: ffff8881ac335600 RCX: 0000000000000080
Nov  2 09:47:26 server kernel: RDX: 0000000000000070 RSI: 0000148f839e0000 RDI: ffff8881f8bc7900
Nov  2 09:47:26 server kernel: RBP: ffffc90024547e38 R08: 0000000000000dc0 R09: 00000000ffffffff
Nov  2 09:47:26 server kernel: R10: ffff88817ba392f8 R11: 0000148fcc9fb5c0 R12: 0000000000000000
Nov  2 09:47:26 server kernel: R13: 0000148fcc9fb5c0 R14: ffff8881bcdf1100 R15: 0000000000000000
Nov  2 09:47:26 server kernel: FS:  0000148fcc9fd6c0(0000) GS:ffff88884e080000(0000) knlGS:0000000000000000
Nov  2 09:47:26 server kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Nov  2 09:47:26 server kernel: CR2: 0000148fa8153028 CR3: 000000045a59e005 CR4: 00000000003726f0
Nov  2 09:47:26 server kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Nov  2 09:47:26 server kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Nov  2 09:47:26 server kernel: Call Trace:
Nov  2 09:47:26 server kernel: <TASK>
Nov  2 09:47:26 server kernel: hailo_vdma_buffer_map+0xa0/0x590 [hailo_pci]
Nov  2 09:47:26 server kernel: hailo_vdma_buffer_map_ioctl+0x10b/0x210 [hailo_pci]
Nov  2 09:47:26 server kernel: hailo_pcie_fops_unlockedioctl+0x1ed/0x5b0 [hailo_pci]
Nov  2 09:47:26 server kernel: vfs_ioctl+0x1b/0x30
Nov  2 09:47:26 server kernel: __do_sys_ioctl+0x52/0x80
Nov  2 09:47:26 server kernel: do_syscall_64+0x68/0xe0
Nov  2 09:47:26 server kernel: entry_SYSCALL_64_after_hwframe+0x76/0x7e
Nov  2 09:47:26 server kernel: RIP: 0033:0x148fd1fc3d5b
Nov  2 09:47:26 server kernel: Code: 00 48 89 44 24 18 31 c0 48 8d 44 24 60 c7 04 24 10 00 00 00 48 89 44 24 08 48 8d 44 24 20 48 89 44 24 10 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1c 48 8b 44 24 18 64 48 2b 04 25 28 00 00
Nov  2 09:47:26 server kernel: RSP: 002b:0000148fcc9fb510 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
Nov  2 09:47:26 server kernel: RAX: ffffffffffffffda RBX: 0000148fcc9fb810 RCX: 0000148fd1fc3d5b
Nov  2 09:47:26 server kernel: RDX: 0000148fcc9fb5c0 RSI: 00000000c0287604 RDI: 0000000000000057
Nov  2 09:47:26 server kernel: RBP: 0000000004e98380 R08: 0000000000000001 R09: 0000000000000000
Nov  2 09:47:26 server kernel: R10: 0000000000001900 R11: 0000000000000246 R12: 0000000004e98410
Nov  2 09:47:26 server kernel: R13: 0000000004e98438 R14: 0000148f839e0000 R15: 0000000004e98380
Nov  2 09:47:26 server kernel: </TASK>

There is a open discussion for this issue:

https://github.com/blakeblackshear/frigate/discussions/18850#discussioncomment-14717742

  • 4 weeks later...

Looks like Google Coral TPU can run YOLO v9 object detection models. I didnt get it to work, but i think it is more related to unraid and where to place the models. Maybe someone else has more luck: https://github.com/dbro/frigate-detector-edgetpu-yolo9

On reddit several guys confirmed that is working.

  • Author
13 hours ago, sylus said:

Looks like Google Coral TPU can run YOLO v9 object detection models. I didnt get it to work, but i think it is more related to unraid and where to place the models. Maybe someone else has more luck: https://github.com/dbro/frigate-detector-edgetpu-yolo9

On reddit several guys confirmed that is working.

I wasn't aware of this, but since the Merge request has already been accepted, I assume this option will be released as a feature in the next version. Just be patient.

  • 1 month later...

Can someone please point this noob in the right direction? I've bought two Tapo c110 cameras and wanted to integrate them into my unraid server. I've setup Frigate through the community applications and got my two cameras working through rtsp. All great. Then I set up mosquitto and tried to link it to Frigate and my Home Assistant docker container for notifications. I did not get this to work. After trying for a while I found out that there's a blueprint for HA. But this blueprint requires a frigate add on in HA installed through HACS. HA on Unraid doesn't have HACS and now I'm stuck. Can someone point me in the right direction?

12 minutes ago, ontosan said:

Can someone please point this noob in the right direction? I've bought two Tapo c110 cameras and wanted to integrate them into my unraid server. I've setup Frigate through the community applications and got my two cameras working through rtsp. All great. Then I set up mosquitto and tried to link it to Frigate and my Home Assistant docker container for notifications. I did not get this to work. After trying for a while I found out that there's a blueprint for HA. But this blueprint requires a frigate add on in HA installed through HACS. HA on Unraid doesn't have HACS and now I'm stuck. Can someone point me in the right direction?

You didn’t provide many details about how you are running HA on Unraid but the HACS docs cover multiple install methods for different scenarios.

https://www.hacs.xyz/docs/use/download/download/#to-download-hacs

22 hours ago, wgstarks said:

You didn’t provide many details about how you are running HA on Unraid but the HACS docs cover multiple install methods for different scenarios.

https://www.hacs.xyz/docs/use/download/download/#to-download-hacs

Exactly the nudge I needed, I found the HA docs and the Frigate docs but didn't look for HACS docs. Thanks so much, it works! And to be fair, I did mention I run HA as a docker container in Unraid :-)

Edited by ontosan

3 minutes ago, ontosan said:

to be fair, I did mention I run HA as a docker container in Unraid :-)

Yeah, apologies. I just missed it.😊

Hi guys, very sorry if this is a stupid request butr for the love no money can i get Frigate installed on my Unraid server.

Now i download through the apps and leave everything on default apart from deleting the Coral and Nvidia stuff it then installs.

I have to click logs quick to get the username and password which it nevers gives, and i have tried more than 10 times to install and delete with different settings in the install part.

I have looked online for guides (idiot proof ones) and videos with out any success.

Could anybody help a noob out please as im loosing the will to live....

20 hours ago, xtremedetailz said:

Hi guys, very sorry if this is a stupid request butr for the love no money can i get Frigate installed on my Unraid server.

Now i download through the apps and leave everything on default apart from deleting the Coral and Nvidia stuff it then installs.

I have to click logs quick to get the username and password which it nevers gives, and i have tried more than 10 times to install and delete with different settings in the install part.

I have looked online for guides (idiot proof ones) and videos with out any success.

Could anybody help a noob out please as im loosing the will to live....

Have you added a config.yml file in your Frigate folder, usally /mnt/user/appdata/frigate?

  • Author
On 1/6/2026 at 5:46 PM, xtremedetailz said:

Hi guys, very sorry if this is a stupid request butr for the love no money can i get Frigate installed on my Unraid server.

Now i download through the apps and leave everything on default apart from deleting the Coral and Nvidia stuff it then installs.

I have to click logs quick to get the username and password which it nevers gives, and i have tried more than 10 times to install and delete with different settings in the install part.

I have looked online for guides (idiot proof ones) and videos with out any success.

Could anybody help a noob out please as im loosing the will to live....

Please give us some info. Copy and paste de error logs for example

5 hours ago, flosken said:

Have you added a config.yml file in your Frigate folder, usally /mnt/user/appdata/frigate?

52 minutes ago, yayitazale said:

Please give us some info. Copy and paste de error logs for example

Ok guys i have now got it installed and can log in and even find a camera. (had to set static ip i think in the camera)

Now the cameras i have are hikvision 4k ones and for the love nor money understand how to edit the yml so i can see the high resolution for recording 24/7 or even when viewing in the frigate viewer.

I appreciate your helps guys.

When i got to config editor this is what i have.

mqtt:
  enabled: false

cameras:
  name_of_your_camera: # <------ Name the camera
    enabled: true
    ffmpeg:
      inputs:
        - path: rtsp://viewer:[email protected].***:554/streaming/channels/102  # <----- The stream you want to use for detection
          roles:
            - detect
        - path: rtsp://viewer:[email protected].***:554/streaming/channels/101
          roles:
            - record
    detect:
      enabled: false # <---- disable detection until you have a working camera feed
      width: 3840 
      height: 2160
detect:
  enabled: true
version: 0.16-0
  • Author
14 hours ago, xtremedetailz said:

Ok guys i have now got it installed and can log in and even find a camera. (had to set static ip i think in the camera)

Now the cameras i have are hikvision 4k ones and for the love nor money understand how to edit the yml so i can see the high resolution for recording 24/7 or even when viewing in the frigate viewer.

I appreciate your helps guys.

When i got to config editor this is what i have.

mqtt:enabled: false

cameras:name_of_your_camera: # <------ Name the camera
    enabled: true
    ffmpeg:
      inputs:
        - path: rtsp://viewer:[email protected].***:554/streaming/channels/102  # <----- The stream you want to use for detection
          roles:
            - detect
        - path: rtsp://viewer:[email protected].***:554/streaming/channels/101
          roles:
            - record
    detect:
      enabled: false # <---- disable detection until you have a working camera feed
      width: 3840 
      height: 2160detect:enabled: trueversion: 0.16-0

Are you following the getting started?

https://docs.frigate.video/guides/getting_started#configuring-frigate

Are you looking in the logs for errors?

  • 1 month later...

hoping for some help as I can not get my GPU ( Nvidia 4060) workingfrigate-logs.txt

When I installed frigate it was using the GPU, then after adding a few camera it just sits idle, and only uses the CPU

Is there something I need to change in the config file?

image.png

  • Author
22 hours ago, rbf351 said:

hoping for some help as I can not get my GPU ( Nvidia 4060) workingfrigate-logs.txt

When I installed frigate it was using the GPU, then after adding a few camera it just sits idle, and only uses the CPU

Is there something I need to change in the config file?

image.png

With such little info is not easy to help you.

...

1 hour ago, yayitazale said:

With such little info is not easy to help you.

...

I'm using it for both, or at least that it what I thought I was doing

I'm using the ghcr.io/blakeblackshear/frigate:stable-tensorrt branch

looks like I was missing

ffmpeg:
hwaccel_args: preset-nvidia from the config file

now I am seeing this when running nvidia-smi

image.png

image.png

I am unable to get the detector to function

when i put this line into the config file

detectors:
tensorrt:
type: tensorrt
device: 0
#This is the default, select the first GPU

Frigate crashes.

Edited by rbf351

  • Author
7 minutes ago, rbf351 said:

I'm using it for both, or at least that it what I thought I was doing

I'm using the ghcr.io/blakeblackshear/frigate:stable-tensorrt branch

looks like I was missing

ffmpeg:
hwaccel_args: preset-nvidia from the config file

now I am seeing this when running nvidia-smi

image.png

image.png

I am unable to get the detector to function

when i put this line into the config file

detectors:
tensorrt:
type: tensorrt
device: 0
#This is the default, select the first GPU

Frigate crashes.

You have configure onnx https://docs.frigate.video/configuration/object_detectors/#onnx

31 minutes ago, yayitazale said:

I ran the command for YOLOv9 for other detectors. It completed with no issues, but there is no /config/model_cache/yolo.onnx file.

the script says exporting to client directory, but where exactly is that?

  • Author
9 minutes ago, rbf351 said:

I ran the command for YOLOv9 for other detectors. It completed with no issues, but there is no /config/model_cache/yolo.onnx file.

the script says exporting to client directory, but where exactly is that?

The same Path where you are running the script I suppose

Just now, yayitazale said:

The same Path where you are running the script I suppose

just found it in the root directory, and moved it to where it needs to be

seems like it's now running properly.

Thank you

  • 4 weeks later...

Hello First I LOVE Frigate and if i cant fix this i ok with it. I hope this is a good place to ask .. been running frigate for some time with no problems all the ways up to v17 now and i seem to be having a problem. the screen as you are watching live only shows 1 ever 2 or 3 sec. nothing has changed here that i know of. Frigate is running on a Server as i UN-Raid server :

Supermicro X9QR7-TF+/X9QRi-F+ , Version 123456789
American Megatrends Inc, Version 3.0a
BIOS dated: Thu 3 Apr 2014 12:00 AM

Processors 4 of Intel® Xeon® CPU E5-4650 v2 @ 2.40GHz

Memory: 128 GiB DDR3 Multi-bit ECC

no Video Card but i have a Coral TPU PCI based TPU installed: 1

detectors:
  coral:
    type: edgetpu
    device: pci:0
    classify: true

the only thing that is running is MQTT server and Plex (No one is using plex but me and it make no differences that i can see )

I only have 3 cameras. I don't under understand why it is so slow all play back is at full speed (even with Audio) the cameras are all POE. They are all on there own wired network with nothing but cameras and Un-Raid 2nd port on the server set to the same address group
Here is the Config From Frigate:

thanks for any help .. I am hoping some one see the problem i cant

Frigat_config.txt

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

Account

Navigation

Search

Search

Configure browser push notifications

Chrome (Android)
  1. Tap the lock icon next to the address bar.
  2. Tap Permissions → Notifications.
  3. Adjust your preference.
Chrome (Desktop)
  1. Click the padlock icon in the address bar.
  2. Select Site settings.
  3. Find Notifications and adjust your preference.