PVE+ vGPU Tesla P4 + Frigate 的简单尝试

hood2000

crazychen 发表于 2024-4-13 07:24
请问您ubuntu用的是哪个版本呀,还是没找到问题出在哪里

ubuntu 22.04
我觉得不应该是系统版本的问题，试试 chmod 777 config.yml  ,看看是否读写权限的问题。
或者把config.yml这行注释掉试试。

另外，每次是否重新删除container重新创建？

新生活

厉害了，学习，谢谢分享，mark

crazychen

hood2000 发表于 2024-4-13 20:45
ubuntu 22.04
我觉得不应该是系统版本的问题，试试 chmod 777 config.yml  ,看看是否读写权限的问题。
或 ...

我把ront目录下的frigate移动到home下重新拉的，哪个错误没了，但又有新的

ensive_goodall
crazychen@ubuntu2:/home/frigate$ sudo docker logs 5a887d19d108
s6-rc: info: service s6rc-fdholder: starting
s6-rc: info: service s6rc-oneshot-runner: starting
s6-rc: info: service s6rc-oneshot-runner successfully started
s6-rc: info: service fix-attrs: starting
s6-rc: info: service s6rc-fdholder successfully started
s6-rc: info: service fix-attrs successfully started
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service trt-model-prepare: starting
s6-rc: info: service log-prepare: starting
s6-rc: info: service log-prepare successfully started
s6-rc: info: service nginx-log: starting
Generating the following TRT Models: yolov7-tiny-288
s6-rc: info: service go2rtc-log: starting
Downloading yolo weights
s6-rc: info: service frigate-log: starting
s6-rc: info: service nginx-log successfully started
s6-rc: info: service go2rtc-log successfully started
s6-rc: info: service go2rtc: starting
s6-rc: info: service frigate-log successfully started
s6-rc: info: service go2rtc successfully started
s6-rc: info: service go2rtc-healthcheck: starting
2024-04-13 22:25:25.489638044  [INFO] Preparing new go2rtc config...
s6-rc: info: service go2rtc-healthcheck successfully started
s6-rc: warning: unable to start service trt-model-prepare: command exited 4
2024-04-13 22:25:25.738726543  [INFO] Starting go2rtc...
2024-04-13 22:25:25.820817678  22:25:25.820 INF go2rtc version 1.8.4 linux/amd64
2024-04-13 22:25:25.821397255  22:25:25.821 INF [api] listen addr=:1984
2024-04-13 22:25:25.821613666  22:25:25.821 INF [rtsp] listen addr=:8554
2024-04-13 22:25:25.821820680  22:25:25.821 INF [webrtc] listen addr=:8555
2024-04-13 22:25:35.494943638  [INFO] Starting go2rtc healthcheck service...

crazychen

最后一顿胡倒腾，总算打开后台页面了，但ffmpeg 出错


2024-04-14 00:46:30.404408461  Traceback (most recent call last):
2024-04-14 00:46:30.404411187    File "/usr/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
2024-04-14 00:46:30.404415765      self.run()
2024-04-14 00:46:30.404418310    File "/usr/lib/python3.9/multiprocessing/process.py", line 108, in run
2024-04-14 00:46:30.404420073      self._target(*self._args, **self._kwargs)
2024-04-14 00:46:30.404421658    File "/opt/frigate/frigate/object_detection.py", line 102, in run_detector
2024-04-14 00:46:30.404456532      object_detector = LocalObjectDetector(detector_config=detector_config)
2024-04-14 00:46:30.404458432    File "/opt/frigate/frigate/object_detection.py", line 53, in __init__
2024-04-14 00:46:30.404459811      self.detect_api = create_detector(detector_config)
2024-04-14 00:46:30.404461218    File "/opt/frigate/frigate/detectors/__init__.py", line 18, in create_detector
2024-04-14 00:46:30.404492962      return api(detector_config)
2024-04-14 00:46:30.404494525    File "/opt/frigate/frigate/detectors/plugins/tensorrt.py", line 220, in __init__
2024-04-14 00:46:30.404496058      self.engine = self._load_engine(detector_config.model.path)
2024-04-14 00:46:30.404497547    File "/opt/frigate/frigate/detectors/plugins/tensorrt.py", line 88, in _load_engine
2024-04-14 00:46:30.404507829      with open(model_path, "rb") as f, trt.Runtime(self.trt_logger) as runtime:
2024-04-14 00:46:30.404527297  FileNotFoundError: [Errno 2] No such file or directory: '/home/frigate/model_cache/tensorrt/8.5.3/yolov5nu.trt'
2024-04-14 00:46:30.404528824  Exception ignored in: <function TensorRtDetector.__del__ at 0x7fc3637b3d30>
2024-04-14 00:46:30.404530065  Traceback (most recent call last):
2024-04-14 00:46:30.404531544    File "/opt/frigate/frigate/detectors/plugins/tensorrt.py", line 239, in __del__
2024-04-14 00:46:30.404656008      if self.outputs is not None:
2024-04-14 00:46:30.404709969  AttributeError: 'TensorRtDetector' object has no attribute 'outputs'
2024-04-14 00:46:31.384495370  [2024-04-14 00:46:31] audio.South                    ERROR   : ffmpeg process is not running,

hood2000

crazychen 发表于 2024-4-14 08:50
最后一顿胡倒腾，总算打开后台页面了，但ffmpeg 出错

是模型文件的问题。在docker compose过程中有问题，就是下面这个提示：
s6-rc: warning: unable to start service trt-model-prepare: command exited

你可以你重新compose试一下，shm_size 设置大一点，1024MB试一下。另外，看一下我主贴第5部分的图片，生成模型文件有个过程，如果logs中没有，就是没有成功。

25989406

但是p4买不到，咳咳。我用tpu还行

crazychen

hood2000 发表于 2024-4-14 17:48
是模型文件的问题。在docker compose过程中有问题，就是下面这个提示：
s6-rc: warning: unable to start  ...

我尝试更改更大尺寸还是不行，模型文件无法生成，模型文件可以手动下载吗？

还有就是提示版本低不知道有影响吗？root@ubuntu2:/home/frigate# docker compose up -d
WARN[0000] /home/frigate/docker-compose.yml: `version` is obsolete

hood2000

crazychen 发表于 2024-4-18 07:37
我尝试更改更大尺寸还是不行，模型文件无法生成，模型文件可以手动下载吗？

还有就是提示版本低不知道有 ...

yml版本只是提醒，没啥影响。

模型文件应该可以下载，另外，如果有fq环境，试一下，很多脚本尤其是涉及到在github上拉文件都可能受影响。，

crazychen

hood2000 发表于 2024-4-18 08:29
yml版本只是提醒，没啥影响。

模型文件应该可以下载，另外，如果有fq环境，试一下，很多脚本尤其是涉及 ...

试了，切换到科学环境下，仍没反应，郁闷至极

hood2000

crazychen 发表于 2024-4-18 12:15
试了，切换到科学环境下，仍没反应，郁闷至极

没有道理啊，我试了重新compose，也能正常下载模型文件。