[Support]: Events not added to the event log but sent via MQTT #9974
-
Describe the problem you are havingI have a notification set on my phone to be notified in case of new event through MQTT. This event was a false-positive (false detection of a person) and when I've looked at the debug feed, I saw that the "person" was considered as stationary for 2 minutes with detection rechecks with a score between about 50% and 86%. The event never appeared in "In progress" events in the GUI, nor in the log after. This is my trigger: And the relevant part to wait for the end of the event and tell in the notification that the clip is ready: This is the event that triggered the automation: And this is the ending event: The "top_score" for the automation is higher (0.86) than the required threshold on Frigate (0.85). I'm doing that because I tune my Frigate+ model with new events that can be false-positive and I don't want to be notified for them. When the event ended, I clicked on my notification to get the replay and I got : So my question is: in which conditions the DB / web UI of Frigate would not record or "discard" an event that was started, updated and ended on MQTT? Version0.13.2 Frigate config fileobjects:
filters:
person:
threshold: 0.85
car:
threshold: 0.80
cameras:
Frigate_Buanderie_Arriere:
ffmpeg:
inputs:
- path: rtsp://XXX:XXXX@XXX/Streaming/channels/XXX
roles:
- detect
- record
zones:
jardin_arriere:
coordinates: 640,360,640,0,0,0,0,360
record:
enabled: true
retain:
days: 1
mode: all
events:
retain:
default: 2
mode: active_objects
detect:
max_disappeared: 50
objects:
track:
- person
filters:
person:
min_area: 1000
snapshots:
enabled: true
bounding_box: true
retain:
default: 7
motion:
improve_contrast: false
mask:
- 170,330,170,360,148,360,150,328Relevant log outputNothing relevantFFprobe output from your cameraNot relevantFrigate stats{
"cameras": {
"Frigate_Buanderie_Arriere": {
"audio_dBFS": 0,
"audio_rms": 0,
"camera_fps": 5,
"capture_pid": 458,
"detection_enabled": 1,
"detection_fps": 0,
"ffmpeg_pid": 463,
"pid": 435,
"process_fps": 5,
"skipped_fps": 0
}
},
"cpu_usages": {
"1": {
"cmdline": "/package/admin/s6/command/s6-svscan -d4 -- /run/service",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"15": {
"cmdline": "s6-supervise s6-linux-init-shutdownd",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"16": {
"cmdline": "/package/admin/s6-linux-init/command/s6-linux-init-shutdownd -c /run/s6/basedir -g 3000 -C -B",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"24": {
"cmdline": "s6-supervise frigate",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"25": {
"cmdline": "s6-supervise go2rtc",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"26": {
"cmdline": "s6-supervise s6rc-fdholder",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"27": {
"cmdline": "s6-supervise nginx",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"28": {
"cmdline": "s6-supervise s6rc-oneshot-runner",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"29": {
"cmdline": "s6-supervise go2rtc-healthcheck",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"30": {
"cmdline": "s6-supervise go2rtc-log",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"31": {
"cmdline": "s6-supervise nginx-log",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"32": {
"cmdline": "s6-supervise frigate-log",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"41": {
"cmdline": "/package/admin/s6-2.11.3.2/command/s6-fdholderd -1 -i data/rules",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"42": {
"cmdline": "/package/admin/s6/command/s6-ipcserverd -1 -- /package/admin/s6/command/s6-ipcserver-access -v0 -E -l0 -i data/rules -- /package/admin/s6/command/s6-sudod -t 30000 -- /package/admin/s6-rc/command/s6-rc-oneshot-run -l ../.. --",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"80": {
"cmdline": "s6-log -b -- T 1 n0 s10000000 T /dev/shm/logs/nginx",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"81": {
"cmdline": "s6-log -b -- T 1 n0 s10000000 T /dev/shm/logs/go2rtc",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"82": {
"cmdline": "s6-log -b -- T 1 n0 s10000000 T /dev/shm/logs/frigate",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"89": {
"cmdline": "/usr/local/go2rtc/bin/go2rtc -config=/dev/shm/go2rtc.yaml",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.1"
},
"108": {
"cmdline": "bash ./run.user go2rtc-healthcheck",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"110": {
"cmdline": "python3 -u -m frigate",
"cpu": "16.0",
"cpu_average": "7",
"mem": "8.4"
},
"118": {
"cmdline": "nginx: master process nginx",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.1"
},
"143": {
"cmdline": "nginx: worker process",
"cpu": "0.1",
"cpu_average": "0",
"mem": "0.0"
},
"144": {
"cmdline": "nginx: worker process",
"cpu": "0.1",
"cpu_average": "0",
"mem": "0.0"
},
"145": {
"cmdline": "nginx: worker process",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.1"
},
"146": {
"cmdline": "nginx: worker process",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.2"
},
"148": {
"cmdline": "nginx: cache manager process",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
},
"296": {
"cmdline": "frigate.logger ",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.5"
},
"377": {
"cmdline": "frigate.recording_manager",
"cpu": "10.1",
"cpu_average": "6",
"mem": "1.6"
},
"386": {
"cmdline": "/usr/bin/python3 -c from multiprocessing.resource_tracker import main;main(83)",
"cpu": "1.2",
"cpu_average": "1",
"mem": "0.1"
},
"387": {
"cmdline": "frigate.detector.coral",
"cpu": "1.0",
"cpu_average": "3",
"mem": "0.7"
},
"389": {
"cmdline": "frigate.output ",
"cpu": "11.2",
"cpu_average": "11",
"mem": "0.8"
},
"396": {
"cmdline": "ffmpeg -f rawvideo -pix_fmt yuv420p -video_size 640x352 -i pipe: -f mpegts -s 1309x720 -codec:v mpeg1video -q 8 -bf 0 pipe:",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.6"
},
"435": {
"cmdline": "frigate.process:Frigate_Buanderie_Arriere",
"cpu": "0.7",
"cpu_average": "0",
"mem": "0.8"
},
"463": {
"cmdline": "ffmpeg -hide_banner -loglevel warning -threads 2 -hwaccel_flags allow_profile_mismatch -hwaccel vaapi -hwaccel_device /dev/dri/renderD128 -hwaccel_output_format vaapi -user_agent FFmpeg Frigate/0.13.2-6476f8a -avoid_negative_ts make_zero -fflags +genpts+discardcorrupt -rtsp_transport tcp -timeout 5000000 -use_wallclock_as_timestamps 1 -i rtsp://*:*@10.0.0.71/Streaming/channels/402 -f segment -segment_time 10 -segment_format mp4 -reset_timestamps 1 -strftime 1 -c copy -an /tmp/cache/Frigate_Buanderie_Arriere@%Y%m%d%H%M%S%z.mp4 -r 5 -vf fps=5,scale_vaapi=w=640:h=352:format=nv12,hwdownload,format=nv12,format=yuv420p -threads 2 -f rawvideo -pix_fmt yuv420p pipe:",
"cpu": "1.1",
"cpu_average": "1",
"mem": "0.4"
},
"407803": {
"cmdline": "sleep 30s",
"cpu": "0.0",
"cpu_average": "0",
"mem": "0.0"
}
},
"detection_fps": 5.8,
"detectors": {
"coral": {
"detection_start": 0,
"inference_speed": 5.99,
"pid": 387
}
},
"gpu_usages": {
"intel-vaapi": {
"gpu": "4.15%",
"mem": "-%"
}
},
"processes": {
"audioDetector": {
"pid": 506
},
"go2rtc": {
"pid": 89
},
"logger": {
"pid": 296
},
"recording": {
"pid": 377
}
},
"service": {
"last_updated": 1708588650,
"latest_version": "0.13.2",
"storage": {
"/dev/shm": {
"free": 993.7,
"mount_type": "tmpfs",
"total": 1024,
"used": 30.3
},
"/media/frigate/clips": {
"free": 291337.6,
"mount_type": "ext4",
"total": 502868.2,
"used": 185914.6
},
"/media/frigate/recordings": {
"free": 291337.6,
"mount_type": "ext4",
"total": 502868.2,
"used": 185914.6
},
"/tmp/cache": {
"free": 936.6,
"mount_type": "tmpfs",
"total": 953.7,
"used": 17
}
},
"temperatures": {},
"uptime": 57991,
"version": "0.13.2-6476f8a"
}
}Operating systemProxmox Install methodDocker Compose Coral versionUSB Network connectionWired Camera make and modelReolink 410w Any other information that may be helpfulNo response |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 17 replies
-
|
I observe the same behavior once without knowing why this can happen |
Beta Was this translation helpful? Give feedback.
-
|
if events start at similar time (like 2 people walking to the door) then a job will be run to clean those events up and only leave one to reduce redundancy, that have be what happened here. |
Beta Was this translation helpful? Give feedback.
-
the clean up job only cleans up events that were saved. has_clip and has_snapshot was false so there was nothing to clean up |
Beta Was this translation helpful? Give feedback.
-
|
Sorry but it's a bit off topic (by my fault). Please let's focus on the issue here. Why the event was there, with a thumbnail and was deleted, or not added at all to the log. |
Beta Was this translation helpful? Give feedback.
-
|
You can have a thumbnail but no snapshot for an event? |
Beta Was this translation helpful? Give feedback.
-
the thumbnail only exists while the event is in progress and isn't saved anywhere once the event ends but there was nothing to save. looking at the code I can see that for an event to be saved the object must have moved. Seems that because the object was detected but never moved at all it was not saved frigate/frigate/object_processing.py Lines 1002 to 1003 in a5a1a54 |
Beta Was this translation helpful? Give feedback.
-
|
That's an interesting idea! When I received the event, the thumbnail was almost the full frame of the camera, so a giant object (this is a valid case for this camera). It was also a false positive, so maybe triggered because of light or shadow, it was initially detected for "a single frame". Then there were subsequent detections with the stationary object and the event finally ended. |
Beta Was this translation helpful? Give feedback.
-
|
I can confirm that |
Beta Was this translation helpful? Give feedback.
-
|
Any suggestion to avoid false-positive in MQTT for this edge case? I think I'll add that to my trigger: I guess it should be expected for a Edit: I've checked 40 last traces and saw that for events that are either |
Beta Was this translation helpful? Give feedback.
-
|
Hello, I’m encountering the same issue: where MQTT events are being generated for objects that are stationary; these objects were falsely detected for a single frame and did not exhibit any subsequent movement. This situation is leading to premature notifications that have nothing attached. I noticed that an answer has been selected for this issue, but I’m wondering if there’s a straightforward solution. Currently, I’m using the recommended Notification Blueprint (SgtBatten). However, it seems I might need to create a custom trigger to handle this specific scenario. I apologize if I’m misunderstanding any aspect of this process. Could the solution be as simple as implementing a |
Beta Was this translation helpful? Give feedback.
-
|
I use a custom trigger and my solution was to add this check |
Beta Was this translation helpful? Give feedback.
-
|
Did that solution hold up over time? I'm not so adept at finding the mqtt message that matches the alert but considering the same. Thanks |
Beta Was this translation helpful? Give feedback.
-
|
Yes, no more issue of that type. Now, I need to migrate to review endpoint and I don't know if the same logic applies. |
Beta Was this translation helpful? Give feedback.
the thumbnail only exists while the event is in progress and isn't saved anywhere once the event ends but there was nothing to save.
looking at the code I can see that for an event to be saved the object must have moved. Seems that because the object was detected but never moved at all it was not saved
frigate/frigate/object_processing.py
Lines 1002 to 1003 in a5a1a54