Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix default model for rknn detector #12807

Merged
merged 1 commit into from
Aug 7, 2024
Merged

Conversation

MarcA711
Copy link
Contributor

@MarcA711 MarcA711 commented Aug 7, 2024

Currently, rknn doesn't use yolonas_s by default and will crash instead, if no model is specified. This was reported here.

I have an unrelated issue but wanted to hear some feedback before opening a separate thread for it:
There is an issue that causes FFmpeg on Rockchip boards to crash sometimes. If this issue occurs, it causes frigates whole detection process to crash silently for this camera. There is no entry in the logs and the sub-process won't restart.
Even though this issue is specific to the rockchip platform, crashes and bugs can happen on all devices. I think it would make sense to implement a health-check in Frigate that periodically checks if all sub-processes for detection and recording still run. Any thoughts on this?

Copy link

netlify bot commented Aug 7, 2024

Deploy Preview for frigate-docs canceled.

Name Link
🔨 Latest commit 84af7aa
🔍 Latest deploy log https://app.netlify.com/sites/frigate-docs/deploys/66b36ff34ee1b70008583768

@NickM-27
Copy link
Sponsor Collaborator

NickM-27 commented Aug 7, 2024

Detection and ffmpeg (detect and recordings) already have health checks. The main thing that doesn't is the recordings maintainer, but that shouldn't have issues unless there is a seg fault.

@NickM-27 NickM-27 changed the base branch from dev to 0.14.1 August 7, 2024 13:03
@MarcA711
Copy link
Contributor Author

MarcA711 commented Aug 7, 2024

The main thing that doesn't is the recordings maintainer, but that shouldn't have issues unless there is a seg fault.

This seems to be the case for me. However, I guess it is a quite rare issue than and only present because of the memory issue on rockchip boards. Thank you.

@NickM-27 NickM-27 merged commit eace948 into blakeblackshear:0.14.1 Aug 7, 2024
10 checks passed
NickM-27 pushed a commit that referenced this pull request Aug 8, 2024
hawkeye217 pushed a commit that referenced this pull request Aug 9, 2024
sandnabba pushed a commit to sandnabba/frigate that referenced this pull request Aug 14, 2024
NickM-27 pushed a commit that referenced this pull request Aug 30, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants