Gstreamer with libcamera for AWS webRTC

0

Hello all,

I am currently trying to set up webRTC streaming with my Raspberry Pi Zero 2 W. I am using a tutorial found here https://catalog.us-east-1.prod.workshops.aws/kinesis-video-streams/en-US/1-introduction , I am using a Raspberry Pi 4 Camera with IR-CUT night vision from amazon (https://www.amazon.com/Raspberry-Camera ... B0768Z87JF). I have set up the AWS CLI on the raspberry Pi properly and enter in the right credentials. I am only able to stream when I enable Legacy camera support, but the lighting is log=w and the camera does not use its full capabilities such as night vision. When I disable the legacy support and test out the camera with libcamera commands the camera uses its full capabilities such as night vision but when I try to stream, I get the error message:

"Error received from element source: Failed to allocate required memory. Debugging information: ../sys/v4l2/gstv4l2src.c(659): gst_v4l2src_decide_allocation (): /GstPipeline:live-kinesis-pipeline/GstV4l2Src:source: Buffer pool activation failed"

Is there any way to stream without enabling legacy camera support, or an alternative to legacy camera support that functions with libcamera instead of raspicamera? if not how can I do this?

已提問 1 年前檢視次數 310 次
1 個回答
0
已接受的答案

Hi, Looking at a few forums/articles online I could understand that "Failed to allocate required memory." is a device specific error. Unfortunately, I was unable to test this out on my end to help you with the right configuration for your setup.

Feel free to reach out to our dedicated KVS support as we might need more debug info from your device to troubleshoot this further. https://docs.aws.amazon.com/kinesisvideostreams/latest/dg/troubleshooting.html#troubleshooting-general

AWS
支援工程師
已回答 1 年前
  • Hello thank you for your reply, after doing some research I realized it is a picamera issue with the new Bullseye version of raspbian, I switched back to Buster and everything is working fine.

您尚未登入。 登入 去張貼答案。

一個好的回答可以清楚地回答問題並提供建設性的意見回饋,同時有助於提問者的專業成長。

回答問題指南