Stuck installing software for Multi-Cam-Adapter v2.2 and two OV5647 cameras

Hi, I’ve hit a wall trying to get the software side set up for the Arducam Multi-Cam-Adapter v2.2 with two OV5647 cameras. The hardware is physically installed—both cameras are plugged in and the ribbon cables are seated as tightly as I can get them—but I’m stuck on the actual install/loading commands. I’ve tried working through it myself but the software side just won’t cooperate. Any pointers on where to start?

Can you start with the Quick Start Guide for the Multi-Camera Adapter Board? For the Quad Camera Adapter Board (B012001) this is the page to follow:

https://docs.arducam.com/Raspberry-Pi-Camera/Multi-Camera-CamArray/Quick-Start-Guide-for-Multi-Adapter-Board/#for-quad-camera-adapter-boardb012001

Work through that and let me know exactly which step fails.

Thanks, I went through the guide but the cameras still aren’t working. I think it’s a software issue. I’m running OctoPrint and I have two v1.3 OV5647 cameras plugged into the top two ports of the multi-cam array board, ribbon connectors seated properly. Here’s what I get when I try libcamera-still:

chase@octoprint:~ $ libcamera-still -t 0 --camera 0
Preview window unavailable
[28:06:55.812717501] [8866] INFO Camera camera_manager.cpp:297 libcamera v0.0.5 +83-bde9b04f
[28:06:55.863953653] [8867] INFO RPI vc4.cpp:437 Registered camera /base/soc/i2c0mux/i2c@1/ov5647@36 to Unicam device /dev/media0 and ISP device /dev/media2
[28:06:55.864015282] [8867] INFO RPI pipeline_base.cpp:1101 Using configuration file '/usr/share/libcamera/pipeline/rpi/vc4/rpi_apps.yaml'
[28:06:55.864724426] [8866] INFO Camera camera.cpp:1033 configuring streams: (0) 1296x972-YUV420
[28:06:55.864956758] [8867] ERROR V4L2 v4l2_videodevice.cpp:1047 /dev/video0[12: cap]: Unable to set format: Device or resource busy
ERROR: *** failed to configure streams ***

And for the second camera:

chase@octoprint:~ $ libcamera-still -t 0 --camera 1
Preview window unavailable
[28:07:49.479973945] [8875] INFO Camera camera_manager.cpp:297 libcamera v0.0.5+83-bde9b04f
[28:07:49.533126343] [8876] INFO RPI vc4.cpp:437 Registered camera /base/soc/i2c0mux/i2c@1/ov5647@36 to Unicam device /dev/media0 and ISP device /dev/media2
[28:07:49.533189342] [8876] INFO RPI pipeline_base.cpp:1101 Using configuration file '/usr/share/libcamera/pipeline/rpi/vc4/rpi_apps.yaml'
ERROR: *** selected camera is not available ***

Thanks for the logs. The camera 0 error shows /dev/video0 is busy, and camera 1 isn’t being selected at all. Can you grab a camera bug report for me? Please reboot the Pi first, then run these commands:

wget -O arducam-camera-bug-report https://github.com/ketai-dhr/arducam-camera-bug-report/releases/download/tools/arducam-camera-bug-report
chmod +x ./arducam-camera-bug-report
./arducam-camera-bug-report

That creates an arducam-bug-report.txt file in the current directory. Attach that file here so I can take a closer look.

Here’s the report. I ran the tool right after a fresh reboot as requested.

arducam-bug-report.txt

Thanks, I’ve had a look at the report. I’ve sent you a screenshot privately showing the exact line/section to delete. Remove that line, then reboot the Raspberry Pi so the change takes effect.

Okay, I deleted the section from the screenshot you sent, rebooted, but it’s still not working. At this point I’m thinking about reloading a fresh copy of OctoPrint and starting from scratch. Quick question on versions—should I use the ‘New Camera Stack’ OctoPrint image, the ‘Stable’ one, or the version on your OctoPrint Solution Page (which is over a year old)? I keep reading that bullseye has built-in support for the multi-camera module and OV5647 v1.3 cameras, so I want to make sure I use the right image.

Before you reflash anything, let’s check one basic thing: the Pi may simply not be detecting the cameras. Run this and post the output:

ls /dev/video0

If /dev/video0 isn’t there, that would explain the ‘selected camera is not available’ and ‘Device or resource busy’ errors. In that case we need to focus on why the cameras aren’t being detected, rather than which OctoPrint image you’re using.