Hi,
my team and I have been working on application the uses 4x ZEDX cameras that we acquired a long time ago. Recently, we have been making a push to finalize our application but have realized that we are encountered frequently failures (about 1 every 30 minutes or so) that forces us to start our entire application in order to recover and it seems to be very consistent.
From what we have been looking at, we see some kind of nvargus failures are causing this but we are unclear why this is occuring. We have also noticed that the failures begin when our application is under motion (a robot) but does not seem to occur when stationary (or atleast far fewer failures). I recently installed a new driver (based on 1.2.0) that I was hoping would help remedy the issue, but it does not seem to have helped. Additionally, whenever we encounter errors - it tends to be 1 of the 4 cameras instead of all cameras at the same time. I have attached a snippet of the log output from the nvargus daemon.
Does anyone have any idea about whats going on here? Any recommendations on what we can do to resolve this issue?
For reference additional reference:
- we are using 4x ZED X cameras connected to a syslogic A4AGX (w/ their latest driver based on 1.2.0)
- our platform is running jetpack 5.1.2 (and we are unable to upgrade to jp6 at the moment)
- we are using the zed ros wrapper to interface with the cameras
nvargus_failures.log (181.7 KB)