Jetson Nano 2GB boot issues

I have been experienced multiple boot issues on my Jetson Nano 2GB board. Basically, I am using this board as a video surveillance system, that detects objects using a tiny-yolo detection model and python script. Videos are written to the filesystem based on what is detected by the model, and these videos are uploaded to an AWS cloud server. The issue that I am facing is that after a few days/ a week, my board shuts off. It stops recording footage and uploading to our server, and the only thing that gets it back online is power cycling the unit. The weird part is that the power light is still on on the unit, but when I plug the unit in to HDMI, I just see a black screen. That makes me think the filesystem is crashing for some reason. I checked in the syslog files to see if there were any clues about what is going wrong, and the only thing I saw that might be related were lines mentioning file system errors, so it is possible that the filesystem has become corrupted. At the specific times when the unit goes offline, I do not notice any lines in syslog that mention crashes or failures; I just see normal system log lines from right before it went offline, and then log lines once it started booting up again after a power cycle. I was just wondering if anyone had any advice/pointers of how to further troubleshoot this issue, or if you think at this point I should just reflash the sd card/try a different sd card. I’m willing to share log files if needed.

Dump the log from uart. It will tell more than syslog.

Hello @WayneWWW, thanks for the reply. Unfortunately I do not have access to the device right now, so I cannot access the uart log. This unit is running a prototype of a project I am developing for my job, and it is in another state at one of our clients locations.

Just to give more context, this unit is running a python script which detects objects using opencv and a tiny-yolo detection model. It writes videos to disk throughout the day, so we can see how our model is performing at one of our clients locations. We have three other jetson nanos with this client, and they are not having issues like this; they are able to stay online consistently. When I run dmesg on the problematic unit, I do see a lot of lines mentioning errors on the file system, so I’m curious if these errors are causing this unit to keep going offline. I’m wondering if my issue is related to the fact that we are reading and writing a lot of data from the filesystem throughout the day, since we are running a neural network, recording footage to disk, and uploading the footage to a remote server.

So overall, I’m really just wondering if anyone has any recommendations/ideas of what is causing that issue online, or if it is hard to tell without being able to dump the uart logs?

If your issue is “the system suddenly reboots by itself”, then we need the uart log to tell.

Ok sounds good, I just wanted to make sure. I think I’ll just have our client mail the unit back to us, so I can troubleshoot with the uart logs like you suggested. If I have any further questions once I am able to get those logs, I will make a new forum post. I appreciate the help.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.