RevPi rebooting after 1m30sec

[lrs]

Hello.

A client of ours purchased 2x RevPi back in 2021 and recently turned in one of the devices as a "no boot". It turns out they meant it keeps crashing/rebooting, which I have confirmed. It's being used as a calendar and control for heating/blinders for meeting rooms or something of that nature. It has a RevPi DIO connected.

Attached is a simple ping loop aimed at the EthB-port, it’s alive and working fine for approximately 1,5-2 minutes and then reboots without any warning.

The PWR led goes from green to red, then off and back on to green. I’ve seen A3 light up orange a few times, but I’m not sure it does so consistently. The disk was full when I first got it, so I removed a few modules that aren't used (Teamviewer and a few others). Could the eMMC be worn out?
Disconnecting the RevPI DIO does not seem to make any difference. I have inspected the circuit board but have not seen anything obviously wrong.

Here’s disk usage:

Code: Select all

pi@revpi:~ $ df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/root       3.5G  3.2G  180M  95% /
devtmpfs        459M     0  459M   0% /dev
tmpfs           463M     0  463M   0% /dev/shm
tmpfs           463M  6.3M  457M   2% /run
tmpfs           5.0M  4.0K  5.0M   1% /run/lock
tmpfs           463M     0  463M   0% /sys/fs/cgroup
/dev/mmcblk0p1   43M   24M   19M  56% /boot
tmpfs            93M     0   93M   0% /run/user/1000

Here’s memory usage when it comes back online after rebooting:

Code: Select all

fpi@revpi:~ $ free -mt				
              total        used        free      shared  buff/cache   available
Mem:            925         122         656           6         145         745
Swap:             0           0           0			
Total:          925         122         656			

Unfortunately, it reboots too quickly to create an image, ref:
https://revolutionpi.com/en/tutorials/i ... all-jessie

I have copied out the logs, and looked through kern.log.1 but can’t see anything that sticks out. It looks like it uncontrollably reboots after enabling the ethernet interfaces. I've attached the logs for the 10th of July, when it was powered on without interruptions or SSH connections. Also attached is config.rsc

What can I do to troubleshoot this? And what’s more, how can I exfiltrate the user/custom data? I'd prefer to do a full image, but I probably won't have time unless I find a cause of the reboots.

[lrs]

I manage to get to 3.7GB with dd before it reboots , using:

Code: Select all

sudo dd if=/dev/mmcblk0 of=/dev/sda bs=64K conv=noerror,sync status=progress

Getting around 24MB/s, would adjusting the block size be helpful at all? USB 2.0 bandwidth should in theory be a little higher.

[lrs]

I sent in an email to support and they were kind enough to gently suggest it was the watchdog rebooting the device.

60 sec timer on the WD and 30-40 seconds for the device to finish booting and starting services is a good fit for the behaviour I've registrered.

Shorting WD to 0V seems to have solved the problem, I have an uptime of 20 minutes now, and will keep an eye on it.

[lrs]

It was the watchdog rebooting the device as intended. It's happy and alive with it connected to 0V. I'll read the fine manual next time, I suppose.

Next problem I encountered was 100% inodes usage. Luckily a more common issue, I found a log folder that was very full and cleared that.