Oh I get the standard disconnects / reconnects in the log, just didn't see them of note to mention as I've had the unit on and off while farting around with it (and doing some mait).
Grepping on all syslogs (including archives) gets me nothing on ftdi or FTDI, or on lsusb - the only things I can see in syslog apart from cron are events I already know about (plugged in a USB powered 12v fan to see if I could overload the power,then unplugged it, disconnection from wlan when I rebooted my router, etc)
Digging about to see if I can find a way to monitor CPU temps (no joy with /sys/class/thermal/thermal_zone0/temp) - at the moment I'm not running a print job as it's 38c (100f) today, but we've got a cool break coming so I can go to town.
There's room to suggest that the plus version of the 3B+ originally was rushed. It had a spate of freeze-up problems. In theory, those were eventually cured with Raspi firmware upgrades. The early advice was to slow down the CPU a bit. That was my earlier advice of upgrading. (I own like... eight Raspi 3B's and five or so zeroes but no 3B+ myself.)
I wanted to post on here even though I know its a bit old, just in case someone (like I did) comes along looking for help. I was getting the same error on an Ender 3 and was assuming that it was the Octoprint or cable causing the issue.
If using Cura 4.1 to slice, there is a problem where the Z speed is incredibly high, change the Z speed max to something reasonable, the ender 3 has a default of like 500 but I set it to 120 mm/sec. Re slice the item, upload it and it may fix the issue.