timbtbitc

I've seen this on a custom backup script of my own invention which ultimately is a whole of server rsync backup to a locally mounted pool of 5 usb disks. Which does the trick great when it works! If I run it from the user scripts it will run for various lengths of time (about 1-3 hours) then crash. If I hand run it from a server prompt - same behavior - crash. If I ssh in using bash cmd line - runs perfectly every time to completion. I'd love to give more diagnostics but when it happens the server is still running all things vms etc and slowly each connection fails. and power cycle is the only way out. In Feb this year - I got told my flash drive was corrupted during one of these crashes and soon after I had to replace the drive. It was corrupted. Today I got told the same thing - up popped notification - your flash drive is corrupt - see here to replace it... but following power cycle we seem to be fine. But I thought I'd better bug report this. I may wind up needing more than my allocated replacements :-} This is a 6 month old 32Gb samsung quality unit. Here's my script #!/bin/bash if [[ $(findmnt -M "/mnt/yottamaster") ]]; then echo "yottamaster IS mounted" nice -n 19 rsync -avzpogt --delete --progress --exclude={'datashedbu','ryconbu'} /mnt/user/* /mnt/yottamaster/datashedbu else echo "yottamaster NOT mounted" fi I realize there's not many clues here Only the firefox tab with the script running crashes - if you run the script in background - we crash but it's been months since I attempted a scheduled run of this - due to the server death and no knowing what has happened. You cant ssh in after the crash, cant power off, open another server window, cant command a reboot, cant access docker, cant ping server,dockers or vms - it looks like the network services and thus ssh take a big hit in the groin. You cant ping the server from anywhere else. Currently open firefox tabs on the server continue to run - only on tab crashes, But you cant use any survivor tabs though they update (ssh/network not completely dead??) Willing to try anything to provide more info I have 3 scripts like this one targets the usb disk shelf, two target to separate unraid servers. They are all set up the same and all were created to be automatically run - but I will say I have hand run them all via bash prompt because I could never get a reliable backup without a mysterious crash. But all run flawlessly by hand. since 6.xx - been a while - I thought it was me and my script but now I'm not so sure timb datashed-diagnostics-20250614-1822.zip