January 7, 20251 yr Hi! This has been going on for a while, but I've just been living with it--but it needs to be fixed. When I try to stop the array, I'm getting stuck on the retry unmounting user share(s) message. When I look in the log, I get this: Jan 7 00:31:45 unServer emhttpd: shcmd (1843561): exit status: 1 Jan 7 00:31:45 unServer emhttpd: shcmd (1843563): /usr/local/sbin/update_cron Jan 7 00:31:45 unServer emhttpd: Retry unmounting user share(s)... Jan 7 00:31:50 unServer emhttpd: shcmd (1843564): /usr/sbin/zfs unmount -a Jan 7 00:31:50 unServer emhttpd: shcmd (1843565): umount /mnt/user Jan 7 00:31:50 unServer root: umount: /mnt/user: target is busy. Jan 7 00:31:50 unServer emhttpd: shcmd (1843565): exit status: 32 Jan 7 00:31:50 unServer emhttpd: shcmd (1843566): rmdir /mnt/user Jan 7 00:31:50 unServer root: rmdir: failed to remove '/mnt/user': Device or resource busy and when running lsof: root@unServer:~# lsof | grep /mnt/user tail 7695 root 1w REG 0,39 139995 10414574175015971 /mnt/user/Logs/unServer-syslog-2024-12-29T043441 Now, I have a Logs directory set to cache -> disk that I use as the log server storage location for other network devices, but it seems that pointing the unraid syslog at that directory is killing me here. I don't want to pile logs onto the Flash, so what's the best way to handle this issue? Manually stopping the array with a kill command works, but I need the shutdown process to work when triggered by the UPS. Much appreciated!! J
January 7, 20251 yr Community Expert I cannot reproduce when saving the log to a zfs pool, are you saving to the array or a pool?
January 7, 20251 yr Author JorgeB, It saves to the cache pool (old btrfs pool), with the mover pushing files to the array.
January 9, 20251 yr Community Expert Also cannot reproduce with a btrfs pool, does that happen every time you stop the array? Also, what release are you running?
January 24, 20251 yr Author Solution On 1/9/2025 at 12:23 PM, JorgeB said: Also cannot reproduce with a btrfs pool, does that happen every time you stop the array? Also, what release are you running? I'm running 6.12.14. The issue was nearly every time, but I finally found the culprit. I'd (some time ago) put a user script to copy the log to the array so I didn't lose the log on a crash. It started on array start, but I guess I didn't make it stop when the array stopped... #!/bin/bash FILENAME="/mnt/user/Logs/unServer-syslog-$(date +%FT%H%M%S)" tail -f /var/log/syslog > $FILENAME So I killed that, and I think that's my issue. Is there a best practice for writing system logs to the array so you're not killing the flash drive?
January 24, 20251 yr Community Expert 42 minutes ago, uiuc_josh said: Is there a best practice for writing system logs to the array so you're not killing the flash drive? Setup syslog server
January 24, 20251 yr Author 28 minutes ago, trurl said: Setup syslog server Oof. RTFM, indeed. Extra stupid thing is that I'd enabled syslog server for my firewall and voip. Just never got rid of the manual tail process script b/c I didn't realize the server logs were going to the same place.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.