March 29, 20251 yr Hi guys, I am having troubles with my Unraid server and cannot figure out what is causing the issue. I do not have anything extraordinary installed or made any big system settings chagnes. Issues that I have: - Syslog shows some error messages I cannot figure out what they mean, e.g.: Freezing user space processes failed after 20.004 seconds (3 tasks refusing to freeze, wq_busy=0) Call Trace: <TASK> __schedule+0x667/0x6bb - When I want to enter the GUI I only get a white page after the server has been running for a day or so - Every time I power down my server and boot it up afterwards Unraid starts a parity check. I have already doubled the timer for shutdown but still the same issue. Any ideas how to tackle one issue after the other? Or might they even be all related to the same root cause? Thanks! syslog
March 29, 20251 yr It looks like you've got a docker container which is continually restarting itself every other second. You should figure out that first. Uptime column on the docker tab should give you the information
March 29, 20251 yr Author Thank you for the quick reply! Which log entries exactly show that? When I go to the docker tab all dockers seem to be running for 15 minutes now after my last restart.
March 29, 20251 yr Author Okay I seem to have found the issue for the permanently restarting docker messages. It was the immich docker which I created per docker compose. The other error messages still keep appearing. Any ideas about the other messages? Mar 29 19:35:50 Tower kernel: Freezing user space processes Mar 29 19:35:50 Tower kernel: Freezing user space processes failed after 20.004 seconds (3 tasks refusing to freeze, wq_busy=0): Mar 29 19:35:50 Tower kernel: task:dockerd state:D stack:0 pid:22831 ppid:1 flags:0x00000006 Mar 29 19:35:50 Tower kernel: Call Trace: Mar 29 19:35:50 Tower kernel: <TASK> Mar 29 19:35:50 Tower kernel: __schedule+0x667/0x6bb Mar 29 19:35:50 Tower kernel: ? super_lock+0x36/0xf0 Mar 29 19:35:50 Tower kernel: ? __pfx_zpl_test_super+0x10/0x10 [zfs] Mar 29 19:35:50 Tower kernel: ? __pfx_set_anon_super+0x10/0x10 Mar 29 19:35:50 Tower kernel: schedule+0x8e/0xc3 Mar 29 19:35:50 Tower kernel: grab_super_dead+0x88/0xc1 Mar 29 19:35:50 Tower kernel: ? __pfx_var_wake_function+0x10/0x10 Mar 29 19:35:50 Tower kernel: sget+0xcd/0x1d0 Mar 29 19:35:50 Tower kernel: zpl_mount+0xa3/0x22a [zfs] Mar 29 19:35:50 Tower kernel: legacy_get_tree+0x24/0x44 Mar 29 19:35:50 Tower kernel: vfs_get_tree+0x1e/0xa5 Mar 29 19:35:50 Tower kernel: path_mount+0x64c/0x72c Mar 29 19:35:50 Tower kernel: do_mount+0x5c/0x8f Mar 29 19:35:50 Tower kernel: __do_sys_mount+0x100/0x12e Mar 29 19:35:50 Tower kernel: do_syscall_64+0x57/0x7b Mar 29 19:35:50 Tower kernel: entry_SYSCALL_64_after_hwframe+0x78/0xe2 Mar 29 19:35:50 Tower kernel: RIP: 0033:0x2039e8e Mar 29 19:35:50 Tower kernel: RSP: 002b:000000c0039ce5b0 EFLAGS: 00000216 ORIG_RAX: 00000000000000a5 Mar 29 19:35:50 Tower kernel: RAX: ffffffffffffffda RBX: 000000c00b570e40 RCX: 0000000002039e8e Mar 29 19:35:50 Tower kernel: RDX: 000000c007b1f470 RSI: 000000c00b570ea0 RDI: 000000c00b570e40 Mar 29 19:35:50 Tower kernel: RBP: 000000c0039ce5f0 R08: 0000000000000000 R09: 0000000000000000 Mar 29 19:35:50 Tower kernel: R10: 0000000000000000 R11: 0000000000000216 R12: 0000000000000147 Mar 29 19:35:50 Tower kernel: R13: 000000c00104e000 R14: 000000c000c66340 R15: 0000000000000027 Mar 29 19:35:50 Tower kernel: </TASK> Mar 29 19:35:50 Tower kernel: task:dockerd state:D stack:0 pid:37768 ppid:1 flags:0x00000006
March 30, 20251 yr Community Expert That call trace suggests an issue with a zfs pool, please post the diagnostics.
March 30, 20251 yr Author So I changed a couple of settings like making the appdata an exclusive share, deleting my docker folder and reinstalling all dockers and the call trace message disappeared. The only issue that is left is that unraid is not able to properly shut down and always does a parity check because of the forced shut down after the timer runs out. The repeating message is: Mar 30 23:49:42 Tower emhttpd: Unmounting disks... Mar 30 23:49:42 Tower emhttpd: shcmd (147333): /usr/sbin/zpool export -f cache Mar 30 23:49:42 Tower root: cannot export 'cache': pool is busy Mar 30 23:49:42 Tower emhttpd: shcmd (147333): exit status: 1 Mar 30 23:49:42 Tower emhttpd: Retry unmounting disk share(s)... Mar 30 23:49:47 Tower emhttpd: Unmounting disks... Mar 30 23:49:47 Tower emhttpd: shcmd (147334): /usr/sbin/zpool export -f cache Mar 30 23:49:47 Tower root: cannot export 'cache': pool is busy Mar 30 23:49:47 Tower emhttpd: shcmd (147334): exit status: 1 Mar 30 23:49:47 Tower emhttpd: Retry unmounting disk share(s)... syslog tower-diagnostics-20250330-2348.zip
March 31, 20251 yr Community Expert 11 hours ago, skel2raw said: Mar 30 23:49:42 Tower root: cannot export 'cache': pool is busy This means something is still using the pool, does it stop if you first disable docker and VM services?
March 31, 20251 yr Author Unfortunately not, even though I can stop all dockers if I try to stop the Array I get the same messages. Edited March 31, 20251 yr by skel2raw
March 31, 20251 yr Community Expert Leave the services disabled, reboot in safe mode, start the array, don't start the services, and see if it stops.
March 31, 20251 yr Author After I did what you suggested I was able to narrow the issue down to the docker service. It seems that at least one of the dockers running is causing the issue. However I have not yet figured out which one(s). However, if I manually stop all dockers the server shuts down without issues. Wondering why Unraid is not able to stop all dockers properly by itself. Edited March 31, 20251 yr by skel2raw
March 31, 20251 yr Community Expert It should, but it's not the first time I've seen this, see if you can find the problem container.
March 31, 20251 yr Author So... unfortunately I was not able to find out what the actual issue is. It seems that all my binhex dockers, cloudfaredtunnel, flaresolver cause this symptom. I have even completely removed two dockers and configured them from scratch but still no success. Any other ideas? I am even considering moving to a new Unraid Beta in the hope that something might be fixed that will prevent this issue from happening. Edited March 31, 20251 yr by skel2raw
March 31, 20251 yr Community Expert That's not normal, I would recommend asking for help in the support thread for one of them, then see if the issue is the same for the other ones.
April 1, 20251 yr Author It keeps getting stranger... I have again deleted all dockers including the docker folder on the cache drive. When I try to shut down the server gets stuck again with the same message, that the "chace" drive is busy even though dockers are not running. What else can there be other than dockers that keep the chace drive busy? Might it be a zfs issue? Edited April 1, 20251 yr by skel2raw
April 5, 20251 yr Author So far I have not been able to resolve this issue. I have even made a factory reset and installed everything form scratch. Same result... I'll try again to remove all plugins and install them one after the other with a couple of hours running between installing each of them. But my hunch is that it might have to do something with the ZFS pool. I did not have this issue before I converted my pool to ZFS.
April 10, 20251 yr Author I have now changed my cache drive from zfs to btrfs and the issue has resolved itself. Now my server is shutting down properly. The weired thing is that my other cache pool I use for downloads is not causing any issues.
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.