"There are no exportable user shares" after upgrading to 6.10.3

[aronwk]

Upgraded from 6.9.2 to 6.10.3 and none of my share are exportable, and several of my file permissions are messed up.
Everything is still there, but with these issues.
Attached are diagnostics.

zerotwo-diagnostics-20220921-1711.zip

[trurl]

What do you get from command line with this?

ls -la /mnt/user

 

[aronwk]

5 minutes ago, trurl said:

What do you get from command line with this?

ls -la /mnt/user

 

this is the output:
 

d-wx-wx-wx  1 nobody users  119 Sep 21 17:25 ./
drwxr-xr-x 27 root   root   540 Sep 21 16:25 ../
drwx------  1 nobody users   43 Nov  5  2018 .Trash-99/
drwxrwxrwx  1 nobody users 2452 Aug  9 14:14 appdata/
drwxrwxrwx  1 nobody users   46 Feb 19  2022 backups/
drwxrwxrwx  1 nobody users   26 Nov 18  2021 domains/
drwxrwxrwx  1 nobody users  372 Sep 21 16:28 downloads/
drwxrwxrwx  1 nobody users   24 Mar  7  2021 firefly_iii/
drwxrwxrwx  1 nobody users   65 Oct 26  2021 fogserver/
drwxrwxrwx  1 nobody users   17 Jul  4 21:22 general/
drwxrwxrwx  1 nobody users   76 Jul  4 21:50 ghidra/
drwxrwxrwx  1 nobody users    6 Jun 27  2020 git/
drwxrwxrwx  1 nobody users  272 Nov 16  2021 isos/
drwxrwxrwx  1 nobody users  234 Oct 19  2021 media/
drwxrwx---  1 nobody users   40 Sep 13 21:25 nextcloud/
drwxrwxrwx  1 nobody users   74 Mar 29 13:10 pre-downloads/
drwxrwxrwx  1 nobody users   20 Dec 30  2021 syncthing/
drwxrwxrwx  1 nobody users    6 Nov  5  2021 syslog/
drwxrwxrwx  1 nobody users   26 Mar  7  2021 system/
drwxrwxrwx  1 nobody users    6 Sep 21 16:29 test/
drwxrwxrwx  1 nobody users   29 Aug  6 22:12 www/

 

[trurl]

 

[aronwk]

nice! that fixed the shares showing up and then I'm trying the docker safe new perms mentioned there right now

[nivagator]

On 9/21/2022 at 5:38 PM, trurl said:

What do you get from command line with this?

ls -la /mnt/user

 

 

Running into this issue multiple times today editing photos in a share. Unraid 6.10.3

 

Here is my output: 

$ ls -la /mnt/user
/bin/ls: cannot access '/mnt/user': Transport endpoint is not connected

 

[JorgeB]

10 hours ago, nivagator said:

/bin/ls: cannot access '/mnt/user': Transport endpoint is not connected

Please post the diagnostics.

[nivagator]

On 10/2/2022 at 3:40 AM, JorgeB said:

Please post the diagnostics.

Diagnostics attached.

galaxy-diagnostics-20221004-1110.zip

[JorgeB]

24 minutes ago, nivagator said:

Diagnostics attached.

shfs crashed, rebooting should bring the shares back, if you are not using NFS disable it since it looks like the crash was related to that.

[nivagator]

I saw that shfs crash and found your post here. Different service, but could it be memory overclocking related?

 

I am using NFS almost exclusively to access shares on my local network. I can replicate this failure nearly on command modifying or deleting images on a Linux client connected via NFS. Its typically on my third or fourth save or delete, the shares crash. Rebooting does restore the shares, but again, I happens consistently. I haven't had this issue with any other usecase and all other services are working as expected. 

 

I've been running this system for almost two years with no problems until this error. 

[JorgeB]

8 minutes ago, nivagator said:

but could it be memory overclocking related?

Unlikely that just NFS would crash, I've never used NFS so cannot really help with that but try updating to v6.11.0, newer kernel might help.

[nivagator]

Updated to 6.11.0 and it crashed immediately on the first attempt deleting files over nfs. This effectively makes the Unraid shares unusable... Anything else I can look into? 

 

Same segfault on shfs

Oct  4 21:41:47 galaxy kernel: shfs[12888]: segfault at 10 ip 000015002048f5c2 sp 000015001f05fc20 error 4 in libfuse3.so.3.12.0[15002048b000+19000] 
Oct  4 21:41:47 galaxy kernel: Code: f4 c8 ff ff 8b b3 08 01 00 00 85 f6 0f 85 46 01 00 00 4c 89 ee 48 89 df 45 31 ff e8 18 dc ff ff 4c 89 e7 45 31 e4 48 8b 40 20 <4c> 8b 68 10 e8 15 c2 ff ff 48 8d 4c 24 18 45 31 c0 31 d2 4c 89 ee 
Oct  4 21:41:47 galaxy kernel: ------------[ cut here ]------------ 
Oct  4 21:41:47 galaxy kernel: nfsd: non-standard errno: -107

 

[JorgeB]

Lets see if @dlandonhas any ideas.

[dlandon]

16 hours ago, nivagator said:

but could it be memory overclocking related?

Are you overclocking memory?

[dlandon]

Several suggestions:

• If you are overclocking memory, don't.  Overclocking memory on a server is not recommended.
• Start by updating all your dockers and plugins.  You are behind in quite a few.  I doubt it will fix anything, but just eliminate that possibility.
• NerdPack is not compatible with 6.11.  Be sure to remove the plugin and reboot.
• If you are on 6.10, remove the NerdPack plugin and reboot as a test.
• Try removing the nvidia drivers plugin.  I saw a reference to nvidia drivers in the segfault.

[nivagator]

Disabled memory overclocking, updated to 6.11, updated all plugins and docker, removed NerdPack. Still occurring.

 

Using Thunar File Manager on the client system. I said the error occurred when deleting files, but technically, i was clicking "Move to Trash". I enabled Delete in the Context Menu and have had no issues when choosing Delete and not Move to Trash. I'm assuming this is relevant, though I'm not experienced enough to know what it means.