Unable to (re)mount SMB shares

[Ryan_Malone]

Issue
I have 2 Proxmox servers that have remote mounts to 2 different NAS' I have (TruneNAS and Unraid) which have been working fine. I have it configured so if one NAS is down, my LXC containers such as Plex, Jellyfin, arrs, etc. can still work because I have a copy pointing at each NAS. I made some network changes on the TrueNAS which required some IP changes. I used 'umount' to unmount the SMB shares in order to create the mount point again, but with a different IP address for the NAS. One server was fine, but the other one won't let me mount the TrueNAS SMB shares manually or through the GUI. The error I'm getting seems like as red herring. I've spent 10+ hours on this and am considering just wiping Proxmox and starting again but thought someone might have some ideas of what other steps I could take to find the root cause of this issue.

Command and resultant error

root@pve3:~# mount -t cifs -o user=myusername,uid=100000,gid=100000 //192.168.30.182/movies /mnt/pve/smashy-movies -v
Password for myusername@//192.168.30.182/movies:
mount.cifs kernel mount options: ip=192.168.30.182,unc=\\192.168.30.182\movies,uid=100000,gid=100000,user=myusername,pass=********
mount error: Server abruptly closed the connection.
This can happen if the server does not support the SMB version you are trying to use.
The default SMB version recently changed from SMB1 to SMB2.1 and above. Try mounting with vers=1.0.
mount error(112): Host is down
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) and kernel log messages (dmesg)

Here is the relevant output from dmesg
[ 1954.434152] CIFS: Attempting to mount //192.168.30.181/movies
[ 1976.169960] CIFS: VFS: cifs_mount failed w/return code = -112
[ 5260.583030] perf: interrupt took too long (2504 > 2500), lowering kernel.perf_event_max_sample_rate to 79000
[ 7786.775036] perf: interrupt took too long (3148 > 3130), lowering kernel.perf_event_max_sample_rate to 63000
[19013.942340] perf: interrupt took too long (3944 > 3935), lowering kernel.perf_event_max_sample_rate to 50000
[45411.107844] CIFS: VFS: Autodisabling the use of server inode numbers on \\192.168.30.170\unraiddata
[45411.107849] CIFS: VFS: The server doesn't seem to support them properly or the files might be on different servers (DFS)
[45411.107849] CIFS: VFS: Hardlinks will not be recognized on this mount. Consider mounting with the "noserverino" option to silence this message.
[77321.249424] CIFS: enabling forceuid mount option implicitly because uid= option is specified
[77321.249427] CIFS: enabling forcegid mount option implicitly because gid= option is specified
[77321.249428] CIFS: Attempting to mount //192.168.30.182/movies
[77342.809337] CIFS: VFS: cifs_mount failed w/return code = -112
[83370.085175] CIFS: enabling forceuid mount option implicitly because uid= option is specified
[83370.085179] CIFS: enabling forcegid mount option implicitly because gid= option is specified
[83370.085180] CIFS: Attempting to mount //192.168.30.182/movies
[83391.515133] CIFS: VFS: cifs_mount failed w/return code = -112
[83584.911092] CIFS: enabling forceuid mount option implicitly because uid= option is specified
[83584.911097] CIFS: enabling forcegid mount option implicitly because gid= option is specified
[83584.911098] CIFS: Attempting to mount //192.168.30.182/movies
[83606.552890] CIFS: VFS: cifs_mount failed w/return code = -112

What I have tried so far
1. Tried mounting from GUI - Not possible
2. Tried enabling SMB1 and connecting - Didn't work
3. Tried specifying SMB1, SMB 2.0, SMB 2.1, and SMB 3.0 when mounting - Same error
4. Restarting SMB service and restarting NAS, and restarting the Proxmox host - No luck
5. Pinging the NAS from the host and from LXC containers - it responds immediately
6. Deleted and recreated the directories I want to mount - No change

All I did was to unmount and then go straight back and manually mount like a minute later so there aren't a whol;e lot of changes where I can think of that could cause this. The other server used the same commands exactly. In fact, I documented each step in my wiki and just copy and pasted after it worked on one, to the other so all the names and mounts would be the same. Any ideas of what I could try in order to pinpoint where the issue might lie are greatly appreciated.

 

[Ryan_Malone]

Bump

 

[bl1mp]

Hi,
which order did you use to restart the devices?
Did you shutdown Server & NAS at the same time for a clean reboot?

SMB is known for hangers on mounting & unmounting (it is related to the oppurtunistic lock feature, if I remember correctly).
And both implementations needs to be compatible

Did you try pinning the SMB version on both sites?
What does the Log of the NAS tell you?

If you are not using it for Windows Clients, maybe yo can consider use NFS4.
The SMB-Protocol is defined by Microsoft, and implemented through diffrent libs in the Linux/Unix World.
NFS is following a RFC-Standard.

Best regards,
Lucas

 

[Ryan_Malone]

Hi,
which order did you use to restart the devices?
Did you shutdown Server & NAS at the same time for a clean reboot?

No I didn't reboot at the same time. I rebooted Proxmox first and then TrueNAS.

SMB is known for hangers on mounting & unmounting (it is related to the oppurtunistic lock feature, if I remember correctly).
And both implementations needs to be compatible

Did you try pinning the SMB version on both sites?

I did.
"3. Tried specifying SMB1, SMB 2.0, SMB 2.1, and SMB 3.0 when mounting - Same error"

What does the Log of the NAS tell you?

I pasted them above

If you are not using it for Windows Clients, maybe yo can consider use NFS4.
The SMB-Protocol is defined by Microsoft, and implemented through diffrent libs in the Linux/Unix World.
NFS is following a RFC-Standard.

NSF is so confusing trying to map our how Proxmox maps IDs from an unprivileged LXC and then how that maps to TrueNAS that I've avoided it. Also, how could the 2 SMB implementations not be compatible when the Proxmox server next to this one works fine with the exact same setttings and so did this one before I unmounted it?

Best regards,
Lucas

 

[bl1mp]

Well, I usually prefer VMs over LXC. Which nihilates the kernel priviledge topic.
Found this post, for mounting NFS with LXC: https://forum.proxmox.com/threads/tutorial-mounting-nfs-share-to-an-unprivileged-lxc.138506/

Maybe FUSE mount is the option that might help you on that.
But it does not solve your current SMB-Problem, Sorry.