Upgraded to fangtooth, cifs mounts stopped working

TrueNAS General
,
1

From Dragonfish upgraded to Electric Eel my cifs mounts via autofs worked, Then upgraded to Fangtooth and cifs mounts stopped working.

I also tried mounting via fstab (old mounts that worked before) don’t work any more.

Linux client is arch.
when mounting via fstab dmesg gives:

sudo mount -a               
mount error(128): Key has been revoked
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs) and kernel log messages (dmesg)

dmesg:

CIFS: enabling forceuid mount option implicitly because uid= option is specified
CIFS: enabling forcegid mount option implicitly because gid= option is specified
CIFS: Attempting to mount //server/mp
CIFS: Status code returned 0xc0000072 STATUS_ACCOUNT_DISABLED
CIFS: VFS: \\server Send error in SessSetup = -128
CIFS: VFS: cifs_mount failed w/return code = -128

autofs gives the same in dmesg as above.

I have not found information about STATUS_ACCOUNT_DISABLED in Truenas.

I have searched the web but not found information how to solve the problem.

2

This may be a case where accounts passwords set to an empty string are breaking SMB setup (because the smbhash in an unexpected value). There was an API escape that allowed empty strings for passwords when users should have been checking the box to disable password authentication.

3

Thanks awalkerix. The user account that made the mounts have a password so this does not solve the problem.

4

Look in /var/log/middlewared.log and fix the affected accounts. There will be log messages that users have invalid hashes.

5

Awlkerix, sorry for not coming back until now. I have been occupied and did power down the truenas scale server. It turned out that the server did not reboot after the upgrade to Fangtooth, or at least no log messages in the /var/log/middlewared.log . So when the server was powered on today it worked perfectly well and there was no log messages with invalid hashes and only one log message that stated version 25.05… have started. A cold boot solved the problem. Thanks Awalkerix to pointing to middleware.log.