From: robin_listas@es.invalid   
      
   On 2025-12-28 22:00, Lawrence D’Oliveiro wrote:   
   > On Sun, 28 Dec 2025 14:33:27 -0000 (UTC), Lars Poulsen wrote:   
   >   
   >> My Fedora system is quite stable, but sometimes library updates are   
   >> a bit out of sync, causing (for example, most recently) the DHCP   
   >> daemon to fail on a "systemctl daemon-reload" at the end. It gets   
   >> fixed when I type "systemctl restart dhcpd", but it would be nice if   
   >> it could do that by itself, rather than waiting for me to figure out   
   >> that the reason half my SONOS speakers have gone offline is because   
   >> DHCP is confused.   
   >   
   > My guess is, some bug in the service dependencies. This can manifest   
   > in timing-related problems where a service will (re)start correctly on   
   > one occasion, but fail on another because something else it needs (but   
   > didn’t properly say it needs) isn’t quite ready yet.   
   >   
   > The error messages should give some hints: mention of errors trying to   
   > connect to some other process (e.g. MySQL/MariaDB, in a case that   
   > happened to me) would be helpful.   
      
      
   This happened to me recently:   
      
   Telcontar:~ # systemctl status smb   
   × smb.service - Samba SMB Daemon   
         Loaded: loaded (/usr/lib/systemd/system/smb.service; enabled; preset:   
   disabled)   
         Active: failed (Result: exit-code) since Wed 2025-12-17 23:09:54 CET;   
   19h ago   
           Docs: man:smbd(8)   
                 man:samba(7)   
                 man:smb.conf(5)   
        Process: 2479 ExecStartPre=/usr/share/samba/update-apparmor-samba-profile   
   (code=exited, status=1/FAILURE)   
            CPU: 77ms   
      
   Dec 17 23:09:54 Telcontar systemd[1]: Starting Samba SMB Daemon...   
   Dec 17 23:09:54 Telcontar update-apparmor-samba-profile[2479]: generating   
   profile sniplet failed   
   Dec 17 23:09:54 Telcontar systemd[1]: smb.service: Control process exited,   
   code=exited, status=1/FAILURE   
   Dec 17 23:09:54 Telcontar systemd[1]: smb.service: Failed with result   
   'exit-code'.   
   Dec 17 23:09:54 Telcontar systemd[1]: Failed to start Samba SMB Daemon.   
   Telcontar:~ #   
      
   I was baffled. I asked ChatGPT:   
      
   It told me to run   
      
      testparm   
      
      sudo /usr/share/samba/update-apparmor-samba-profile   
      
   The first one said:   
      
      Telcontar:~ # testparm   
      Load smb config files from /etc/samba/smb.conf   
      lpcfg_do_global_parameter: WARNING: The "domain logons" option is deprecated   
      Loaded services file OK.   
      Weak crypto is allowed by GnuTLS (e.g. NTLM as a compatibility fallback)   
      
      ERROR: both 'wins support = true' and 'wins server = ' cannot   
   be set in the smb.conf file. nmbd will abort with this setting.   
      
      Server role: ROLE_STANDALONE   
      
      Press enter to see a dump of your service definitions   
      [...]   
      
      
   So chatgpt told me that in here:   
      
      wins server = 192.168.6.16   
      wins support = No   
      
   I should remove the second line. It also advised to remove   
      
      domain logons = Yes   
      
      
   It later said:   
      
      
   Why the AppArmor message was misleading   
      
   The failure message:   
      
      generating profile sniplet failed   
      
   is not an AppArmor bug — it’s a configuration validation failure caused by   
   Samba rejecting the config.   
      
   This is a known pain point on openSUSE.   
      
   --   
   Cheers, Carlos.   
   ES🇪🇸, EU🇪🇺;   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)