Scale Smb broadcast

[tojezota]

Hello all,

I have upgraded to TrueNAS scale and have found that TrueNAS does not seem to broadcast smb either in windows explorer or Mac finder.
I have to connect to it manually I.e smb//: ipaddress.

I have tried in proxmox virtually or bare metal and neither work. Once connected manually I do not have share issues.
I did not have this with core and hoping it just a setting I’ve missed and someone can assist me..

Thanks

 

[Kris Moore]

We'll need to get a ticket created on this so we can investigate. Please open an issue on https://jira.ixsystems.com and attach a debug file (System -> Advanced -> Download Debug)

 

[sretalla]

Well, it should be if you have Network Global Configuration settings with WS-discover and mDNS set to on.

It might be that you need to restart the zeroconf service.

 

[Southpaw1496]

I am also experiencing this issue. I looked for the Zeroconf service and I couldn't find it, does it need to be restarted using the shell?

I made sure to check my settings, and both mDNS and WS-discover are enabled. Was a ticket created in the end?

 

[Southpaw1496]

After some head-scratching, I have managed to verify that while TrueNAS is broadcasting the _http._tcp. type, it is not broadcasting the _smb._tcp. type, which is why macOS can't discover it. I don't have a Windows machine to test if WS_discover is working or not.

 

[Southpaw1496]

I have found NAS-118579 which seems to describe this issue, after I fixed it by mistake.

 

[mistermanko]

This just hit me and I pulled my hair out trying to fix it. Simply touch the avahi smb service file and smb time machine is appearing again. It's beyond me that this is still not fixed...

 

[anodos]

This just hit me and I pulled my hair out trying to fix it. Simply touch the avahi smb service file and smb time machine is appearing again. It's beyond me that this is still not fixed...

It's heavily dependent on user network configuration. A misbehaving mDNS client (or misconfigured networking appliance) can basically trigger a "passive observation of failure" in other clients causing the mDNS SRV records to get dropped. Restarting SMB restarts avahi and makes it work a little while longer until it gets killed again in these less-than-felicitous setups.

 

[mistermanko]

Restarting the smb service did not change that behavior, it was the "touch the ahvahi smb service file" that changed it back to normal.
Could it have something to do with VLANs? I was fiddling with a vlan setup in TN and believe that could have caused it because before the same setup worked with Time machine for 1 year+ uninterrupted.

 

[anodos]

Restarting the smb service did not change that behavior, it was the "touch the ahvahi smb service file" that changed it back to normal.
Could it have something to do with VLANs? I was fiddling with a vlan setup in TN and believe that could have caused it because before the same setup worked with Time machine for 1 year+ uninterrupted.

Touching avahi smb service file has same effect as restarting the mDNS service because avahi daemon sets an inotify flag on the service files and reloads automatically if they're touched.

 

[mistermanko]

What is the scope on this getting fixed? The jira issue seems kinda stale?

 

[ragametal]

I have a mixed environment with Windows and Mac clients. The Windows clients can discover the truenas scale server when they go to "Networks" without a problem.

The Macs, on the other hand, cannot find it when they go to finder-->network but they can connect to it manually if they go to finder-->go-->connect to server.

i followed the advise of touching the avahi SMB file by SSH into the truenas scale server and opening the file in the vi text editor
sudo vi /etc/avahi/services/SMB.service

Once the document was displayed on my screen i hit "esc" and then :wq to write and close the document. As soon as i did that the truenas scale server was discoverable on the mac clients.

Worked like magic, but bad magic at that since i have to do this every time the server is restarted.

Is there a more permanent workaround to this besides manually "touching" the SMB.service file?


Edit:
Before "Touching" the file, i confirmed that the avahi service was up and running by using the command sudo service avahi-daemon status.

I even tried to restart the avahi service manually by sudo service avahi-daemon restart and this didn't work. "Touching" the file has been the only solution i have found so far.

 

[Kris Moore]

Checking the Jira issue, looks like the fix is merged and will be in 23.10.0 next week.

 

[ragametal]

Checking the Jira issue, looks like the fix is merged and will be in 23.10.0 next week.

Thanks for the update.

 

[ragametal]

Just to report that after upgrading to TrueNAS-SCALE-23.10.0.1, this issue seems to be resolved on my system.
I figured I should report something that is not a problem anymore.