In my last post the last feedback I had was from Vadim and Benjamin that they would be watching the thread to see how things were going. I made several posts indicating I had crashes and offered information and screenshots that I hoped would be helpful and I did not get any response. When I just went to add a follow up to the post I see that it is now closed since my last post was over a week ago.
This is one more example of the shortcomings of this “customer service” model.
I have had some wonderful feedback from other users over the course of my Roon journey but at the end of the day, at some point there should be a team at Roon that is big enough to support these issues themselves. It feels like I we are using the postal service to send troubleshooting ideas back and forth by mail.
Can someone on the Roon team look at the thread and either reopen it or just respond to this one with any feedback.
I was out of town but now I have been back for several days. In the good news, the Nucleus does not appear to have been crashing. In the past, after it would crash I would have to restart everything, and I could tell about when it crashed because my backups would have stopped.
I have had about 3 weeks worth of solid operation where the backups are happening everyday. That is the good news.
Now I am having another problem. Maybe this was always present, but because the system was operating so intermittently I didn’t realize it, but I am having an end point that keeps disappearing. The main system we listen through has a Meridian 251 powered controller as an endpoint. What is happening now is that during playback it will stop playing, and it will disappear from the list of endpoints and then after 30-90 seconds it will show up again. The music does not restart, I have to manually press the play button. Sometimes it will still show up as a source, but the play button will be grayed out and nothing will happen if I press it.
I looked in the forums and saw that several years ago there was quite a few people that were having problems with Meridian endpoints. It seemed like one recommendation was updating a switch, but wasn’t sure if there was an actual resolution that was reached between Roon and Meridian. It has happened 3 times in the past two hours as I was trying to listen to music. It happened right around 10:15 specifically, don’t recall the other exact times. When it happens, I can use any of the other endpoints just fine it is only this Meridian unit that is showing problems. Has this issue been resolved between Meridian and Roon? We have a second Meridian endpoint that is a non-powered controller, we don’t use it as often, so not sure if it is affected also.
Thanks for the follow-up! I’m glad to hear the crashing has subsided. For your meridian issues, we’re seeing the Meridian 251 is sending unsolicited packets. This is normal for Meridian’s discovery process, but seeing it frequently in the logs often indicates that the device is re-announcing itself to the network because it thinks the previous session was dropped.
Since the music stops and requires a manual restart, the Nucleus has definitely “forgotten” the playback state due to a total drop in communication. Try these steps in order:
Assign Static IPs / DHCP Reservations: Network audio endpoints (like the 251 and the 218) hate it when their IP addresses shift, or when a router "probes" them.
Log into your router and set a DHCP Reservation for the Nucleus and the Meridian 251.
This ensures that even after a reboot, they always live at the same "address," preventing the "grayed out" play button (which usually means Roon knows the device exists but can't find its current IP).
Check for "IGMP Snooping”: If you are using managed network switches (like Cisco, Ubiquiti, or Netgear ProSafe), ensure IGMP Snooping is turned OFF.
Meridian and Roon both rely on multicast traffic. IGMP Snooping often tries to be "smart" by blocking this traffic to certain ports, which causes endpoints to disappear every few minutes as the switch "prunes" the connection.
Simplify the Signal Path (The "Bypass" Test): To rule out a failing ethernet cable or a bad switch port:
Plug the Meridian 251 and the Nucleus directly into your main router using fresh Cat6 cables.
Update DNS Settings: Because your logs show 11-second delays for web requests, your Nucleus is likely timing out on its internal tasks.
In your router settings, change your DNS providers to Google DNS (8.8.8.8) or Cloudflare (1.1.1.1). This often clears up the "lag" the server experiences when trying to verify licenses or fetch metadata.
Let me know if any of the above help, thank you Dave!
Thanks for the suggestions. I will be unable to try them for a few days but hopefully at the end of the week I will be able to make the recommended changes.
