Roon core intermittently disconnects while playing music

Roon Core Machine

Nucleus purchased on store.roonlabs.com

Networking Gear & Setup Details

Third Floor: The Roon Nucleus is connected by CAT6 Ethernet cable to a Luxul 12-port/8 PoE+ Gigabit Managed Switch model AMS-1208P. The switch is connect by Ethernet cable.

Ground Floor: The switch is plugged by CAT6 cable to a Luxul Epic 5 Router. The router connects by Ethernet to the Xfinity xFI Gateway Modem model name CGM4331COM by vendor Technicolor.

Connected Audio Devices

dCS Rossini Apex DA converter connect by Ethernet to the third floor switch. Balanced cables to an integrated amplifier. iPhone and iPads used as remotes.

Number of Tracks in Library

9,633 tracks

Description of Issue

Each time I use Roon, I receive an intermittent warning message on my remote that the Roon core is disconnected. The screen changes to the Roon icon for a couple minutes. Then, the artist and track I was previously listening to will spontaneously repopulate the screen. I then press play to start the music again.

A second problem occurs when I intermittently receive the message that Qobuz or Tidal are loading too slowly. Roon skips the song and move to the next track.

Both problems occurred even after changing the switch to a TP link unmanaged network switch.

I have no other streaming problems gear other than the Nucleus. When I use the Mosaic app with the dCS DAC, I can stream Tidal and Qobuz without any problems.

@support @nucleus

Hi @Craig_Logan,

The unit is running out of memory. From your completed support template, I can safely say it’s not your number of tracks since I’d expect your Nucleus to handle up to 100k tracks. I discussed this with my fellow techs and we will need to escalate the issue to our developers to figure out why this is happening.

I will get back to you as soon as I can. Likely tomorrow.

Regards,
Wes

Thank you.

1 Like

Hi @Craig_Logan,

We couldn’t find anything definitive that we could pin down as the cause. All we have are working hypotheses on the matter.

  1. Significant tagging is taking place in the database and something is getting corrupted.

  2. The number of zones connecting to your core shows to be 93 on our base model Nucleus.

For our first thought, we think it’s wise to step away from your database and try to use Roon with a fresh one for a good 3 days to see what happens. Here are the steps for that:

  • Create a Backup of your current Roon Database
  • Stop RoonServer from running in Nucleus’s WebUI
  • Navigate to your Nucleus’ Database Location
  • Find the folder that says “RoonServer”
  • Rename the “RoonServer” folder to “RoonServer_old”
  • Restart the RoonServer in the WebUI to generate a new Database
  • On the Roon Remotes, press “Use another Core” and connect to the new database

As far as #2 goes, we should be able to handle this many devices. The only factor here could be how many of these devices you have grouped into each zone and how many zones you’re attempting to play to at any given time. We didn’t see evidence that this was the cause but I can’t help but keep coming back to this as a potential cause. I did hardware testing for Roon for some time and at the peak of this testing I had 34 devices associated with my core. I will list steps for refreshing RAAT. Please note, if you do this you will need to re-enable all of your devices and set up any zones again. You will also lose any and all DSP settings so please be advised:

  • Create a Backup of your current Roon Database
  • Stop RoonServer from running in Nucleus’s WebUI
  • Navigate to your Roon’s Database Location
  • Find the folder that says “RAATServer”
  • Rename the “RAATServer” folder to “RAATServer_old”
  • Restart the RoonServer in the WebUI to generate a new RAATServer

Thanks for your help. I will work on this and let you know if it works.

I usually only have one or two zones going when listening to music.

I’ve taken the steps for problem 1. This has exacerbated the problem further. Every few seconds, I get a warning message which state “Waiting for Roon Core”. I can’t login.

Hi @Craig_Logan,

Sorry, it took me some time to respond. Is your issue ongoing? I checked in on your core and see that is reporting to Roon in real-time. Is the screenshot from your Mac?

Thanks,
Wes

I was able to login. I’ve noticed far fewer interruptions in the sound. I believe I only noticed a couple of hangups when playing music within the first couple hours of following your advice.

Could the system have still been refreshing itself when I initially noticed problems logging back in?

Hi @Craig_Logan,

It’s hard to say but entirely possible. To my knowledge, there was no downtime with Roon account servers or any other service.

If you do happen to notice interruptions in sound going forward, please note the date/time/track played and let me know. I can further isolate the problem in diagnostics with that information and will be able to offer additional guidance.

Thanks,
Wes

The problem of intermittent disconnections from the Nucleus persists. The disconnections are shorter and less frequent.

Hi @Craig_Logan,

Did you happen to note the information above? I’d like to take a look at the diagnostics logging but can’t comb through all of the logging to find the problem. Please let me know. I would like to see this issue disappear altogether!

Wes

Ok. It just happened when listening to The Weeknd’s song Take My Breathe. 2:15 pm EST.

Hi @Craig_Logan,

Here’s what I found:

02/06 19:12:10 Warn: [Chord Electronics 2Go] [zoneplayer/raat] long rtt sync Chord Electronics 2Go: realtime=807344385917 rtt=67500us offset=260742385us delta=50720us drift=49055us in 15.324s (3201.250ppm, 11524.501ms/hr)
02/06 19:12:10 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":18354}
02/06 19:12:10 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":24000}
02/06 19:12:11 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":22257}
02/06 19:12:11 Trace: [raatserver] [RaatServer Craigs-Mac-mini @ 10.37.129.2:9200] connecting (attempt 5)
02/06 19:12:11 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":24000}
02/06 19:12:12 Info: [stats] 23462mb Virtual, 1789mb Physical, 605mb Managed, 578 Handles, 128 Threads
02/06 19:12:12 Trace: [Chord Electronics 2Go] [Lossless 71.7x, 24/48 MQA TIDAL FLAC => 24/48 MQA] [100% buf] [PLAYING @ 0:14/5:39] Take My Breath - The Weeknd
02/06 19:12:12 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":9697}
02/06 19:12:13 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":417}
02/06 19:12:13 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":13074}
02/06 19:12:13 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":24000}
02/06 19:12:14 Warn: [Chord Electronics 2Go] [zoneplayer/raat] long rtt sync Chord Electronics 2Go: realtime=811430134667 rtt=81500us offset=260739134us delta=44313us drift=45804us in 19.399s (2361.189ppm, 8500.281ms/hr)
02/06 19:12:14 Trace: [Chord Electronics 2Go @ 192.168.5.107:34007] [raatclient] GOT [35] {"status":"Dropout","samples":24000}
02/06 19:12:14 Warn: [Chord Electronics 2Go] [zoneplayer/raat] Too many dropouts (>3s dropped out in the last 30s). Killing stream
02/06 19:12:14 Trace: [Chord Electronics 2Go] [zoneplayer/raat] too many dropouts. stopping stream

So it looks like it’s not really a core disconnect problem but a network latency-related one. As you can see in the next log (same timeframes as above), the Chord seems to be losing its network connection. This is reflected in the "bw limit:0kb

02/06 19:12:52 Info: FTMSI-B 1 FileCache qo/5B754262 dwStatus:EndOfFileReached files:0 accessTimeOut:True priorities: ('zoneplayer:70':1) --> bw limit:0kbps
02/06 19:12:52 Info: FTMSI-B 2 FileCache qo/0B2B89A1 dwStatus:AllBlocksDownloaded files:0 accessTimeOut:True priorities: ('zoneplayer:70':2) --> bw limit:0kbps
02/06 19:12:52 Info: FTMSI-B 3 FileCache ti/48508D5B dwStatus:AllBlocksDownloaded files:0 accessTimeOut:True priorities: ('zoneplayer:72':23) --> bw limit:0kbps
02/06 19:12:52 Info: FTMSI-B 4 FileCache ti/D20FD161 dwStatus:AllBlocksDownloaded files:0 accessTimeOut:True priorities: ('zoneplayer:76':4) --> bw limit:0kbps
02/06 19:12:52 Info: FTMSI-B 5 FileCache ti/08E87B10 dwStatus:AllBlocksDownloaded files:0 accessTimeOut:True priorities: ('zoneplayer:72':22) --> bw limit:0kbps
02/06 19:12:52 Info: FTMSI-B 6 FileCache ti/787F0890 dwStatus:AllBlocksDownloaded files:0 accessTimeOut:True priorities: ('zoneplayer:76':3) --> bw limit:0kbps
02/06 19:12:52 Info: FTMSI-B 7 FileCache ti/BEF8F2F0 dwStatus:FirstBlockRetrieved files:0 accessTimeOut:False priorities: ('zoneplayer:84':18) --> bw limit:0kbps

Is the issue happening with the Chord and Chord alone? Is it a part of a zone you’ve created, and if so, does removing the Chord from the zone make any difference?

Let me know.

Thanks,
Wes

Hello Wes,

Thanks for the update. I have multiple Roon endpoints. The problem happens with all of them - dCS Rossini DAC, Naim Uniti Atom, Chord Poly, iPhone, iPad, Sonos, etc.

Hi @Craig_Logan,

I only have one idea before I can point to a network issue. If this doesn’t change the situation we’ll need to look at simplifying some elements of the network and testing. One thing I might suggest right away is ensuring the Xfinity gateway is indeed in bridge mode. I see at least one instance of the Mac having an Xfinity address and that implies it might not be in bridge mode.

This will clear your core’s cache and I am hoping this will prevent us from having to go any further!

  • Exit out of Roon
  • Stop RoonServer via the Web UI
  • Find and open your Roon database
  • Navigate to RoonServer/Cache
  • Move the contents of the /Cache folder elsewhere, like your desktop
  • Try restarting Roon and verify if the issue still occurs

Regards,
Wes

Good morning. I followed your advice and am still experiencing the same problem.

Hi @Craig_Logan,

I am still seeing network timeouts. Was this what you were listening to today when it happened?

“Album of the year J. Cole”?

02/08 18:12:37 Trace: [Naim Audio Uniti Atom HE] [Lossless, 24/44 MQA TIDAL FLAC => 24/44 MQA] [100% buf] [PLAYING @ 1:56/2:19] Album of the Year (Freestyle) - J. Cole
02/08 18:12:37 Info: [stats] 23185mb Virtual, 1169mb Physical, 489mb Managed, 544 Handles, 89 Threads
02/08 18:12:42 Trace: [Naim Audio Uniti Atom HE] [zoneplayer/raat] sync Naim Audio Uniti Atom HE: realtime=65439427733287 rtt=500us offset=65310084733us delta=-385us drift=-3864us in 121.7955s (-31.726ppm, -114.212ms/hr)
02/08 18:12:42 Trace: [Naim Audio Uniti Atom HE] [Lossless 99.8x, 24/44 MQA TIDAL FLAC => 24/44 MQA] [100% buf] [PLAYING @ 2:01/2:19] Album of the Year (Freestyle) - J. Cole
02/08 18:12:43 Debug: FTMSI-B-OE ti/EE05680A exit thread signalled
02/08 18:12:43 Debug: FTMSI-B ti/EE05680A download status: AllBlocksDownloaded accessTimeout:True openFiles:1 prev:(FirstBlockRetrieved,True,1)
02/08 18:12:43 Info: FTMSI-B ti/EE05680A: allocated bw changed from 51200 to 0 kbps
02/08 18:12:43 Info: FTMSI-B 1 FileCache ti/80FFC500 dwStatus:AllBlocksDownloaded files:1 accessTimeOut:False priorities: ('zoneplayer:40':1) ('zoneplayer:50':1) --> bw limit:0kbps
02/08 18:12:43 Info: FTMSI-B 2 FileCache ti/EE05680A dwStatus:AllBlocksDownloaded files:1 accessTimeOut:True priorities: ('zoneplayer:50':2) --> bw limit:0kbps

This time it was the Naim but these network issues aren’t the source of your issues, it’s the out of memory messages we’re seeing in diagnostics.

We might want to see if the issues happen on a fresh database. I would like to have you try this if we could. Make sure you back up your DB to a safe location before embarking on this and once done, refrain from re-importing the DB to test the situation. If you would, please run Roon for 24 hours or until you duplicate the issue before re-importing.

  • Create a Backup of your current Roon Database
  • Stop RoonServer from running in Nucleus’s WebUI
  • Navigate to your Nucleus’ Database Location
  • Find the folder that says “RoonServer”
  • Rename the “RoonServer” folder to “RoonServer_old”
  • Restart the RoonServer in the WebUI to generate a new Database

Thanks,
Wes

Hello Wes,

We tried these same steps about 8 days ago when you made the suggestion earlier in the thread. Unfortunately, It didn’t work for me.

Since the Nucleus is running out of memory, do I need to add RAM?

It’s possible it could help. My main concern was that you have a modest library and it’s happening. When we did this before, did you re-import your DB backup?

Can you try disabling the AirPlay instances in Roon? If nothing else, at least the ones that are Roon Ready devices that aren’t used via AirPlay, and let’s see if anything changes.

Wes