Roon Stops Playing Randomly To Pi Endpoint (DietPi & IQAudio Dac+)

@support

Hello,

Have build up a couple of Pi endpoints recently, one driving an Allo DigiOne and the other an IQAudio Dac+. I cannot say if they are both behaving this way (Pi with Allo DigiOne has hardly been used), but the IQAudio Pi seems to stop playing randomly. I cannot find any link with file type, size or format. Nor does it stop after any particular set period of time. It is just as if someone has pressed the pause button, it restarts instantly when the play button is pressed. I set it up with a 24 hour playlist at 6pm last night and it managed to play through for about 14 hours (so stopped at 8-9am today approx) 16 seconds into the start of a track. The first day I started using it is seemed to go for several hours or longer before stopping.

Everything is on the latest build and the PI was only built up last weekend with build v157. I have never been using a remote when it stops so cannot say if Roon produces any error messages when it does stop.

All details of my system with some screenshots are at the following link.

Please access my logs as required to look into this as I do not want to invest in any more Pi endpoints if this is going to be an issue.

It would help if you could post some additional info, particularly about your setup and network.

Hi,

it is all in the linked dropbox folder. There is a txt file with it all in.

Are you running DietPi or DietPi with the Allo GUI on the Pi?

Hi, just the plain DietPi (v157 I think released 19th October ish). The AlloGui seems to have a few teething issues so I have left it alone for now. Maybe the one to use for the next build. :slight_smile:
Only the IQAudio board and Roon were selected at setup, nothing else. Everything in the system is wired ethernet.

Cheers. I installed the build with the Allo web gui and my audio seems to cut out from time to time. But I changed too many things at once. I upgraded to 157. I installed the GUI and I added an iPurifier (which caused issues because it pushed my power consumption beyond the Pi power supply so had to get a new one). So it is interesting what you’re reporting. It could be Roon. It could be the Pi. As for the GUI - I’m not really using it. I like the netdata software. If I can install that separately I might get rid of the gui.

I did a search for similar issues here but whilst there are quite a few people complaining about dropouts (I have not heard any) I haven’t seen many ‘randomly stopping’ ones. Those I have seen have different hardware to me so are not necessarily comparable.

Hi @Anthony_B ---- Thank you for the report and sharing your feedback with us. Very appreciated!

Moving forward, confirming that I have downloaded the content of the provided dropbox link above (thank you), and have also went ahead and enabled diagnostics on your account so we can have a closer into what is going with your setup.

In the mean time, based on the information found in the “rock readme.text” can you please verify the make/model of unmanged switch you are implementing?

-Eric

Hi Eric,

thanks for the quick reply; there are several models of unmanaged switches in the network:
ROCK & NAS->Linksys/Cisco SE2500->Netgear GS308 -> NetgearGS308 -> Virginmedia Router

There is one managed switch in the network, but no Roon traffic passes through it, a Netgear GS108T. It is connected to the GS308 immediately after the Virginmedia Router. It is easily temporarily disconnected if need be.
Roon has been playing fine through the Pi since I restarted playback about 4 hours ago.

Hi @Anthony_B ----- Thank you for the follow up! I appreciate you providing the requested feedback!

We have received the diagnostic report from your system and I am currently waiting on some feedback from our techs in regard to the information found in the log files.

Furthermore, you mentioned things have been playing fine for a few hours since your last update. How are things currently holding up? I know that this behavior has been sporadic and it has been hard to pin down the variable(s) which are causing these stops to occur but when you do have a successful listening session versus when you encounter stoppages, is there anything that you are doing differently in relation to your “use” of Roon?

-Eric

Hi Eric,

since my last post it has been playing continuously, but I have been out at work so if it did stop and restart I wouldn’t know. The only difference with the ‘soak’ playing it is doing at the moment is that it is from a long queue and all from the same NAS (Classical music folder is all on one NAS). When I started using the Pi/Dac+ on Saturday it was left playing an album and then left to go into radio mode, so the music formats and sources could vary under those conditions (I have two NAS). During that afternoon it stopped a few times which is how I noticed there was an issue. I generally checked this on an android phone Remote and also used this to search for another CD to play. It restarts straight away by pressing Play.

I was doing the 24h soak test to eliminate one NAS or the other from causing issues. I have not yet got onto the other NAS as the first one I chose has the issue (though not as bad as my first afternoon of use). I have read though quite a few accounts of music stopping when Android Remotes go to sleep and it did occur that something like this could be going on.

It is a very common scenario for me though to start play on one remote then use another to change tracks or stop/start play. Remotes can be on android phones, Win 10 or Mac OSX. This could be a challenging set of experiments unless the logs show something useful. If you can see the logs from Saturday afternoon, they should be of most use. All that music would probably have been from the other NAS, so perhaps that is a clue.

I have done a quick network diagram today to show the audio bits of my network:

As it has been running nearly 24h now without incident, I have stopped playback and replaced the queue with another 24h one but this time with music from the other NAS.

1 Like

Hi Eric,

music still playing through the night from the other NAS (Synology DS415 Play). However, the Wife had Sonos playback stop on her last night (around 9pm) and it would not restart (iPad Remote). The Play/Pause button was unresponsive and she had just changed albums. Playback stopped a minute or so into the first track. It was the soundtrack for Local Hero by Mark Knopfler and would have been sourced from the DS415 Play NAS. That should be in the logs somewhere and maybe it is the same issue? It also showed an error message about ‘Automated backups: Scheduled backup failed’, which I haven’t seen before. I have a screen grab of that if you need it.

Playback just stopping has happened before with Sonos as well as the usual ‘Roon has lost control of the audio device’ error message. It is one of the reasons I was moving away from Sonos onto the Pi/DAC+


Hi @Eric
Just an update. After replacing the queue last night with one coming from the other NAS (Synology DS415 play) the music kept going throughout the night but seemed to stop just after I left for work, so somewhere between 8 and 10 am UK time I think. It stopped during a Portishead album a few tracks from the end. This should show up in the logs easy enough. I restarted playback at around 5.30 pm and it started straight away as it always does.

Hi @Anthony_B ---- Thank you for the continued feedback and more importantly, thank you for your patience. My apologies for the slow response.

Moving forward, confirming that I have received your recent upload containing your network configuration. Thank you! Furthermore, I had a chance to discuss this behavior you are experiencing with some of our techs and one of our senior developers yesterday to get some feedback on the issue.

Our techs noted that they are seeing multiple traces in the logs that seem to suggest that there is something occurring on your network that is having influence on this issue you are experiencing with the mentioned endpoint. In light of this the team has asked if you could please perform the following test for us:

From the information in the provided diagram, your core machine (i.e NUC i7) and NASs are located in the garage. Can you please move the Pi close to the core (i.e on the same switch) with a new network cable, and confirm if you still make the mentioned observation.

The goal of this test should be to isolate just the Pi endpoint and remove as much complexity from the chain of communication as possible. All things considered of course.

-Eric

Hi @Eric,

many thanks for looking into this so swiftly. The ‘something occurring on your network that is having an influence on this’ sounds rather ominous! Whatever it is I would love to know the details so I can get it fixed. I work in IT so don’t spare the detail. I will be away from home over the weekend but will reconfigure as you suggest and put it on soak test again. There is plenty of other stuff on the network (the diagram just showed the audio bits) but nothing you wouldn’t expect these days (TVs, PVR, blu ray player and home automation hubs etc).

Regarding the error message I received regarding scheduled backups failing: This was explained when I looked in the Settings for backup and it stated that it could not backup whilst music was playing. As it was on a Playback soak test that explains it.

@Eric
Hi,
Pi now reconfigured as suggested around 9.30pm. Both NASs, ROCK and the Pi-DAC+ are on the same switch in the garage and are on a 24h playlist. All connections to the switch were remade and seemed sound, with a new cable for the Pi used. The playlist will move onto radio mode when it finishes. I won’t be able to report on progress until sometime Sunday, so have a good weekend. :slight_smile:

Hi @Eric ,

Well, the new configuration ran for approximately 11 and a half hours before stopping all by itself. This would place the stop time at around 9am ish Saturday morning. I am sure it has stopped at that similar time on a few occasions before so something is maybe going on at that time on the network or with one of the devices? There was no-one at home using the network and there usually isn’t at that time during the week.
It stopped about 16 seconds from the end of a track, Millennium by William Orbit. It started instantly when I pressed the Play button, as it always does. That would have come from the Synology DS415 Play NAS.
I would be interested to know what the logs reveal about this stoppage and whether it is consistent with previous incidents, and where it points the finger of suspicion.
Update: it stopped around 8.30 Monday morning, playing Jeff Wayne’s War of the Worlds album.

Hi @Eric,

I have some more information. I happened to be watching Roon remote when it did it this morning. At 8.28 it lost connection to the core. It reconnected after 20 seconds or so with playback stopped and was doing a full rescan of all watched folders. This is typically what happens if Roon Server is manually restarted or crashes and restarts. I cannot tell if it did that as the uptime indicator is not present on the ROCK status page anymore. The main OS has been running for 18 days according to status page. So what is happening at this time every 24h to cause a server restart (if that is what is happening)?

Hi @Anthony_B ----- Thank you for the updates and my apologies for the slow response. I have been away from the office for a few days and I am currently catching up on any open support issues.

Moving forward, I have went ahead and re-enabled diagnostics on your account so we can receive an updated report which will contain the mentioned time frames in your latest posts. Once available I will then pass the diagnostics report over to our techs for evaluation to see if we can identify any commonalities in the traces found at the mentioned time frames.

-Eric

Hi @Eric,

slight change of behaviour this morning. I had music playing to both the PI/DAC+ and a Sonos Connect. At 8.28 a full rescan of the watched folders started but music kept playing and no loss of connection to the Core.