Hi @Anthony_B ----- Thank you for the update, I appreciate the insight.
I just checked our servers again and I am still not seeing the updated diagnostics report from your core machine I can see the request has been placed but nothing has made it to our servers unfortunately.
Anthony, would it be possible for you to manually upload a set of your Roon core logs to us via the instructions found here? I want to make sure we have an updated set of logs from the tests you have been performing.
Great news. Apart from the incident already mentioned at 8:28 this morning, music has been playing all day apart from a hiccup at around 9.24. Sonos had a bit of a pause but restarted itself OK after 10 seconds or so, but the Pi went into pause mode and required a manual press of the play buttonâŚ
do you have an update on this yet? Although I am no longer actively monitoring the disconnects, it is still clear that the Pi regularly stops playing, but a Sonos unit tends to carry on. The Pi issue was captured in the logs I sent. The network has been returned to its original configuration.
Hi @Anthony_B ----- Thank you for touching base with me and more importantly, thank you for your patience while our tech team has been reviewing the last set of provided logs.
Moving forward, our techs have informed me that they are noticing that these stops are almost always occurring after Roon is initiating a scheduled rescan of your your ânetwork basedâ watch folders. In light of this, the team has asked if you would kindly perform the following test (see below), to confirm if the rescan could be what is triggering this behavior.
Please edit your âautomatic rescan intervalâ on all network watch folder to: âonly on startupâ.
Restart the application.
Wait while the application finishes rescanning the network watch folder.
OK thanks for looking into this, will do. Is it the number of folders that is the issue? Alternatively, if this is a current âfeatureâ of ROCK, can it be set to rescan at 2am everyday when I wonât notice any issues?
I had to reboot the NUC today (via web interface) as when I returned from work an hour ago the Core could not be found by either the WIN 10 or Android remotes. I think this is the first time this has happened. Restarting Roon Server itself made no difference.
Will report back when it has had a few days soak test.
not a great start to the test: I started a 24h playlist on Kitchen-Pi and Kitchen Sonos. They both stopped after around 30 minutes (approx 9pm). This was around the time that a third stream was started to the Bedroom Sonos. According to the wife, this stopped itself too shortly after starting (âRoon lost control of the playback deviceâ or something to that effect) but she restarted it.
So, it is possible that either:
the start of the third stream to the bedroom Sonos caused the kitchen devices to stop
or something caused all three streams to stop at the same time
or all three stoppages are separate incidents
I have restarted the kitchen streams so weâll see how far they get this time.
Have only had time for limited testing but under the circumstances it may be long enough. After my last post:
Bedroom Sonos playback started approx 9pm 8/11/17
Playback restarted 8/11/17 for both Kitchen Sonos and Kitchen PI @9.30pm approx:
In the morning 9/11/17:
Kitchen Pi had stopped with 20:32 remaining of 24h playlist - so approx 3 and half hours playing time before stopping.
Restarted at 5am 09/11/17 with 24h playlist
Kitchen Sonos had stopped with 19:46 remaining - so approx 4h15m playing time before stopping.
Restarted at 5am 09/11/17 with 24h playlist
Bedroom Sonos was left playing overnight and was still playing
When I got home from work:
09/11/17
Kitchen Pi had stopped with 19:45 remaining so approx 4h15 playing time before stopping at about 9.15am
Bedroom Sonos still playing
Kitchen Sonos still playing
I updated the NASs later that evening so abandoned further testing (as they need a reboot after update). They are normally an update or three behind current, but are now up to date.
Changes made to ROCK seem to have resulted in a more unstable system. As Nas and network have stayed the same, Iâm thinking it is ROCK that has the issues.
So, what to do now?
If this was a Windows machine it would have been put out of its misery by now with a full reinstall, so I am thinking along those lines for ROCK. Is this the best course of action though?
I did a different test yesterday. At around 10am I loaded up a memory stick with a load of Christmas music (well it had to happen soon anyway!) and put this straight into the back of the NUC. The aim of this was to rule out any network or NAS issues. After it was all scanned and identified in Roon I started it on a random play through that selection only to the Kitchen Pi. At the same time I started random plays of other music to both Bedroom and Kitchen Sonos endpoints. All three played for at least 12 hours or so until bedtime. In the morning both Kitchen Pi and Bedroom Sonos had stopped at some point in the night. Kitchen Sonos was still going.
nearly 23 hours later, the Pi is still going strong but both Sonos units stopped during the night (so managed at least 15 hours). Interestingly they both stopped at the very start of their respective tracks. One was in Radio mode and the other was still working through a playlist.
All music was cd or mp3 quality and was fed solely from the USB stick.
@Anthony_B ---- Thank you for the update and more importantly, thank you for the diligence!
I have enabled diagnostics on your account so we can have a closer look into what was going on during the mentioned troubleshooting exercises you had performed.
I will be sure to share the teamâs thoughts/findings once they have completed their analysis.
many thanks. Iâm not sure what the testing proves as it doesnât seem to show anything definitive and of course there could be different failure modes at work here. Anyway, the Pi is still playing 36+ hours later⌠I will try to make a note of any further random stoppages and will leave the 24h rescan set to off for now for the watched folders.
Hi @Anthony_B ââ Thank you again for your patience here while we have been discussing the next steps forward in addressing this behavior you are experiencing. I know how frustrating these types of issues can be so thank you for hanging in there with us.
Moving forward, the test you conducted with the USB stick connected to the NUC is an interesting data point. Our feeling so far has been that the dropouts you are experiencing could possibly be linked to the amount of traffic on your network, or they could be related to the network performance of the NAS, and how fast Roon is able to read its data. If the USB stick works consistently, that would appear to support the latter theory.
These types of issues are always hard to pin down when there are many variables in play and when the behavior itself seems to be semi-sporadic in nature. The most logical step forward hear (in my most humble opinion) is going to be to shrink down your setup as much as possible (all things considered of course) and expand back out to try and suss out the variable(s) that are causing these stops in playback. If I was troubleshooting this, the next step would be to get everything on the floor in the middle of the room: NUC (w/usb attached storage), router, and affected endpoints â the simplest possible setup. If things worked in that configuration I would then add the NAS back to the mix. If that yields a positive result, start to add in whatever complexity was removed, one piece at a time, and see where the problem recurs.
I will tell you that network troubleshooting is probably the most frustrating and painful part of the support we do here. Itâs frustrating for everyone involved, and itâs one of the reasons weâve worked so hard to make sure Roon is robust on all sorts of networks. Eliminating potential causes one-by-one is the only real solution, and myself and our team are ready to enumerate the best places to start. Let me know what you think Anthony.