Bugzilla – Bug 7294
Current song marker doesn't always switch to next page and update song with "notes" marker
Last modified: 2008-12-18 11:12:53 UTC
SC 7 - build 17695, Server on Windows XP, client on Windows Vista, IE 7 With a long playlist it doesn't always switch pages when the current song moves to the second or third pages. The "notes" icon is not displayed on the first page (where it is sitting). Manually clicking to the correct page does display the "notes" icon. Attempting to shut down the browser displays a script error (see attached screen shot).
Created attachment 2981 [details] Screenshot of error message when shutting down browser
Michael: can you take a look at this?
Are the two issues related? Do you only see the error message, when the page changed doesn't happen? I can't reproduce neither issue: tested with Opera, Safari, FF2, IE6 and IE7 on Mac and Windows, skipping songs using a controller to trigger the page change back and forth. It always correctly displayed the song marker, and I never go that error message.
Could you try to reproduce the issue with a different browser? Or the same browser using a different server (even if it's only the IP address instead of localhost)
Works for WXP using FF 2.0.0.12. Page data moved to page 2 and updated song 1 with the notes icon. Scrollbar, however, remained at the bottom. I had to scroll up manually to check if first song was showing as current. This is also works for me with IE7. What I did notice, however, is that with a particularly large playlist, it seems to take a long time for the jsonrpc.js response. I was probably 30 seconds into the track before there was a full update on the playlist.
Doesn't always occur, unfortunately. When I last noticed it, it was about 10 songs into the 3rd page while the browser was on the first page. I'm not 100% on the error message and the out of sync. page being related. I've had the error message on a couple of other occasions but I hadn't noticed if the page was out of sync. or not (and it is too late to check afterwards). On both recent occasions when I've had the page out of sync. occur I've also gotten the error message. I haven't seen a case when the page was out of sync. and I haven't gotten the error. Seems to have a high correlation. All of my playlists (usually, but not always, created by MusicIP) are multiple pages long. I don't usually save them so there were dynamically created one way or another. In this last case I was adding a bunch of recently acquired albums to play them. One thought: screen saver? It would kick in on the client while still on the first page. Client box is Vista, remember - server is XPsp2. Is there any additional logging that I could turn on? I'll try changing the IP address to the server.
kdf - what are particularly long playlists? I don't see any issue with 900 songs, skipping by hundreds. Doug - you mention the screensaver. I doubt it's the cause of the issue. But is the computer going into power saving mode? Is it the same machine which is running SC? I could imagine the browser getting confused when it loses network connection due to power saving. Could this be the issue you're seeing?
In my case, it was just under 800. I was using softsqueeze on a P4 3G with 512M of RAM. It is often noticeably slow, performance wise. Other systems would probably only show this problem with many more tracks in the playlist. I suspect much of the problem I have with that setup is the forced virus scanner which triggers on every access to every file.
5 pages of 50 would be typical - perhaps a bit on the high side. It seems like I keep tripping on some of the more obscure page related issues so I figured I was one of the odd people that did long playlists - perhaps that is an incorrect conclusion. The client and server are 2 different boxes (server is XP). SC is run as a service and there is no current logged in session on the SC box, typically. The screen saver is on the Vista/client box. It certainly had gone into screen saving mode (bubbles), it almost certainly had gone into the power saving mode where it turns off the screen one or more times. I don't believe that it had fully gone to sleep, though, but I can try that to confirm.
Reopen if somebody can find a reproduceable case.
This bug is being closed since it was resolved for a version which is now released! Please download the new version of SqueezeCenter (formerly SlimServer) at http://www.slimdevices.com/su_downloads.html If you are still seeing this bug, please re-open it and we will consider it for a future release.