Bug 9278 - FLAC playback sometimes skips a track
: FLAC playback sometimes skips a track
Status: RESOLVED WORKSFORME
Product: Logitech Media Server
Classification: Unclassified
Component: Streaming From SlimServer
: 7.1
: PC Windows Vista
: -- normal (vote)
: Investigating
Assigned To: Squeezebox QA Team email alias
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2008-08-24 02:47 UTC by James Pearce
Modified: 2009-09-15 18:29 UTC (History)
0 users

See Also:
Category: ---


Attachments
Fragment of server.log showing problem (9.00 KB, text/plain)
2008-08-24 02:47 UTC, James Pearce
Details
Full server log showing the problem occuring (49.99 KB, application/x-zip-compressed)
2008-08-27 14:14 UTC, James Pearce
Details

Note You need to log in before you can comment on or make changes to this bug.
Description James Pearce 2008-08-24 02:47:53 UTC
Created attachment 3875 [details]
Fragment of server.log showing problem

This is a problem I've only seen happen with FLAC, but that's not to say it doesn't occur with MP3 under the right conditions.

My player is a SqueezeBox 3, connected via Cat5. Firmware is reported as 101.

If I skip through FLAC tracks, sometimes one of them won't play. The player will hang for 5 seconds and then play the audio from the next track, even though it reports it is playing the correct track.

I've run the network test - all seems OK.
The hang can occur on any FLAC file, so it doesn't appear to be a corrupt file problem.

I've attached part of my server log for the period in which the problem occurred. The pertinent lines seem to be:

08-08-24 10:21:33.6753] Slim::Player::Source::readNextChunk (2460) Reduced chunksize to 18142 at end of file (24692446 - 24674304)
[08-08-24 10:21:33.6760] Slim::Player::Source::readNextChunk (2460) Reduced chunksize to 0 at end of file (24692446 - 24692446)
[08-08-24 10:21:33.6765] Slim::Player::Source::readNextChunk (2527) end of file or error on socket, opening next song, (song pos: 24692446(tell says: . 24692446), totalbytes: 24692446)
[08-08-24 10:21:33.6768] Slim::Player::Source::gotoNext (1350) Opening next song...



SqueezeCenter Information
-------------------------
SqueezeCenter Version: 7.1 - 22170 @ Mon Jul 28 13:36:57 PDT 2008 - Windows Vista - EN - cp1252
Server IP address: 192.168.1.69
Perl Version: 5.8.8 MSWin32-x86-multi-thread
MySQL Version: 5.0.22-community-nt
Platform Architecture: 586
Hostname: optimus
Server Port Number: 9000
Total Players Recognized: 1

Player Information
------------------
Name: Squeezebox
Model: Squeezebox v3
Firmware: 101
The IP address for this player is: 192.168.1.68:21387
The Ethernet MAC address for this player is: 00:04:20:12:6e:4a
Comment 1 Chris Owens 2008-08-25 10:03:25 UTC
James could you attach the full log, and make sure that os.files, player.firmware, and player.display and player.source are turned on.  Thanks!
Comment 2 James Pearce 2008-08-27 14:14:03 UTC
Created attachment 3904 [details]
Full server log showing the problem occuring
Comment 3 James Pearce 2008-08-27 14:19:30 UTC
I've just uploaded an attachment showing the problem occurring. Anecdotally, with more logging turned on, it took longer for the problem to occur.

The problem occured at 22:10:25, towards the end of the log, as I skip through tracks and get to "Sweet Harmony (Streelife DJs Remix)". The player shows this as playing, but you can see from the logs that something bad happens, and it ends up playing "Push Up Word Up (Ashton AC Slater Remix)"

I've tried swapping the ethernet cable, but the same problem occurs.
Comment 4 James Richardson 2008-08-28 11:48:22 UTC
Thanks for the logs James, assigning to unassigned so this can be discussed at the next bug meeting.
Comment 5 James Richardson 2008-09-10 13:10:28 UTC
Please re-test with 7.2 release version and post new logs if you still see the error.
Comment 6 Chris Owens 2008-09-10 16:11:36 UTC
Andy notes there's nothing interesting in the log.  Several people suggest that this is a duplicate of another bug where skipping through FLAC files causes an off-by-one error somewhere.
Comment 7 Chris Owens 2008-09-10 16:12:44 UTC
Alan notes that if there is a decoder underrun event without a track start event that would be an interesting log entry.
Comment 8 Chris Owens 2008-09-15 09:44:07 UTC
On hold waiting for the reporter's comments.
Comment 9 James Richardson 2008-10-02 07:38:01 UTC
Moving monitor bugs