Bug 10392 - Bitrate limiting does not work (with softsqueeze)
: Bitrate limiting does not work (with softsqueeze)
Status: RESOLVED FIXED
Product: Logitech Media Server
Classification: Unclassified
Component: Softsqueeze
: 7.3.1
: PC Windows Server 2003
: -- major (vote)
: Future
Assigned To: Unassigned bug - please assign me!
http://forums.slimdevices.com/showthr...
:
Depends on: 10451
Blocks: 10439
  Show dependency treegraph
 
Reported: 2008-12-19 02:27 UTC by Yuly Milner
Modified: 2009-01-16 10:55 UTC (History)
1 user (show)

See Also:
Category: ---


Attachments
Regular log (447.46 KB, text/plain)
2008-12-19 02:27 UTC, Yuly Milner
Details
Console output part (13.36 KB, text/plain)
2008-12-19 02:28 UTC, Yuly Milner
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Yuly Milner 2008-12-19 02:27:36 UTC
Created attachment 4477 [details]
Regular log

Hello

The problem is described in forums (see url)

After upgrade to 7.3 I tried to stream to my SS. First thing I noticed CPU on SC went 100% for about 20 seconds (taken by lame process). The audio started. About 30 sec before the track end lame kicked in again. However, the first track did not finish - it was cut off about 10 seconds before the end, and the next track started to play. This went on with every track.

I rolled back to 7.2.1 - everything perfect again.

To test I've started the server from command line with player.source=debug. I was unable to capture the console messages into file, so I only have part of them (as log2.txt). Regular log is attached as log.txt. Here is what seems to indicate a problem:

SW: 2008-12-19 12:01:33.923 MoveDataThreadProc for step 2 got 4096 bytes, about
to write data.
SW: 2008-12-19 12:01:33.923 MoveDataThreadProc for step 2 about to call ReadFile
.
SW: 2008-12-19 12:01:33.938 MoveDataThreadProc for step 2 got 1311 bytes, about
to write data.
SW: 2008-12-19 12:01:33.938 MoveDataThreadProc for step 2 about to call ReadFile
.
SW: 2008-12-19 12:01:34.017 MoveDataThreadProc for step 2 failed reading with er
ror 109.
SW: 2008-12-19 12:01:34.017 MoveDataThreadProc for step 2 ending.
SW: 2008-12-19 12:01:34.017 Timeout Process/Thread for step 2 died.
SW: 2008-12-19 12:01:34.017 Tidying up
SW: 2008-12-19 12:01:34.017  Process/thread 2 stopped
SW: 2008-12-19 12:01:34.017 Watchdog expired
SW: 2008-12-19 12:01:34.017 Waiting for process step 0 to terminate
SW: 2008-12-19 12:01:34.017 Waiting for process step 1 to terminate
SW: 2008-12-19 12:01:34.032 Thread for step 2 streamed   1306 blocks totalling 0
051951F (5346591) bytes
SW: 2008-12-19 12:01:34.032 Socketwrapper has terminated.



For me it looks like something makes transcoding work on the whole file at once (hence the high CPU in the beginning), and this is failing.

My version is:
Version: 7.3.1 - 24367 @ Thu Dec 18 10:00:25 PST 2008
Operating system: Windows Server 2003 - EN - cp1251
Platform Architecture: 586
Perl Version: 5.8.8 - MSWin32-x86-multi-thread
MySQL Version: 5.0.22-community-nt

I'm streaming flac to softsqueeze 3.7 on another machine. Bitrate limiting is configured to 160K, level 4.
Comment 1 Yuly Milner 2008-12-19 02:28:14 UTC
Created attachment 4478 [details]
Console output part
Comment 2 Moonbase 2008-12-19 05:43:25 UTC
Wild guess: Did you set Crossfade? If so, can you reproduce this without crossfading?
Comment 3 Yuly Milner 2008-12-19 07:56:28 UTC
(In reply to comment #2)
> Wild guess: Did you set Crossfade? If so, can you reproduce this without
> crossfading?
> 

Made me sit search for a while. In fact crossfade is not even supported for softsqueeze at the squeezecenter side (is this intentional?) In the sofsqueeze menu itself the crossfade is set to "none", so no, this was not the reason.
Comment 4 James Richardson 2008-12-19 10:19:48 UTC
Soft squeeze is now a Community supported add-on, moving but to monitor.

Patches welcome
Comment 5 Yuly Milner 2008-12-23 05:16:08 UTC
So lets see what we have here: Softsqueeze is community supported application; hence no support from Logitech. OK, understood.
Then, here comes Squeezeplay. This is in beta now, so again no (official) support. Can you see where it leads to? There is no supported PC-based client! I do not think this is right way to treat the community!

Anyhow, Squeezeplay shows the same symptoms (see messages below).

SW: 2008-12-23 15:11:35.392 MoveDataThreadProc for step 2 got 4096 bytes, about
to write data.
SW: 2008-12-23 15:11:35.392 MoveDataThreadProc for step 2 about to call ReadFile
.
SW: 2008-12-23 15:11:35.423 MoveDataThreadProc for step 2 got 4096 bytes, about
to write data.
SW: 2008-12-23 15:11:35.423 MoveDataThreadProc for step 2 about to call ReadFile
.
SW: 2008-12-23 15:11:35.470 MoveDataThreadProc for step 2 got 4096 bytes, about
to write data.
SW: 2008-12-23 15:11:35.470 MoveDataThreadProc for step 2 about to call ReadFile
.
SW: 2008-12-23 15:11:35.470 MoveDataThreadProc for step 2 got 723 bytes, about t
o write data.
SW: 2008-12-23 15:11:35.470 MoveDataThreadProc for step 2 about to call ReadFile
.
SW: 2008-12-23 15:11:35.548 MoveDataThreadProc for step 2 failed reading with er
ror 109.
SW: 2008-12-23 15:11:35.548 MoveDataThreadProc for step 2 ending.
SW: 2008-12-23 15:11:35.548 Timeout Process/Thread for step 2 died.
SW: 2008-12-23 15:11:35.548 Tidying up
SW: 2008-12-23 15:11:35.563  Process/thread 2 stopped
SW: 2008-12-23 15:11:35.563 Watchdog expired
SW: 2008-12-23 15:11:35.563 Waiting for process step 0 to terminate
SW: 2008-12-23 15:11:35.563 Waiting for process step 1 to terminate
SW: 2008-12-23 15:11:35.579 Thread for step 2 streamed   1190 blocks totalling 0
04A52D3 (4870867) bytes
SW: 2008-12-23 15:11:35.579 Socketwrapper has terminated.

Comment 6 James Richardson 2008-12-23 08:04:29 UTC
(In reply to comment #5)

> Anyhow, Squeezeplay shows the same symptoms (see messages below).
> 

Thank you for the validation, opening a separate bug 10439 for SqueezePlay, which is a supported software player.
Comment 7 James Richardson 2009-01-07 13:29:11 UTC
Fixed with Change 24415