Bug 11779 - Sync broken on Belkin F5D8236-4 Router when 802.11e QOS is enabled
: Sync broken on Belkin F5D8236-4 Router when 802.11e QOS is enabled
Status: RESOLVED WORKSFORME
Product: Logitech Media Server
Classification: Unclassified
Component: Sync
: 7.3.2
: Macintosh MacOS X 10.5
: -- normal (vote)
: 7.4.0
Assigned To: Ross Levine
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2009-04-13 07:34 UTC by Matt Wise
Modified: 2009-09-08 09:11 UTC (History)
4 users (show)

See Also:
Category: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Matt Wise 2009-04-13 07:34:52 UTC
Over the weekend at my inlaws we were trying to sync an SB3 and Boom playing standard MP3 files. They would work fine for a while but randomly start re-buffering for no apparent reason. The two players both had decent WiFi signal (one at 75%, one at 55%). The Macbook serving them had great signal and was reporting 120+mbit transmit rate to the router. 

After disabling QOS and removing an additional router from the house (an Airport Express B/G model that was acting as an additional access point on one end of the house) all of the problems went away. 

One interesting bit was that the systems seemed extremely responsive 99% of the time... not until just a few seconds before they had to did we see any slowness in the text on the screens. 
rebuffer
Comment 1 Andy Grundman 2009-04-13 07:39:05 UTC
I think we should look at using the DSCP bits in IP packets for our streaming HTTP traffic from SC, if possible.  Possibly also in slimproto packets from the player side.
Comment 2 Chris Owens 2009-04-13 09:51:17 UTC
QA to try to reproduce.
Comment 3 Blackketter Dean 2009-04-13 13:59:05 UTC
Can you attach a sniff?
Comment 4 Ross Levine 2009-06-03 17:17:32 UTC
What are standard mp3 files? I can't reproduce this with 128 or 192kbit. QoS enabled or disabled, 3 IP3K players synced, no problems. 

Wise Matt what are the chances it could have had something to do with the wireless access point?
Comment 5 Ross Levine 2009-06-04 15:42:30 UTC
Talked to Wise Matt, whatever issue occurred may have not been related to QoS. But I opened bug 12275 as a catch-all QoS related bug for ongoing investigation of potential QoS issues.
Comment 6 Matt Wise 2009-06-04 15:47:16 UTC
Agreed... it could have just been a weird situation that cause the  
problem I saw. If QA cannot reproduce this, then lets just keep an eye  
on possible QOS issues in the future.