PDA

View Full Version : What debug boxes to tick



timmorris
2006-03-10, 00:29
I'm finding more and more often that my Squeezebox just stops playing halfway through a track for no apparent reason. I was running 6.2.2, I've now downgraded to 6.2.1. What debug boxes should I select in order to find out what is causing the problem?

Signal strength is only showing between 50 and 60% which I find surprising as the PCs in the room show between very good and excellent.

Tim

timmorris
2006-03-10, 02:40
Log output on fail:

2006-03-10 09:36:35.5306 Backtrace:

frame 0: Slim::Player::Squeezebox2::directHeaders (/PerlApp/Slim/Networking/Slimproto.pm line 413)
frame 1: Slim::Networking::Slimproto::process_slimproto_fra me (/PerlApp/Slim/Networking/Slimproto.pm line 244)
frame 2: Slim::Networking::Slimproto::client_readable (/PerlApp/Slim/Networking/Select.pm line 111)
frame 3: Slim::Networking::Select::select (slimserver.pl line 634)
frame 4: main::idle (slimserver.pl line 38)
frame 5: PerlSvc::Startup (perlsvc.pl line 1481)
frame 6: PerlSvc::_startup (slimserver.pl line 0)
frame 7: (eval) (slimserver.pl line 0)

2006-03-10 09:36:45.8248 00:04:20:06:36:ce: Decoder underrun while this mode: play
2006-03-10 09:36:45.8255 Got a track starting event
2006-03-10 09:36:45.8256 Song 1 had already started, so it's not longer in the queue
2006-03-10 09:36:45.8258 Song 2 has now started playing
2006-03-10 09:36:45.8261 Song queue is now 2
2006-03-10 09:36:48.6720 00:04:20:06:36:ce: Underrun while this mode: play

What does underrun mean?

Tim

timmorris
2006-03-10, 03:34
2006-03-10 10:30:31.6751 00:04:20:06:36:ce: Decoder underrun while this mode: play
2006-03-10 10:30:41.7124 00:04:20:06:36:ce: Underrun while this mode: play

bpa
2006-03-10, 04:46
Buffer Underrun generally means server couldn't deliver packets to the SB in time so the SB's buffer ran out.

If you sending WAV it will load your network the heaviest also if your server uses wireless to the router it will increase utilisation of the wireless.

Enable the network health plugin and look at the buffer fill after playing a few songs.

Edit:
However I think decoder underrun may be produced in the server - if that is the case try enabling d_source and look at the log.

timmorris
2006-03-10, 05:07
I've deliberately used the the box that is in a position to deliver the files quickest as the server as it is hard wired into the switch. I'm trying to listen to FLAC encoded files over the network. The problem doesn't seem to happen if I close down any web browsers pointing at the slim server.

Buffer fill isn't an issue, I'll enable d_source, should I turn anything else on while I'm at it?

Tim

bpa
2006-03-10, 05:18
I'm a bit out of my depth in this "underrun" area.

I've just searched the sources and you should only get those "underrun" messages if you already have d_source enabled. However get a full log (from selection to playing) with d_source and post it may help others to comment.

timmorris
2006-03-10, 05:26
I'll see what it turns up, the error is a bit unpredictable. Is the log visible in the browser the whole thing, and where is it? Stupid question I know but a search for log.txt doesn't turn up much!

I've turned off virus scanning on read to see if that has an impact, but it doesn't seem to. The main culprit appears to be the slim-server web browser.

I'm having fun on my desktop PC ripping CDs using Mareo and 4 different levels of compression ready for some blind testing over the weekend.

Tim

bpa
2006-03-10, 05:54
Turn on debugging d_source from the server settings/ debugging web page.

At the top LHS of the page there is a para of text - click on the hyperlinked word "here" to open the log window.

Alternatively start slimserver from a cmd prompt (dos box) and the log message will appear in the window.

timmorris
2006-03-10, 06:04
I got that far I just thought there might actually be a text based log file somewhere. I have no idea what is causing it. It has nothing to do with loading on the server as I had the task manager open when it crapped out this time, and it was pretty much idle. I've copied more than is needed into a proper text file.

Tim

bpa
2006-03-10, 07:23
I am not familiar with these message and I should see what it looks like on my system but at a guess it looks like the file could be shorter than expected (at 12:47:35.4487) and that may be why there is a decoder underrun.

If that is the case then you should get the underrun error with this file - try playing by itself.



2006-03-10 12:41:49.0958 openSong: opening file F:\Users Shared Folders\Tim\My Documents\My Music\Atomic- The Very Best Of Blondie\Atomic '98 (Tall Paul Remix).mp3
2006-03-10 12:41:49.0964 seeking in 2465 into F:\Users Shared Folders\Tim\My Documents\My Music\Atomic- The Very Best Of Blondie\Atomic '98 (Tall Paul Remix).mp3
2006-03-10 12:41:49.0970 Streaming with format: mp3
2006-03-10 12:41:49.1072 00:04:20:06:36:ce New play mode: play
2006-03-10 12:41:49.1089 00:04:20:06:36:ce: Current playmode: play
2006-03-10 12:41:49.1224 We need to send 0 seconds of silence...
2006-03-10 12:41:49.1226 sending 0 bytes of silence
2006-03-10 12:41:59.1063 Got a track starting event
2006-03-10 12:41:59.1065 Song 19 had already started, so it's not longer in the queue
2006-03-10 12:41:59.1067 Song 20 has now started playing
2006-03-10 12:41:59.1070 Song queue is now 20
2006-03-10 12:47:35.4487 Reduced chunksize to 24666 at end of file (9789530 - 9764864)
2006-03-10 12:47:36.6599 Reduced chunksize to 0 at end of file (9789530 - 9789530)
2006-03-10 12:47:36.6601 end of file or error on socket, opening next song, (song pos: 9789530(tell says: . ), totalbytes: 9789530)
2006-03-10 12:47:36.6603 opening next song...

timmorris
2006-03-10, 07:29
It isn't the file I get the same error randomly whatever I'm playing. I have moved the access point to a higher shelf though so the signal strength is around 65-70% now. What sort of signal strength is required for reliable operation? I was hoping to install these in the house next door which has more than the odd weak spot, certainly weaker than 50% on the Squeezebox display...

If it continues I can hook up a WAP54G Wireless ethernet bridge on a temporary basis just to rule out packet loss. I'd have thought though that it would have shown up as a drop-out rather than just a stop in play.

Tim

bpa
2006-03-10, 07:37
I rechecked the decoder underrun message and it seems to be generated by a message (STMd) from the SB so I would turn the network health plugin to see if it shown anything. Also try running the network test.

To see if network bandwidth has anything to do with it try changing Bit rate limiting under Player Setting / Audio to a smaller value. This will transcode all music to a player down to the rate limit.

timmorris
2006-03-10, 09:34
I'm up to the high 60s for signal strength - the signal strength meter on my pocket PC shows 75% at the same location. My desktop PC, notebook, PocketPC and printer all connect without problems, and I can stream 1.5Mbit videos from the internet over the wireless connection I have to be careful that I don't overlap as we use 3 access points in total (big house) but I'll try changing the channel. Maybe I've just got a duff unit. In order to determine whether or not it is signal strength that is causing the problem I have two avenues I can persue:

1. Use an external adapter

2. Use a long piece of CAT 5 to wire directly into the socket via a 4 port switch - I've got lots of bits of kit lying around.

Is there any particular software build/firmware combo you'd recommend?

Oh, and thanks for all your help. I've asked Kevin Pearsall at Slimdevices to look in to see if he can shed some light.

Quick edit - I can run at 100% all the time at 2k using net-test which is more bandwidth than FLAC takes up.

There is a third option. Leave it and play with the Meridian G68 I have on loan. It's a great piece of kit - so good I've already decided to buy one!

Tim

bpa
2006-03-10, 09:42
From another post I'm guessing it is an SB3 you have so why don't you try the "network test" under the plugins menu. It does a load test in situ at different rates.

As you aware, different devices have different antenna so a good signal for one device does not mean it is good for all - only a test on the SB is really meaningful.

timmorris
2006-03-10, 09:46
Our posts crossed, I've installed the Net-test plugin - I can run at 2k showing 100% all the time, and at 3k at around the mid 90s. Which debug parameter shows buffer use?

Tim

bpa
2006-03-10, 10:00
There are two types of perf test.

1. A feature of SB2/SB3 I think which can be found under plugins menu on the player "Network Test" - this enables you to test different network speeds - sustained.

2. A network health monitor which can be enabled from the Home webpage under Help - "Server & Network Health". This plugins gives details about buffer usage once it is enabled.

timmorris
2006-03-10, 11:49
(1) is an installable plug-in (just a single file) and I've been using (2) since the start which to my mind errs on the pessimistic side, though not I suppose if you want FLAC without timing errors (jitter).

I hope I'm not putting a curse on this but I changed the channel from 6 to 1 and it appears to be OK it has just played an entire FLAC encoded CD for the first time (David Gilmour - On An Island, out this week, if you liked Pink Floyd it's a must buy) Given that is does seem to need a clean signal I think I'm going to have to put another Access Point in the loft as they want three Squeezeboxes next door.

Thanks for all your help today, I do appreciate it, Let's hope I continue to make progress so I can go through the listening tests tomorrow, and then start ripping on Sunday. I'm comparing FLAC, CBR 320k and VBR -0. Given that I've got a Pro Logic II processor which heavily relies on phase if you want to extract multi-channel music from a two channel source and compression destroys phase I've got an inkling I'll be using flac.

I really appreciate all the help I've been given in the last couple of weeks, it really has put a gloss on the initial phase of ownership. I've been made to feel the same about the Meridian G68 - the pre-sales technical support team have spent ages on the phone helping me tweak room correction profiles.

This really is a wandering off topic post. I've always been attracted by the idea of a colour Pronto NG but 500 has just been too much to stump up. I came across a firm in the states that sells Philips refurbished TSU7000 (essentialy the same as an RU980, it just uses a different frequency if you want to use the RF->IR multi-room device. http://www.refurbdepot.com/productdetails2.cfm?Product_ID=3268&refererid=nextag

Tim

Mark Lanctot
2006-03-10, 13:51
I hope I'm not putting a curse on this but I changed the channel from 6 to 1 and it appears to be OK it has just played an entire FLAC encoded CD for the first time (David Gilmour - On An Island, out this week, if you liked Pink Floyd it's a must buy) Given that is does seem to need a clean signal I think I'm going to have to put another Access Point in the loft as they want three Squeezeboxes next door.

If you have a laptop you can do some experiments to find the strongest signal. See http://wiki.slimdevices.com/index.cgi?NetworkProblemsSecondGuide



This really is a wandering off topic post. I've always been attracted by the idea of a colour Pronto NG but 500 has just been too much to stump up. I came across a firm in the states that sells Philips refurbished TSU7000 (essentialy the same as an RU980, it just uses a different frequency if you want to use the RF->IR multi-room device. http://www.refurbdepot.com/productdetails2.cfm?Product_ID=3268&refererid=nextag

Tim

If you're going to be spending that much, look at a Nokia 770, which you can also use for web browsing...

timmorris
2006-03-10, 19:03
I'm not sure whether wireless is the issue - I get 100% at 2k and well over 95 at 3k - the access point is less than 20 feet from the Squeezebox and is fitted with 7dBi high gain antennae.

I already have a PocketPC together with Pocket WinC to check for Wireless dead spots, and the IR on phones and pocketPCs isn't powerful enough for my needs. I already have a B&W Pronto that I've custom programmed, I just thought it might be nice to inject some colour.

The next thing I'm going to try is flashing the Access Point with hacked firmware (which will probably microwave my balls) - if the problem persists then the wireless network isn't the problem. Any other ideas?

Tim

timmorris
2006-03-10, 20:33
I've now increased the output power from 22mW to 84mw - combined with the high gain antennae I'm now transmitting well beyond the EC conformity regs, BUT I have a modern building with foil backed insulation so it isn't going to escape.

The Squeezebox now reports an 84% signal strength. I'll do some tests tomorrow - if it continues to fail it is something else that is causing the dropouts.

Tim

Mark Lanctot
2006-03-10, 21:13
I have read on the DD-WRT forums that increasing signal power is not all that effective. The signal-to-noise ratio remains the same.

High gain antennae are recommended though.

timmorris
2006-03-10, 21:15
Squeezebox
Please queue up several tracks to play on this player and start them playing. Then press the Reset Counters link above to clear the statistics and update this display.

Summary
Control Connection : OK
Streaming Connection : OK
Player Signal Strength : OK
Buffer Fullness : OK
Server Response Time : OK


Warnings
This player is performing normally.


--------------------------------------------------------------------------------

Player Performance : Squeezebox
The graphs shown here record the long term trend for each of the player performance measurements below. They display the number and percentage of measurements which fall within each measurement band.

It is imporant to leave the player playing for a while and then assess the graphs.


Player Signal Strength
This graph shows the strength of the wireless signal received by your player. Higher signal strength is better. The player reports signal strength while it is playing.
< 10 : 0 : 0%
< 20 : 0 : 0%
< 30 : 0 : 0%
< 40 : 0 : 0%
< 50 : 0 : 0%
< 60 : 0 : 0%
< 70 : 0 : 0%
< 80 : 482 : 29% ##############
< 90 : 1152 : 71% ###################################
< 100 : 0 : 0%
>=100 : 0 : 0%
max : 85.000000
min : 71.000000
avg : 79.796818

Buffer Fullness
This graph shows the fill of the player's buffer. Higher buffer fullness is better. Note the buffer is only filled while the player is playing tracks.
Squeezebox1 uses a small buffer and it is expected to stay full while playing. If this value drops to 0 it will result in audio dropouts. This is likely to be due to network problems.

Squeezebox2 uses a large buffer. This drains to 0 at the end of each track and then refills for the next track. You should only be concerned if the buffer fill is not high for the majority of the time a track is playing.

Playing remote streams can lead to low buffer fill as the player needs to wait for data from the remote server. This is not a cause for concern.

< 10 : 50 : 3% #
< 20 : 7 : 0%
< 30 : 7 : 0%
< 40 : 8 : 1%
< 50 : 8 : 1%
< 60 : 7 : 0%
< 70 : 11 : 1%
< 80 : 8 : 1%
< 90 : 15 : 1%
< 100 : 1324 : 92% #############################################
>=100 : 0 : 0%
max : 99.999905
min : 0.000000
avg : 93.901291

Control Connection
This graph shows the number of messages queued up to send to the player over the control connection. A measurement is taken every time a new message is sent to the player. Values above 1-2 indicate potential network congestion or that the player has become disconnected.
< 1 : 1574 :100% ##################################################
< 2 : 0 : 0%
< 5 : 0 : 0%
< 10 : 0 : 0%
< 20 : 0 : 0%
>=20 : 0 : 0%
max : 0.000000
min : 0.000000
avg : 0.000000


--------------------------------------------------------------------------------

Server Performance
The graphs shown here record the long term trend for each of the server performance measurements below. They display the number and percentage of measurements which fall within each measurement band.
Server Response Time
This graph shows the length of time between slimserver responding to requests from any player. It is measured in seconds. Lower numbers are better. If you notice response times of over 1 second this could lead to problems with audio performance.
The cause of long response times could be either other programs running on the server or slimserver processing a complex task.

< 0.002 : 13656 : 87% ###########################################
< 0.005 : 1765 : 11% #####
< 0.01 : 29 : 0%
< 0.015 : 54 : 0%
< 0.025 : 72 : 0%
< 0.05 : 63 : 0%
< 0.1 : 4 : 0%
< 0.5 : 9 : 0%
< 1 : 0 : 0%
< 5 : 0 : 0%
>=5 : 0 : 0%
max : 0.155321
min : -0.016665
avg : 0.001222

Timer Accuracy
Slimserver uses a timer mechanism to trigger events such as updating the user interface. This graph shows how accurately each timer task is run relative to the time it was intended to be run. It is measured in seconds.
Timer tasks are scheduled by the server to run at some point in the future. As only one timer task can run at once and the server may also be performing other activity, timer tasks always run slightly after the time they are scheduled for. However if timer tasks run significantly after they are scheduled this can become noticable through delay in the user interface.

< 0.002 : 4300 : 54% ###########################
< 0.005 : 663 : 8% ####
< 0.01 : 1218 : 15% #######
< 0.015 : 1342 : 17% ########
< 0.025 : 320 : 4% ##
< 0.05 : 84 : 1%
< 0.1 : 13 : 0%
< 0.5 : 2 : 0%
< 1 : 0 : 0%
< 5 : 0 : 0%
>=5 : 0 : 0%
max : 0.108551
min : 0.000000
avg : 0.005037

Timer Task Duration
This graph shows how long each timer task runs for. It is measured in seconds. If any timer task takes more than 0.5 seconds this is likely to impact the user interface.
< 0.002 : 6358 : 80% ########################################
< 0.005 : 1569 : 20% #########
< 0.01 : 4 : 0%
< 0.015 : 1 : 0%
< 0.025 : 6 : 0%
< 0.05 : 0 : 0%
< 0.1 : 4 : 0%
< 0.5 : 0 : 0%
< 1 : 0 : 0%
< 5 : 0 : 0%
>=5 : 0 : 0%
max : 0.054021
min : -0.015232
avg : 0.001009


It is still stopping half way through a track, and I think my room would be a radiation hazard if I pumped the power up any higher.

I'll upgrade to the latest daily build and see if that helps.

timmorris
2006-03-11, 04:12
As per the attached I get a BIG increase in SNR 10dB is huge as it's a logarithmic scale. (well I would attach it if I could do it without jumping through hoops - you'll just have to trust me)

Latest daily build doesn't help, which means:

1. I've got an obscure hardware or software configuration which is allowing a bug to manifest, or

2. I've got a faulty player

I suppose the easy way to check is to install the server on another PC. I'll try later. If that is the case I will need some help tracking down what is going on.

Tim

bpa
2006-03-11, 05:13
I use HyperWRT and increasing power diud not help - I used high gain antenna to fix a poor area.

However your buffer stats in indicate that most of the time signal is fine but there is a 3% where is nearly buffer empty - I would think that is associated with a burst of something disrupting the wireless. Poor wireless gives a spread of values across graph and not the two columns with a big gap in between.

Using a wired connection is the only sure test of this - you could move SB closer to router for the wired test and use earphones.

timmorris
2006-03-11, 06:30
The 3% where nearly empty is a function of the Squeezebox 3 - it empties the buffer when you get to the end of a track.

The only other thing I can think of is altering the FLAC encoding scheme I'm using.

I've uninstalled the server software re-booted and removed any vestigial files and I'm about to re-install. It worked when I first used it so there must be something different.

Re-flashing the firmware has allowed me to correct a problem in another area of the house - a PC with a high gain antenna was struggling to connect at 11Mbits. I use 3 WAP 54s all with High gain antennae to cover a house which is 55m from one end to the other.

Net-stumbler shows a decent increase in SNR - and rather than just ramping it up to 84mW the power output is selectable. I've emailed a mate at Cisco to see if the limit is purely regulatory or if there's a danger of causing harm!

Tim

timmorris
2006-03-11, 07:59
It appears that your first thoughts were correct - un-installing and re-installing the server software with 6.2.1 (which automatically downgraded the firmware in the player to 28) appears to have solved the problem.

I'm a great believer in self help - is there a document somewhere that lists what each of the debug options refers to?

I've said it before - while the software is fantastic the documentation is poor. Someone from Slimdevices needs to sit down and provide a full set of end-user documentation.

Tim

Mark Lanctot
2006-03-11, 08:49
Net-stumbler shows a decent increase in SNR - and rather than just ramping it up to 84mW the power output is selectable. I've emailed a mate at Cisco to see if the limit is purely regulatory or if there's a danger of causing harm!


The limit is regulatory. The danger is that the wireless components will overheat. DD-WRT, the firmware I'm using, recommends 70 mW as safe - the default is 28 but DD-WRT allows it to be set as high as 251 mW!

Given that many on the DD-WRT forums are running 100 mW+, I'd say your 84 mW is safe. Do you notice any additional heat being given off by the router?

timmorris
2006-03-11, 08:54
No excessive heat generation - it isn't noticeably any warmer than standard.

Tim

wickedbob
2006-04-08, 15:08
I'm a great believer in self help - is there a document somewhere that lists what each of the debug options refers to?

I've said it before - while the software is fantastic the documentation is poor. Someone from Slimdevices needs to sit down and provide a full set of end-user documentation.


I absolutely agree with you.
End-user documentation needs some major improvement.

I appear to be getting stuttering with internet radio, local MP3 and FLAC playback and just can't get anything to play through squeezenetworks.

"Official" tech support has been disappointing and hasn't solved my problems. Since I asked for a refund Slimdevices hasn't answered any of my emails.

Thanks for your posts Tim, I will use some of these diagnostic tools and your experiences to diagnose where the problems lie in my household.