[Icecast] fallen too far behind, removing
Kurt
Support at lsdcode.com
Tue Aug 9 12:51:17 UTC 2016
Morning all, I set up IceCast several years ago on centos 6. seemed
easy, now I am trying to install it on Centos 7.
For the life of me, I cant do it, I have read several STEP by steps on
how to configure, but I am pounding my head against the wall, I am a
Windows IT guy, thus I am looking for a step by step detail way of doing
it. All the examples I find, fail at some point or another. Any one have
a link to an idiots procedure ?
Kurt, AMR.fm
On 8/2/2016 4:30 AM, Roman wrote:
> Hi,
>
> For some reason I see this in my logs:
>
> [2016-08-02 12:43:02] INFO source/send_to_listener Client 169
> (127.0.0.1) has fallen too far behind, removing
>
> The more interesting is that it is the ffmpeg, that is listening to
> the source stream being removed.
>
> My setup is:
>
> Stream generator:
> DEVA TX, streaming 44,1 khz PCM (pcm16le) stream to icecast2 server
> Transcoding:
> VM with ffmpeg reading this stream from source mountpoint and
> transcodes it to 3 MP3 streams (320, 192, 96 kbps)
>
> Using: Icecast 2.4.2
> ffmpeg: version N-50577-g777b305-static
>
> I've tried two different VMs and two different DEVA TX devices to
> troubleshoot this issue, but I've gave up.
>
> I've also tried to get this stream with ffmpeg first, than transcode
> to same streams. The result was the same.
>
> I actually run two different setups at this moment: production and
> testing. Both run same versions and different HW and VMs are on
> different physical hosts
>
> To avoid total disconnects of clients in production, I run a backup
> mountpoint that plays saved stream file, when source listener gets
> disconnected. At this moment all listeners are being moved to the
> /backup stream and than back, when source listener connects again. To
> reconnect I use a local small php script, that cheks every second if
> ffmpeg stream is running. If not - sends me an e-mail and restarts the
> ffmpeg.
>
> I understand that some listeners could be removed due to latency,
> network errors or something like that. But I'm speaking of ffmpeg main
> stream listener, that connects from same physical network (different
> subnets) in case of Test env and from perfect (if we speak of latency)
> place in productions (same physical devices between listener and deva
> tx, but different subnets)
>
> Any ideas or help to debug this one?
>
> ffmpeg line at testing VM: (for production it generates 3 streams
> instead of 1 here)
>
> #!/bin/sh -e
> su vlc -c "FFREPORT=file=/home/vlc/`date +%Y-%m-%d-%H-%M-%S`-mp3.log
> screen -A -m -d -S ffMP3 /home/vlc/ffmpeg/latest/ffmpeg -f s16le -ar
> 44100 -ac 2 -timeout 2000000 -i http://localhost:8500/source
> -reconnect_at_eof -reconnect_streamed -b:a 192k -compression_level 0
> -ac 2 -ar 44100 -af "volume=3dB" -f mp3 -ice_name 'LiveFM @ 192'
> icecast://source:pass@localhost:8500/LiveFM_320"
>
> icecast.xml
>
> <icecast>
> <limits>
> <clients>1000</clients>
> <sources>3</sources>
> <threadpool>5</threadpool>
> <queue-size>524288</queue-size>
> <client-timeout>30</client-timeout>
> <header-timeout>15</header-timeout>
> <source-timeout>10</source-timeout>
> <burst-on-connect>1</burst-on-connect>
> <burst-size>65535</burst-size>
> </limits>
>
> <mount>
> <mount-name>/backup</mount-name>
> <hidden>1</hidden>
> <server_name>backup</server_name>
> <charset>UTF8</charset>
> <no-yp>1</no-yp>
> </mount>
>
>
>
> <mount>
> <mount-name>/LiveFM_320</mount-name>
> <charset>UTF8</charset>
> <password>secret</password>
> <fallback-mount>/backup</fallback-mount>
> <fallback-override>1</fallback-override>
> <no-yp>1</no-yp>
> </mount>
>
>
>
> --
> Best regards,
> Roman.
>
>
> _______________________________________________
> Icecast mailing list
> Icecast at xiph.org
> http://lists.xiph.org/mailman/listinfo/icecast
---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xiph.org/pipermail/icecast/attachments/20160809/3741e242/attachment.htm>
More information about the Icecast
mailing list