similar to: On demand relay in icecast 2?

> On Sun, 2004-02-15 at 19:52, Joern Nettingsmeier wrote: > > Michael Smith wrote: > > > On Thursday 12 February 2004 03:05, v b wrote: > > > > > >>I'm using this on icecat 1.3.12 but it doesn't seem to work with icecast > > >>2.0.0. The relay starts as soon as icecast is started and does not drop, > > >>even if there are no

I think this might be a problem: Say I have a master stream called /aac with 20 max listeners I also have a on-demand relay with (master-redirect) enabled. Now the plan is, when the master stream is full it would redirect to the slave stream (a 302), but the slave stream cannot connect to the master as its currently full. So somehow the icecast server has to allow the relay stream to connect

Thanks for the suggestion. It turns out that debug for ices told me nothing but debug for icecast 2.2.0 showed that it was terminating the source at the same place in the playlist, apparently due to a lack of trailing metadata in one particular file -- icecast saw it as end of stream. After checking code updates on the TRAC system, I installed the kh branch of icecast (kh13) and this seems

Michael Smith wrote: > On Thursday 12 February 2004 03:05, v b wrote: > >>I'm using this on icecat 1.3.12 but it doesn't seem to work with icecast >>2.0.0. The relay starts as soon as icecast is started and does not drop, >>even if there are no clients. I've been over and over the docs and can't >>figure out what I'm doing wrong. >> >

here's my icecast.xml <icecast> <limits> <clients>10</clients> <sources>2</sources> <threadpool>5</threadpool> <queue-size>102400</queue-size> <client-timeout>30</client-timeout> <header-timeout>15</header-timeout> <source-timeout>10</source-timeout> </limits> <authentication> <!--

Hello, We are searching for a possibility of "aliassing without wasting bandwidth". It is described in the following section which we found at http://open4all.info/ossa/_Icecast_Streaming_Server. # What happens is that when a listeners connects to your server and # requests the /laika stream, then your icecast server will connect # as a client to my machine, and then feed my stream to

On Thu, 2005-02-17 at 02:42, Geoff Shang wrote: > Hi, > > Speex does belong to the Xiph family of codecs. > > My understanding is that Icecast can stream Ogg Speex. I'm not aware of > any streamers that can stream it to Icecast though. I would think that at > least Ogg123 (compiled with Speex support) and Illiminable's DirectShow > filters would be able to

I run icecast 2.0.0 on freebsd 5.2.1 and an avermedia 206 studio. Briliant :-) Yesterday I gave icecast-2.0-kh34 a try. Starts extrem faster than version 2.0.0 and i could set metatags for ogg. ... but it coredumps every 10 secs. Could i help with maybe a dump or a hand for testing. would be a plesure to me :-) regards Karl-Heinz --- >8 ---- List archives: http://www.xiph.org/archives/

With multiple relay slaves and a growing list of streams, I am looking for ways to conserve bandwidth. The current behavior has each relay slave "pulling" each stream from the master server. Is anyone working on offering a modification to this behavior where the relay slaves will only pull the stream from the master as needed? So the slave would always be aware of the streams'

On Mon, 06 Dec 2004 14:42:59 +0000, Karl Heyes wrote: > On Mon, 2004-12-06 at 09:31, adam wrote: >> alo >> >> i have seen some mention of an ices patch with support for jack...anyone >> know where i can get this? > > The latest is in SVN under http://svn.xiph.org/icecast/branches/kh and > tarballs are available in http://mediacast1.com/~karl. You will need

Geoff Shang wrote: > Kelvin Chu wrote: > >> After reading the docs, I was under the impression that icecast2 does >> not do on demand relaying? > > > Oh, on demand *relaying*! different kettle of fish. > > On demand relaying is available in Karl's branch and has just been > checked into the mainline development code. > > Geoff. > Tarballs

Hello all, I am trying out version 2.3RC3 but it seems it has the same bug as all versions before (including 2.2 KH9 which I used last) When trying to relay an on-demand stream with icecast and also limiting, I always find that icecast counts completely wrong. The first listener is not counted, but icecast connects to the master server, gets the stream and relays. If a second client

> On Sun, 2004-02-15 at 19:52, Joern Nettingsmeier wrote: > > Michael Smith wrote: > > > On Thursday 12 February 2004 03:05, v b wrote: > > > > > >>I'm using this on icecat 1.3.12 but it doesn't seem to work with icecast > > >>2.0.0. The relay starts as soon as icecast is started and does not drop, > > >>even if there are no

Hello All, I would like to run a script whenever a source connects to my icecast server, somelike on source connect run xyz script. is this possible? What i would like to achive is to have an email sent out whenever a source connects and also update a database with the fact it connected. any hints or tips on how to achive this would be welcomed many thanks in advance stuart nixon

hi, i am desperately trying to get this version of ices up and running on debian etch. i used the apt repository at http://www.fbriere.net/debian/ices2/ but i found this also affects the "official" source package at http://mediacast1.com/~karl/ when i try to run an edited version of either the example file provided with the source package or the deb at /usr/share/doc/ices2-kh/ i

alo i have seen some mention of an ices patch with support for jack...anyone know where i can get this? adam

On Aug 29, 2004, at 8:54 PM, Karl Heyes wrote: > The hidden tag applies to <mount> sections, and therefore works with > relays or source clients. BTW I've done some more changes to the svn > branch to better handle the hidden mounts, so either use svn or wait > for > kh53 Karl, are there any appreciable differences between the KH branch and what could be considered

> > just so you know, --enable-vorbis-updates is only a compile flag for the KH > > branch, not the mainline trunk. And I think it would be great to have this > > code be factored out of KH and integrated into the mainline trunk...This > > also seems like an odd thing to have as a compile flag, what was the > > motivation for that ? oddsock > That switch

On Mon, 2005-09-19 at 21:43, Milos Wimmer wrote: > > If there's no fallback link from /cro-region-256.ogg to > > /z-cro-region.ogg then it's wrong, but I'd prefer to see the xml and > > error log (level 4) before saying it's a bug. If there is a fallback > > from /cro-region-256.ogg to /cro-region.ogg with override enabled then > > it could happen.

In message "[icecast] icecast-2.0-kh34" on 24.04.2004, Karl Heyes <karl@xiph.org> writes: KH> On Fri, 2004-04-23 at 09:58, Wild Karl-Heinz wrote: KH> ... [snip] >> ... but it coredumps every 10 secs. KH> There is a segv bug in kh34 triggered when clients move from one mount KH> to another. That should be fixed now in kh35. the result with gdb .... (gdb)