mammoth-devel Mailing List for Mammoth

I'm also having a try at getting it to build. Latest trunk, latest 
boost, latest tbb. Boost looks like it does some cool stuff, but also 
appears to be a major PITA.

Found one change needed so far:
Error message-> mammoth/include/mammoth/common/pch.hpp:70:26: error: 
boost/pfto.hpp: No such file or directory
changed:
#include <boost/pfto.hpp>
to:
#include <boost/serialization/pfto.hpp>

Now to figure out:
CMakeFiles/mammoth-server.dir/main.cpp.o: In function `T.15022':
main.cpp:(.text+0x11c8): undefined reference to 
`boost::system::get_system_category()'

and

CMakeFiles/mammoth-server.dir/__/core/common/memory_management.cpp.o: In 
function `global constructors keyed to _Znwj':
memory_management.cpp:(.text+0x224): undefined reference to 
`boost::system::get_generic_category()'

I'm hoping we can get Mammoth to function properly, maybe add some 
functionality to it. I'd really like to end up with something that might 
give rtmpd or the commercial servers some competition but with proper 
documentation.


On 11/17/2011 05:05 AM, Javier Castillo II wrote:
> As this is the mammoth-devel list, I must admit I am still working on
> reviving mammoth server for fun and have gotten it to compile and I
> will post to this list once I get the server to load the configuration
> file, or in a couple of days if I get stumped. While debugging it,
> mammoth server is not able to load the plugin directory, and find the
> vod_service Plug-in library, and I am trying to determine if this is
> due to the boost calls I updated from the deprecated function calls.
> I don't have anything against Java, and I agree, when done right, it's
> not that bad, but I will keep Java limited to my home computer, and I
> will continue to work on mammoth as I follow the mindset of "Maybe we
> can do it right." What would life be without a hobby.
> As with any open source project, any help is welcome. Just don't know
> what happened to the project admin.
> Also, just to note, the license for rtmpd is GNU GENERAL PUBLIC
> LICENSE Version 3, 29 June 2007 as of the source for crtmpserver-690
> from their site, and yes, the documentation is lacking.
>

As this is the mammoth-devel list, I must admit I am still working on
reviving mammoth server for fun and have gotten it to compile and I
will post to this list once I get the server to load the configuration
file, or in a couple of days if I get stumped. While debugging it,
mammoth server is not able to load the plugin directory, and find the
vod_service Plug-in library, and I am trying to determine if this is
due to the boost calls I updated from the deprecated function calls.
I don't have anything against Java, and I agree, when done right, it's
not that bad, but I will keep Java limited to my home computer, and I
will continue to work on mammoth as I follow the mindset of "Maybe we
can do it right." What would life be without a hobby.
As with any open source project, any help is welcome. Just don't know
what happened to the project admin.
Also, just to note, the license for rtmpd is GNU GENERAL PUBLIC
LICENSE Version 3, 29 June 2007 as of the source for crtmpserver-690
from their site, and yes, the documentation is lacking.

Em 15/11/2011 16:16, D Mahoney escreveu:
> You mean besides the fact that it's written in Java?
Yes
> I personally would like to have access to a video platform that 
> reasonably documented, has a reasonable license, and isn't condemned 
> to be a resource hog just by virtue of the JVM it sits on top of. No, 
> I'll put in some time helping out with the Mammoth project. Maybe we 
> can do it right.
Java is not too bad, i was a java-hatter in the past, but when it's 
well-done, works very fine.
>
> On 11/15/2011 11:37 AM, Tiago Jacobs - iMDT wrote:
>> Also...  The license of rtmpd is undefined yet (they dont publish 
>> licensing on website).
>>
>> Why not use red5?
>>
>> Tiago
>

You mean besides the fact that it's written in Java? I personally would 
like to have access to a video platform that reasonably documented, has 
a reasonable license, and isn't condemned to be a resource hog just by 
virtue of the JVM it sits on top of. No, I'll put in some time helping 
out with the Mammoth project. Maybe we can do it right.

On 11/15/2011 11:37 AM, Tiago Jacobs - iMDT wrote:
> Also...  The license of rtmpd is undefined yet (they dont publish 
> licensing on website).
>
> Why not use red5?
>
> Tiago

Also...  The license of rtmpd is undefined yet (they dont publish 
licensing on website).

Why not use red5?

Tiago


Em 15/11/2011 04:52, Dan M escreveu:
> The problem with rtmpd is that the documentation is almost nonexistent.
>
> On 11/13/2011 08:24 PM, Tiago Katcipis wrote:
>> I strongly suggest you to take a look at rtmpd:
>>
>> http://www.rtmpd.com/
>>
>> I used mammoth a long time ago but i got several problems running 
>> more that 80 simultaneous connections (like segmentation fault :-), 
>> rtmpd worked perfectly using considerably less CPU/memory.
>>
>> Best Regards,
>> Tiago Katcipis
>>
>> On Sun, Nov 13, 2011 at 4:11 PM, Javier Castillo II 
>> <j.c...@gm... <mailto:j.c...@gm...>> wrote:
>>
>>     Not sure how long this project has remained dormant, but at the
>>     moment
>>     I have checked out the trunk from svn and am working on getting this
>>     to compile. I have updated the boost path calls from the deprecated
>>     functions on my local copy:
>>     http://www.boost.org/doc/libs/1_45_0/libs/filesystem/v3/doc/deprecated.html
>>     At the moment, It looks like the libavformat calls to get the
>>     AVFormatContext values is also broken. I am currently reviewing the
>>     documentation on how to work out these errors:
>>     ./trunk/include/mammoth/stream/av_demuxer.hpp:380:21: error:
>>     'AVFormatContext' has no member named 'track'
>>     and similar for title, genre, etc etc.
>>     I am using ffmpeg 0.8.5 and boost 1.47.0.
>>     I am wondering, is this project dead? If so, I would like to know
>>     why.. not enough time, conflict of interests, etc etc.
>>     Otherwise I plan to get this compiled and continue developing on it.
>>
>>     Other ToDo list:
>>     Work on the cmake files in order to get it to find ffmpeg even
>>     without
>>     pkg-config as my system had ffmpeg installed from the repo, but
>>     didn't
>>     have a .pc file installed for PKGCONFIG to find it.
>>     Workaround: Manually edit the CMakeCache.txt file to have
>>     FFMPEG_INCLUDE_DIR:PATH=FFMPEG_INCLUDE_DIR-NOTFOUND set to
>>     FFMPEG_INCLUDE_DIR:PATH=/usr/include/ffmpeg or wherever your system
>>     path to the includes are.
>>
>>     ------------------------------------------------------------------------------
>>     RSA(R) Conference 2012
>>     Save $700 by Nov 18
>>     Register now
>>     http://p.sf.net/sfu/rsa-sfdev2dev1
>>     _______________________________________________
>>     Mammoth-devel mailing list
>>     Mam...@li...
>>     <mailto:Mam...@li...>
>>     https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>>
>>
>>
>> ------------------------------------------------------------------------------
>> RSA(R) Conference 2012
>> Save $700 by Nov 18
>> Register now
>> http://p.sf.net/sfu/rsa-sfdev2dev1
>>
>>
>> _______________________________________________
>> Mammoth-devel mailing list
>> Mam...@li...
>> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Save $700 by Nov 18
> Register now
> http://p.sf.net/sfu/rsa-sfdev2dev1
>
>
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel

The problem with rtmpd is that the documentation is almost nonexistent.

On 11/13/2011 08:24 PM, Tiago Katcipis wrote:
> I strongly suggest you to take a look at rtmpd:
>
> http://www.rtmpd.com/
>
> I used mammoth a long time ago but i got several problems running more 
> that 80 simultaneous connections (like segmentation fault :-), rtmpd 
> worked perfectly using considerably less CPU/memory.
>
> Best Regards,
> Tiago Katcipis
>
> On Sun, Nov 13, 2011 at 4:11 PM, Javier Castillo II 
> <j.c...@gm... <mailto:j.c...@gm...>> wrote:
>
>     Not sure how long this project has remained dormant, but at the moment
>     I have checked out the trunk from svn and am working on getting this
>     to compile. I have updated the boost path calls from the deprecated
>     functions on my local copy:
>     http://www.boost.org/doc/libs/1_45_0/libs/filesystem/v3/doc/deprecated.html
>     At the moment, It looks like the libavformat calls to get the
>     AVFormatContext values is also broken. I am currently reviewing the
>     documentation on how to work out these errors:
>     ./trunk/include/mammoth/stream/av_demuxer.hpp:380:21: error:
>     ‘AVFormatContext’ has no member named ‘track’
>     and similar for title, genre, etc etc.
>     I am using ffmpeg 0.8.5 and boost 1.47.0.
>     I am wondering, is this project dead? If so, I would like to know
>     why.. not enough time, conflict of interests, etc etc.
>     Otherwise I plan to get this compiled and continue developing on it.
>
>     Other ToDo list:
>     Work on the cmake files in order to get it to find ffmpeg even without
>     pkg-config as my system had ffmpeg installed from the repo, but didn't
>     have a .pc file installed for PKGCONFIG to find it.
>     Workaround: Manually edit the CMakeCache.txt file to have
>     FFMPEG_INCLUDE_DIR:PATH=FFMPEG_INCLUDE_DIR-NOTFOUND set to
>     FFMPEG_INCLUDE_DIR:PATH=/usr/include/ffmpeg or wherever your system
>     path to the includes are.
>
>     ------------------------------------------------------------------------------
>     RSA(R) Conference 2012
>     Save $700 by Nov 18
>     Register now
>     http://p.sf.net/sfu/rsa-sfdev2dev1
>     _______________________________________________
>     Mammoth-devel mailing list
>     Mam...@li...
>     <mailto:Mam...@li...>
>     https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Save $700 by Nov 18
> Register now
> http://p.sf.net/sfu/rsa-sfdev2dev1
>
>
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel

Similarly rtmplite -- http://code.google.com/p/rtmplite/ -- is a very
good solution being well maintained by a very active group.

On Sun, Nov 13, 2011 at 7:24 PM, Tiago Katcipis <kat...@in...> wrote:
> I strongly suggest you to take a look at rtmpd:
>
> http://www.rtmpd.com/
>
> I used mammoth a long time ago but i got several problems running more that
> 80 simultaneous connections (like segmentation fault :-), rtmpd worked
> perfectly using considerably less CPU/memory.
>
> Best Regards,
> Tiago Katcipis
>
> On Sun, Nov 13, 2011 at 4:11 PM, Javier Castillo II
> <j.c...@gm...> wrote:
>>
>> Not sure how long this project has remained dormant, but at the moment
>> I have checked out the trunk from svn and am working on getting this
>> to compile. I have updated the boost path calls from the deprecated
>> functions on my local copy:
>>
>> http://www.boost.org/doc/libs/1_45_0/libs/filesystem/v3/doc/deprecated.html
>> At the moment, It looks like the libavformat calls to get the
>> AVFormatContext values is also broken. I am currently reviewing the
>> documentation on how to work out these errors:
>> ./trunk/include/mammoth/stream/av_demuxer.hpp:380:21: error:
>> ‘AVFormatContext’ has no member named ‘track’
>> and similar for title, genre, etc etc.
>> I am using ffmpeg 0.8.5 and boost 1.47.0.
>> I am wondering, is this project dead? If so, I would like to know
>> why.. not enough time, conflict of interests, etc etc.
>> Otherwise I plan to get this compiled and continue developing on it.
>>
>> Other ToDo list:
>> Work on the cmake files in order to get it to find ffmpeg even without
>> pkg-config as my system had ffmpeg installed from the repo, but didn't
>> have a .pc file installed for PKGCONFIG to find it.
>> Workaround: Manually edit the CMakeCache.txt file to have
>> FFMPEG_INCLUDE_DIR:PATH=FFMPEG_INCLUDE_DIR-NOTFOUND set to
>> FFMPEG_INCLUDE_DIR:PATH=/usr/include/ffmpeg or wherever your system
>> path to the includes are.
>>
>>
>> ------------------------------------------------------------------------------
>> RSA(R) Conference 2012
>> Save $700 by Nov 18
>> Register now
>> http://p.sf.net/sfu/rsa-sfdev2dev1
>> _______________________________________________
>> Mammoth-devel mailing list
>> Mam...@li...
>> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Save $700 by Nov 18
> Register now
> http://p.sf.net/sfu/rsa-sfdev2dev1
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>

I strongly suggest you to take a look at rtmpd:

http://www.rtmpd.com/

I used mammoth a long time ago but i got several problems running more that
80 simultaneous connections (like segmentation fault :-), rtmpd worked
perfectly using considerably less CPU/memory.

Best Regards,
Tiago Katcipis

On Sun, Nov 13, 2011 at 4:11 PM, Javier Castillo II <
j.c...@gm...> wrote:

> Not sure how long this project has remained dormant, but at the moment
> I have checked out the trunk from svn and am working on getting this
> to compile. I have updated the boost path calls from the deprecated
> functions on my local copy:
> http://www.boost.org/doc/libs/1_45_0/libs/filesystem/v3/doc/deprecated.html
> At the moment, It looks like the libavformat calls to get the
> AVFormatContext values is also broken. I am currently reviewing the
> documentation on how to work out these errors:
> ./trunk/include/mammoth/stream/av_demuxer.hpp:380:21: error:
> ‘AVFormatContext’ has no member named ‘track’
> and similar for title, genre, etc etc.
> I am using ffmpeg 0.8.5 and boost 1.47.0.
> I am wondering, is this project dead? If so, I would like to know
> why.. not enough time, conflict of interests, etc etc.
> Otherwise I plan to get this compiled and continue developing on it.
>
> Other ToDo list:
> Work on the cmake files in order to get it to find ffmpeg even without
> pkg-config as my system had ffmpeg installed from the repo, but didn't
> have a .pc file installed for PKGCONFIG to find it.
> Workaround: Manually edit the CMakeCache.txt file to have
> FFMPEG_INCLUDE_DIR:PATH=FFMPEG_INCLUDE_DIR-NOTFOUND set to
> FFMPEG_INCLUDE_DIR:PATH=/usr/include/ffmpeg or wherever your system
> path to the includes are.
>
>
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Save $700 by Nov 18
> Register now
> http://p.sf.net/sfu/rsa-sfdev2dev1
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>

Looks like this method of getting the metadata is now deprecated. I am
currently reviewing the ffmpeg documentation to figure out how to make
the proper call using AVFormatContext::metadata (
http://ffmpeg.org/doxygen/0.6/structAVFormatContext.html#f81395915e863cb72f950ea890c30efd
) and has a method av_metadata_get()
http://ffmpeg.org/doxygen/0.6/metadata_8c-source.html#l00026

This is in relation to the related errors when compiling with related
errors for the additional metadata:
include/mammoth/stream/av_demuxer.hpp:380:21: error: ‘AVFormatContext’
has no member named ‘track’
Here:
                if (m_format_ctx->track != 0)
                        metadata->set_int("track", m_format_ctx->track );
                if (m_format_ctx->title[0] != '\0')
                        metadata->set_string("title", m_format_ctx->title);
                if (m_format_ctx->author[0] != '\0')
                        metadata->set_string("author", m_format_ctx->author);
                if (m_format_ctx->copyright[0] != '\0')
                        metadata->set_string("copyright",
m_format_ctx->copyright);
                if (m_format_ctx->comment[0] != '\0')
                        metadata->set_string("comment", m_format_ctx->comment);
                if (m_format_ctx->album[0] != '\0')
                        metadata->set_string("album", m_format_ctx->album);
                if (m_format_ctx->year != 0)
                        metadata->set_int("year", m_format_ctx->year);
                if (m_format_ctx->genre[0] != '\0')
                        metadata->set_string("genre", m_format_ctx->genre);

Not sure how long this project has remained dormant, but at the moment
I have checked out the trunk from svn and am working on getting this
to compile. I have updated the boost path calls from the deprecated
functions on my local copy:
http://www.boost.org/doc/libs/1_45_0/libs/filesystem/v3/doc/deprecated.html
At the moment, It looks like the libavformat calls to get the
AVFormatContext values is also broken. I am currently reviewing the
documentation on how to work out these errors:
./trunk/include/mammoth/stream/av_demuxer.hpp:380:21: error:
‘AVFormatContext’ has no member named ‘track’
and similar for title, genre, etc etc.
I am using ffmpeg 0.8.5 and boost 1.47.0.
I am wondering, is this project dead? If so, I would like to know
why.. not enough time, conflict of interests, etc etc.
Otherwise I plan to get this compiled and continue developing on it.

Other ToDo list:
Work on the cmake files in order to get it to find ffmpeg even without
pkg-config as my system had ffmpeg installed from the repo, but didn't
have a .pc file installed for PKGCONFIG to find it.
Workaround: Manually edit the CMakeCache.txt file to have
FFMPEG_INCLUDE_DIR:PATH=FFMPEG_INCLUDE_DIR-NOTFOUND set to
FFMPEG_INCLUDE_DIR:PATH=/usr/include/ffmpeg or wherever your system
path to the includes are.

i would recommend you to use rtmpd, mammoth seems to be abandoned.

http://www.rtmpd.com/

best regards,
Katcipis

On Mon, Sep 27, 2010 at 9:22 PM, חץ בן חמו <he...@be...> wrote:

> Hello,
> I just discovered openfms and I'm trying to compile it on my Fedora
> machine, but it fails.
>
> CMake Error: The following variables are used in this project, but they are
> set to NOTFOUND.
> Please set them or make sure they are set and tested correctly in the CMake
> files:
> FFMPEG_INCLUDE_DIR (ADVANCED)
>    used as include directory in directory /home/hetz/mammoth
>    used as include directory in directory /home/hetz/mammoth/src
>    used as include directory in directory /home/hetz/mammoth/src/core
>    used as include directory in directory /home/hetz/mammoth/src/server
>    used as include directory in directory /home/hetz/mammoth/modules
>    used as include directory in directory /home/hetz/mammoth/modules/live
>    used as include directory in directory /home/hetz/mammoth/modules/vod
>
> -- Configuring incomplete, errors occurred!
>
> Is there a way to tell cmake to look at /usr/include/ffmpeg? the files are
> there :)
>
> Also, is there some documentation how to use OpenFMS? if not, I'll be happy
> to ask few more questions and write some docs..
>
> Thanks,
> Hetz
>
>
> ------------------------------------------------------------------------------
> Start uncovering the many advantages of virtual appliances
> and start using them to simplify application deployment and
> accelerate your shift to cloud computing.
> http://p.sf.net/sfu/novell-sfdev2dev
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>


-- 
http://www.getgnulinux.org/windows

Hello,
I just discovered openfms and I'm trying to compile it on my Fedora machine,
but it fails.

CMake Error: The following variables are used in this project, but they are
set to NOTFOUND.
Please set them or make sure they are set and tested correctly in the CMake
files:
FFMPEG_INCLUDE_DIR (ADVANCED)
   used as include directory in directory /home/hetz/mammoth
   used as include directory in directory /home/hetz/mammoth/src
   used as include directory in directory /home/hetz/mammoth/src/core
   used as include directory in directory /home/hetz/mammoth/src/server
   used as include directory in directory /home/hetz/mammoth/modules
   used as include directory in directory /home/hetz/mammoth/modules/live
   used as include directory in directory /home/hetz/mammoth/modules/vod

-- Configuring incomplete, errors occurred!

Is there a way to tell cmake to look at /usr/include/ffmpeg? the files are
there :)

Also, is there some documentation how to use OpenFMS? if not, I'll be happy
to ask few more questions and write some docs..

Thanks,
Hetz

the last update is this one:
http://mammothserver.org/2009/12/27/update/#more-48

after that we couldnt get any answer on this mailing list anymore =/

the last commit is from paulo pizarro that works with me:
http://mammoth.svn.sourceforge.net/viewvc/mammoth/

if you want to take the project to you it will be fine... but i think no one
will help you to develop it, it seems to be abandoned =/.

i still recommend you to contribute with rtmpd, the guy that maintains it is
always on the mailing list and is always trying to help anyone using it.

best regards,
Katcipis

On Sat, Sep 25, 2010 at 12:15 PM, Victor Dudochkin <
dud...@gm...> wrote:

> I have a some ideas of development:
> 1. Implement a libevent support for mammoth. Fast networking like nginx.
> 2. Implement full support of red5 services like a fitcDemo etc. It for easy
> deployment mammoth with standard samples of red5 like a videoconferencing
> and chat.
> 3. Support packaging (ebuild) for gentoo linux.
>
> Dudochkin Victor
>
>
> ------------------------------------------------------------------------------
> Start uncovering the many advantages of virtual appliances
> and start using them to simplify application deployment and
> accelerate your shift to cloud computing.
> http://p.sf.net/sfu/novell-sfdev2dev
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>


-- 
http://www.getgnulinux.org/windows

i do not want to stop you from working on mammoth...but the development
seems to be completely stoped. We used to use it on one of our solutions,
but we have several problems with segmentation fault when running more that
100 clients at the same time on our service.

so we found out rtmpd, it does the same thing with the benefit that a lot of
people are using it and there is active development. Our solution runned
fine with 150 clients on rtmpd and consumed less CPU than mammoth with 100
clients.

http://www.rtmpd.com

best regards,
Katcipis

On Sat, Sep 25, 2010 at 12:06 PM, Victor Dudochkin <
dud...@gm...> wrote:

> I want join to development of mammoth. I patched mammoth and succesefuly
> started it on linux.  Sorry about my bad english. Dudochkin Victor.
>
>
> ------------------------------------------------------------------------------
> Start uncovering the many advantages of virtual appliances
> and start using them to simplify application deployment and
> accelerate your shift to cloud computing.
> http://p.sf.net/sfu/novell-sfdev2dev
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>


-- 
http://www.getgnulinux.org/windows

I have a some ideas of development:
1. Implement a libevent support for mammoth. Fast networking like nginx.
2. Implement full support of red5 services like a fitcDemo etc. It for easy
deployment mammoth with standard samples of red5 like a videoconferencing
and chat.
3. Support packaging (ebuild) for gentoo linux.

Dudochkin Victor

I want join to development of mammoth. I patched mammoth and succesefuly
started it on linux.  Sorry about my bad english. Dudochkin Victor.

mam...@li...

Many thanks to Katcipis, i find the rtmpd which may be well used in my
project,, but the website shut down today, could you please send me the code
and documents if you ever downloaded them. thanks

Let me say clearly about my project. please give me more advises, thanks

Our project is that "Our program processes the video from the CCD camera,
and want to publish the live video(processed) on the internet.the image
after processing is RGB format,

do you ever met this question before?  please give me some suggestions~~
thanks.

Qingyun



On Wed, Jul 7, 2010 at 3:58 PM, 李庆赟 <qin...@gm...> wrote:

> Hi, boys,
>
> How can i run a demo of mammoth, because i have a project need to publish
> the videos in the hard disk to the users in the intranet.
>
> Thanks a lot for your help, if you have a mammoth in running, please give
> me the address, i want to see the result
>
> Rain
>
>
>

i recommend rtmpd:

http://www.rtmpd.com/

we used mammoth on a project here where i work and it gave us a lot of
problems, and the project simply seems to be frozen, rtmpd is being updated
almost daily and is working great on our solution :-).

best regards,
Katcipis

On Wed, Jul 7, 2010 at 4:58 AM, 李庆赟 <qin...@gm...> wrote:

> Hi, boys,
>
> How can i run a demo of mammoth, because i have a project need to publish
> the videos in the hard disk to the users in the intranet.
>
> Thanks a lot for your help, if you have a mammoth in running, please give
> me the address, i want to see the result
>
> Rain
>
>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Mammoth-devel mailing list
> Mam...@li...
> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>
>


-- 
http://www.getgnulinux.org/windows

Hi, boys,

How can i run a demo of mammoth, because i have a project need to publish
the videos in the hard disk to the users in the intranet.

Thanks a lot for your help, if you have a mammoth in running, please give me
the address, i want to see the result

Rain

Hi there,

I've been looking for an rtmp server for a while (and had even started
writing my own) but mammoth seems to suit my needs very well.

I've downloaded and compiled and so far it works well with the exception of
h.264 and aac playback. After some research it appears adobe has changed the
handshake so the open spec handshake will not work with these two formats.

I've read through the red 5 handshake code and updated handshake.hpp and
handshake.cpp to preform this new handshake. I've attached the two files. It
seems to work with aac audio but the video is not displayed. So i dont know
if there's something wrong with what i've written or if there's a problem
with writing the video frame.

Feel free to add / change my code.

I have also emailed ope...@ad... to seek clarification on f4v support
in the open spec.  I haven't heard anything back since last friday though.

I hope it's helped.

David

Hi Alexander,

I realized load tests with mammith and when I enable more than 100/120
streams mammoth dies.
I'll start to analyze this error, then It may be interesting I perform these
tests in version 0.3.
When you will publish the new release? Can I help you with tests?

Thanks,

Paulo Pizarro

Once a while I get the question if this project is still alive.. Yes, it is.
Although the last commit seems like ages ago, I’ve got a whole bunch of
updates waiting for a final commit that will bump mammoth to 0.3.0. In that
is a complete rewrite of the protocol layer, handshaking, amf0/amf3,
addition of a new variant type, working rtmps/rtmpte support and thread
local support used by serialization and context storage. On a protocol level
the output stream is now completely asynchronous and all the ugly red5
leftovers have been removed (things like channel, header, etc..). With so
many changes and additions I’m currently in the process of writing
unit-tests. Although things are looking good, optimistically it will still
take a month (very little time these days) before I can start committing,
please be patient..

Hi Alexander,

I implemented a voip service.
I used the ffmpeg to transcode a rtp stream to flash.
On attachment have a patch file for this service.

To teste:

1) Execute the command above to generate a rtp stream:

ffmpeg -f oss -i /dev/dsp -f rtp -ar 8000 -acodec pcm_mulaw rtp://
127.0.0.1:5000\?localport=4000

2) Create a file "test.sdp" with lines above:

v=0
o=- 0 0 IN IP4 127.0.0.1
t=0 0
s=No Name
c=IN IP4 127.0.0.1
m=audio 5000 RTP/AVP 8
b=AS:64

3) Copy this file to directory: /<mammoth_install>/sdp

4) Add the voip_service to settings.xml file:

     <service>
      <name>voip</name>
      <plugin>voip_service</plugin>
      <storage_folder>/opt/mammoth/sdp/</storage_folder>
      <chunksize>4096</chunksize>       <!-- override the global chunksize
-->
    </service>

5) Open a Flash player, connect to mammoth and play "test.sdp" ;-)

Best regards,

Paulo Pizarro

Hi Alex,Folks

Are there any instructions on how to see a demo of mammoth-server.
Currently I can see if I stream a video using mammoth and use red5-client to
see that video then it does not play that video and gets stuck.

-Srini

Hi Alexander,

2009/5/27 Alexander Wichers <dev...@wi...>

> Hi Paulo,
>
> Thank you for your contribution; your patch has been applied.


Nice.


>
>
> Let me know if you need any further help with understanding the code.


Yes, I need. Do you have a class diagram?

What I need to do first? A rtp_stream derived from stream_base?

Thank you,

Paulo


>
>
> Cheers,
> Alexander
>
>
> Paulo Pizarro wrote:
>
>> Hi Alexander,
>>
>> Congratulations for good job.
>>
>> I'm interested on contributing to the project.
>> My first interest is on the "rtp publishing (using FFmpeg)".
>>
>> I fixed switch behavior for -f and -V command line.
>>
>> Get the patch on the attachment.
>>
>> The switch on main() does not have any data to follow for -f and -V
>> command,
>> so you should not test if argnum+1 < argc.
>>
>> And this patch check if the length of argv is not 2.
>>
>> Invalid options: - f
>>
>> Best regards,
>>
>> Paulo Pizarro
>>
>>
>> ------------------------------------------------------------------------
>>
>>
>> ------------------------------------------------------------------------------
>> Register Now for Creativity and Technology (CaT), June 3rd, NYC. CaT is a
>> gathering of tech-side developers & brand creativity professionals. Meet
>> the minds behind Google Creative Lab, Visual Complexity, Processing, &
>> iPhoneDevCamp as they present alongside digital heavyweights like Barbarian
>> Group, R/GA, & Big Spaceship. http://p.sf.net/sfu/creativitycat-com
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Mammoth-devel mailing list
>> Mam...@li...
>> https://lists.sourceforge.net/lists/listinfo/mammoth-devel
>>
>
>