Send Motion-user mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.sourceforge.net/lists/listinfo/motion-user
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Motion-user digest..."
Today's Topics:
1. Re: On Alarm Recording - No Motion (Javier Marcon)
2. Re: On Alarm Recording - No Motion (Mike Wilson)
3. Re: : track.c compile errors (now link errors) (Jim Dodgen)
----------------------------------------------------------------------
Message: 1
Date: Wed, 29 Apr 2015 14:51:06 -0300
From: Javier Marcon <[email protected]>
Subject: Re: [Motion-user] On Alarm Recording - No Motion
To: Motion discussion list <[email protected]>
Message-ID: <[email protected]>
Content-Type: text/plain; charset=windows-1252
Why don't you publish it on github?, so if another person needs
something similar, he can download it without bothering you for the code.
Regards,
Javier.
El 29/04/15 a las 14:23, Bill Clay escibi?:
> Pablo,
>
> Over the last four years, I have gradually developed a modified version
> of motion v. 3.2.12 that may be able to do what you want. It is not
> "plug and play" and would require you to have sysadmin skills to install
> and C programming skills if you needed to tweak my mods.
>
> A few details:
>
> 1. motion optionally provides pre-event image recording, but Kenneth
> warns against long pre-event capture because it hinders timely response
> to events as motion writes out its buffer of the N frames preceding the
> event (a motion configuration parameter). If your frame rate is modest
> (I run 4 frames/sec), you MIGHT be able to reliably get 5 seconds of
> pre-event recording. I've never tried more than 5 pre-event frames (ca.
> 1 sec.), but IIRC, my mods do slightly improve the logic that tries to
> smooth out the load of catch-up writing of pre-event frames once an
> event is signaled.
>
> 2. I added an external alarm input (via a memory-mapped file) that
> triggers still and/or video recording and on-<event> scripts just as if
> motion had been detected in the images.
>
> 3. The external alarm input requires a separate daemon to signal
> alarms. Mine is written in Python and uses Phidgets USB devices
> (http://www.phidgets.com) to monitor ordinary commercial PIR/microwave
> motion sensors.
>
> 4. Probably the biggest hurdle to your large (for motion) pre-event
> recording requirement is doing database INSERTs for recorded frames
> (optional; ignore this point if you don't need it). In unmodified
> motion, database updates run synchronously in the same thread with image
> capture and recording. Even if your DBMS runs on the same box as motion
> (mine is at a remote site, which makes this a real problem), catch-up
> pre-event DB inserts may substantially slow the frame capture rate
> during the first few seconds of the event, which is surely not what you
> want.
>
> My mods include a new module (about 1 kloc) for efficient (PREPAREd SQL
> and binary, not character, data transmission), high-reliability (robust
> recovery from network outages and DBMS bounces), asynchronous (separate
> thread) DBMS updates, but only for PostgreSQL. I imagine someone
> proficient with MySQL C-language APIs could adapt its asynchronous
> features to that DBMS without too much trouble.
>
> This is probably more complexity than you want to deal with, but if
> you're interested, I can email you a tarball of this modified version of
> motion 3.2.12 and my Python external alarm handler.
>
> Regards,
> Bill Clay
>
> ------------------------------------------------------------------------------
> One dashboard for servers and applications across Physical-Virtual-Cloud
> Widest out-of-the-box monitoring support with 50+ applications
> Performance metrics, stats and reports that give you Actionable Insights
> Deep dive visibility with transaction tracing using APM Insight.
> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
> _______________________________________________
> Motion-user mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/motion-user
> http://www.lavrsen.dk/twiki/bin/view/Motion/WebHome
>
------------------------------
Message: 2
Date: Wed, 29 Apr 2015 18:47:34 +0000
From: Mike Wilson <[email protected]>
Subject: Re: [Motion-user] On Alarm Recording - No Motion
To: Motion discussion list <[email protected]>
Message-ID:
<caaveucoajuomjerm1wfccvqu7bzidpmynw0pvjzin9_u21j...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
I'd like to see it on github as well. It would be nice to pull the threaded
db stuff into the main branch.
On Wed, Apr 29, 2015 at 1:52 PM Javier Marcon <[email protected]>
wrote:
> Why don't you publish it on github?, so if another person needs
> something similar, he can download it without bothering you for the code.
>
> Regards,
>
> Javier.
>
> El 29/04/15 a las 14:23, Bill Clay escibi?:
> > Pablo,
> >
> > Over the last four years, I have gradually developed a modified version
> > of motion v. 3.2.12 that may be able to do what you want. It is not
> > "plug and play" and would require you to have sysadmin skills to install
> > and C programming skills if you needed to tweak my mods.
> >
> > A few details:
> >
> > 1. motion optionally provides pre-event image recording, but Kenneth
> > warns against long pre-event capture because it hinders timely response
> > to events as motion writes out its buffer of the N frames preceding the
> > event (a motion configuration parameter). If your frame rate is modest
> > (I run 4 frames/sec), you MIGHT be able to reliably get 5 seconds of
> > pre-event recording. I've never tried more than 5 pre-event frames (ca.
> > 1 sec.), but IIRC, my mods do slightly improve the logic that tries to
> > smooth out the load of catch-up writing of pre-event frames once an
> > event is signaled.
> >
> > 2. I added an external alarm input (via a memory-mapped file) that
> > triggers still and/or video recording and on-<event> scripts just as if
> > motion had been detected in the images.
> >
> > 3. The external alarm input requires a separate daemon to signal
> > alarms. Mine is written in Python and uses Phidgets USB devices
> > (http://www.phidgets.com) to monitor ordinary commercial PIR/microwave
> > motion sensors.
> >
> > 4. Probably the biggest hurdle to your large (for motion) pre-event
> > recording requirement is doing database INSERTs for recorded frames
> > (optional; ignore this point if you don't need it). In unmodified
> > motion, database updates run synchronously in the same thread with image
> > capture and recording. Even if your DBMS runs on the same box as motion
> > (mine is at a remote site, which makes this a real problem), catch-up
> > pre-event DB inserts may substantially slow the frame capture rate
> > during the first few seconds of the event, which is surely not what you
> > want.
> >
> > My mods include a new module (about 1 kloc) for efficient (PREPAREd SQL
> > and binary, not character, data transmission), high-reliability (robust
> > recovery from network outages and DBMS bounces), asynchronous (separate
> > thread) DBMS updates, but only for PostgreSQL. I imagine someone
> > proficient with MySQL C-language APIs could adapt its asynchronous
> > features to that DBMS without too much trouble.
> >
> > This is probably more complexity than you want to deal with, but if
> > you're interested, I can email you a tarball of this modified version of
> > motion 3.2.12 and my Python external alarm handler.
> >
> > Regards,
> > Bill Clay
> >
> >
> ------------------------------------------------------------------------------
> > One dashboard for servers and applications across Physical-Virtual-Cloud
> > Widest out-of-the-box monitoring support with 50+ applications
> > Performance metrics, stats and reports that give you Actionable Insights
> > Deep dive visibility with transaction tracing using APM Insight.
> > http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
> > _______________________________________________
> > Motion-user mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/motion-user
> > http://www.lavrsen.dk/twiki/bin/view/Motion/WebHome
> >
>
>
>
> ------------------------------------------------------------------------------
> One dashboard for servers and applications across Physical-Virtual-Cloud
> Widest out-of-the-box monitoring support with 50+ applications
> Performance metrics, stats and reports that give you Actionable Insights
> Deep dive visibility with transaction tracing using APM Insight.
> http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
> _______________________________________________
> Motion-user mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/motion-user
> http://www.lavrsen.dk/twiki/bin/view/Motion/WebHome
>
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
Message: 3
Date: Wed, 29 Apr 2015 12:00:34 -0700
From: Jim Dodgen <[email protected]>
Subject: Re: [Motion-user] : track.c compile errors (now link errors)
To: tosiara <[email protected]>, [email protected]
Message-ID:
<CAJ2wgcpa=v8eoxs1jtx4_pbmurq_eserj7y0qqgoykhdr6j...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Thanks,
That worked with ffmpeg 2.6.2, still got a the duplicate symbol with
0.5.15, not a problem no value in using old versions.
BTW the list manager still not sending replies
*Jim Dodgen*
On Wed, Apr 29, 2015 at 10:16 AM, tosiara <[email protected]> wrote:
> To solve linking error add extra libs to motion's Makefile:
>
> -lswresample -lva
>
> On Wed, Apr 29, 2015 at 7:58 PM, Jim Dodgen <[email protected]> wrote:
>
>> Sorry I did not reply sooner, this just arrived via the digest, and I had
>> not gotten a email on it so this will not be in the correct thread.
>>
>> Ok, I put in the include and that got everything to compile. I have built
>> two versions of ffmpeg from source the latest and oldest ones I could
>> find 2.6.2 and 0.5.15.
>>
>> Now during the link I'm getting errors. pretty much related to the
>> version of ffmpeg I'm using Looks like a ffmpeg version problem and
>> here are the link issues with each ffmpeg.
>>
>> What version of ffmpeg is required? and if it is this picky why not pull
>> a good version into the project?
>>
>> Notes: I change ffmpeg versions with a simple "make install". I'm
>> running ubuntu 14.04 LTS. 32 bit. I'm compiling motion from the trunk
>> version 5.6.4
>>
>> here are the link errors.
>>
>> --- using ffmpeg 2.6.2 ---------
>>
>> Linking Motion...
>>
>> --------------------------------------------------------------------------------
>>
>> gcc -L/usr/local/lib -o motion motion.o logger.o conf.o draw.o
>> jpegutils.o vloopback_motion.o video.o video2.o video_common.o netcam.o
>> netcam_ftp.o netcam_jpeg.o netcam_wget.o track.o alg.o event.o picture.o
>> rotate.o webhttpd.o stream.o md5.o ffmpeg.o sdl.o netcam_rtsp.o -lm
>> -lpthread -L/usr/lib/i386-linux-gnu -lSDL -ljpeg -L/usr/local/lib
>> -lavformat -lavcodec -lavutil -lm -lz -lsqlite3
>> /usr/local/lib/libavcodec.a(opusdec.o): In function
>> `opus_decode_subpacket':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:374: undefined reference to
>> `swr_is_initialized'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function `opus_decode_frame':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:220: undefined reference to
>> `swr_is_initialized'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function `opus_init_resample':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:161: undefined reference to
>> `swr_init'
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:167: undefined reference to
>> `swr_convert'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function `opus_decode_frame':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:234: undefined reference to
>> `swr_convert'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function `opus_flush_resample':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:115: undefined reference to
>> `swr_convert'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function
>> `opus_decode_subpacket':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:406: undefined reference to
>> `swr_close'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function `opus_decode_flush':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:567: undefined reference to
>> `swr_close'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function `opus_decode_close':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:589: undefined reference to
>> `swr_free'
>> /usr/local/lib/libavcodec.a(opusdec.o): In function `opus_decode_init':
>> /home/jim/ffmpeg-2.6.2/libavcodec/opusdec.c:642: undefined reference to
>> `swr_alloc'
>> /usr/local/lib/libavcodec.a(vaapi.o): In function
>> `ff_vaapi_commit_slices':
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:101: undefined reference to
>> `vaCreateBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:110: undefined reference to
>> `vaCreateBuffer'
>> /usr/local/lib/libavcodec.a(vaapi.o): In function
>> `ff_vaapi_render_picture':
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:52: undefined reference to
>> `vaUnmapBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:65: undefined reference to
>> `vaBeginPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:69: undefined reference to
>> `vaRenderPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:73: undefined reference to
>> `vaRenderPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:78: undefined reference to
>> `vaEndPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:56: undefined reference to
>> `vaUnmapBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:61: undefined reference to
>> `vaUnmapBuffer'
>> /usr/local/lib/libavcodec.a(vaapi.o): In function
>> `ff_vaapi_commit_slices':
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:101: undefined reference to
>> `vaCreateBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:110: undefined reference to
>> `vaCreateBuffer'
>> /usr/local/lib/libavcodec.a(vaapi.o): In function `alloc_buffer':
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:129: undefined reference to
>> `vaCreateBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:131: undefined reference to
>> `vaMapBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:129: undefined reference to
>> `vaCreateBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:131: undefined reference to
>> `vaMapBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:129: undefined reference to
>> `vaCreateBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:131: undefined reference to
>> `vaMapBuffer'
>> /usr/local/lib/libavcodec.a(vaapi.o): In function `destroy_buffers':
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:38: undefined reference to
>> `vaDestroyBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:38: undefined reference to
>> `vaDestroyBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:38: undefined reference to
>> `vaDestroyBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:38: undefined reference to
>> `vaDestroyBuffer'
>> /usr/local/lib/libavcodec.a(vaapi.o): In function
>> `ff_vaapi_render_picture':
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:52: undefined reference to
>> `vaUnmapBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:65: undefined reference to
>> `vaBeginPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:69: undefined reference to
>> `vaRenderPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:73: undefined reference to
>> `vaRenderPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:78: undefined reference to
>> `vaEndPicture'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:56: undefined reference to
>> `vaUnmapBuffer'
>> /home/jim/ffmpeg-2.6.2/libavcodec/vaapi.c:61: undefined reference to
>> `vaUnmapBuffer'
>>
>>
>> -------------- using ffmpeg 0.5.15 -------------------
>>
>> Linking Motion...
>>
>> --------------------------------------------------------------------------------
>> gcc -L/usr/local/lib -o motion motion.o logger.o conf.o draw.o
>> jpegutils.o vloopback_motion.o video.o video2.o video_common.o netcam.o
>> netcam_ftp.o netcam_jpeg.o netcam_wget.o track.o alg.o event.o picture.o
>> rotate.o webhttpd.o stream.o md5.o ffmpeg.o sdl.o netcam_rtsp.o -lm
>> -lpthread -L/usr/lib/i386-linux-gnu -lSDL -ljpeg -L/usr/local/lib
>> -lavformat -lavcodec -lavutil -lm -lz -lsqlite3
>> /usr/local/lib/libavformat.a(file.o):(.data+0x40): multiple definition of
>> `file_protocol'
>> ffmpeg.o:(.data+0x0): first defined here
>> collect2: error: ld returned 1 exit status
>> make: *** [motion] Error 1
>>
>>
>> Date: Wed, 29 Apr 2015 08:13:31 +0300
>> From: tosiara <[email protected]>
>> Subject: Re: [Motion-user] track.c compile errors
>> To: Motion discussion list <[email protected]>
>> Message-ID:
>> <
>> cachtdwt2xtnlhs-nkmk9n2dmbfp3cmu1kdcs-41m6xchwkc...@mail.gmail.com>
>> Content-Type: text/plain; charset="utf-8"
>>
>> Could be this issue: https://github.com/sackmotion/motion/issues/11
>>
>> Try adding #include <linux/videodev2.h> at the top of track.c
>>
>>
>>
>> On Wed, Apr 29, 2015 at 12:13 AM, Jim Dodgen <[email protected]> wrote:
>>
>> > when i'm doing a make I get a lot of errors. What am I missing?
>> >
>> > ---cut here ---
>> >
>> > track.c: In function ?servo_center?:
>> > track.c:566:18: warning: variable ?ret? set but not used
>> > [-Wunused-but-set-variable]
>> > unsigned int ret = 0;
>> > ^
>> > track.c: In function ?uvc_center?:
>> > track.c:936:29: error: storage size of ?control_s? isn?t known
>> > struct v4l2_control control_s;
>>
>> *Jim Dodgen*
>>
>>
>>
>>
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
------------------------------------------------------------------------------
One dashboard for servers and applications across Physical-Virtual-Cloud
Widest out-of-the-box monitoring support with 50+ applications
Performance metrics, stats and reports that give you Actionable Insights
Deep dive visibility with transaction tracing using APM Insight.
http://ad.doubleclick.net/ddm/clk/290420510;117567292;y
------------------------------
_______________________________________________
Motion-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/motion-user
End of Motion-user Digest, Vol 107, Issue 4
*******************************************
