19 Jul
2011
19 Jul
'11
1:02 a.m.
Hi all,
I'm trying to build an audio logger- a machine to record audio that can stay
running all the time. I've been attempting to use Rotter and Jack for this,
but I'm getting the same problems across a variety of distros (Arch, Ubuntu,
Ubuntu Studio, CentOs &c.). I'm using an Edirol FA-101, attached to a P4
3ghz machine with a second SATA drive as an audio recording volume. I
launch jack, with a big period so as to avoid xruns:
jackd -v -d firewire -r 44100 -n 3 -p 4096 2> jack_stderr.log >
jack_stdout.log &
Then I launch multiple instances of Rotter, one per channel, six total:
rotter -f flac -c 1 -n channelone -d 500 -L flat -v
/secondsatadrive/channelone
.
.
.
rotter -f flac -c 1 -n channelsix -d 500 -L flat -v
/secondsatadrive/channelsix
Then I launch esjit and patch physical input one to rotter instance
channelone, etc.
This works, for a while. Jackd still gets xruns, but the recording happens
and files are written properly. However, after several hours and a whole
lot of xruns, it stops working. Files are still written and named properly,
but they are only a few hundred k in size and contain no audio. The error
logs show the xruns, but nothing else (no error messages, except for xruns).
The output of Rotter also does not show anything wrong. Clearly, jackd is
breaking somehow; anyone have clues? Any way to get more info from jackd?
What am I doing wrong? This problem is the same across several distros
(see above), and occurs whether or not I use the -R realtime flag.
19 Jul
19 Jul
1:27 a.m.
There is a limit to the size of a wav file. Are you writing a file too
large after the time period?
On Mon, Jul 18, 2011 at 4:02 PM, Eric Steinberg
<eric.steinberg(a)gmail.com> wrote:
Hi all,
I'm trying to build an audio logger- a machine to record audio that can stay
running all the time. I've been attempting to use Rotter and Jack for this,
but I'm getting the same problems across a variety of distros (Arch, Ubuntu,
Ubuntu Studio, CentOs &c.). I'm using an Edirol FA-101, attached to a P4
3ghz machine with a second SATA drive as an audio recording volume. I
launch jack, with a big period so as to avoid xruns:
jackd -v -d firewire -r 44100 -n 3 -p 4096 2> jack_stderr.log >
jack_stdout.log &
Then I launch multiple instances of Rotter, one per channel, six total:
rotter -f flac -c 1 -n channelone -d 500 -L flat -v
/secondsatadrive/channelone
.
.
.
rotter -f flac -c 1 -n channelsix -d 500 -L flat -v
/secondsatadrive/channelsix
Then I launch esjit and patch physical input one to rotter instance
channelone, etc.
This works, for a while. Jackd still gets xruns, but the recording happens
and files are written properly. However, after several hours and a whole
lot of xruns, it stops working. Files are still written and named properly,
but they are only a few hundred k in size and contain no audio. The error
logs show the xruns, but nothing else (no error messages, except for xruns).
The output of Rotter also does not show anything wrong. Clearly, jackd is
breaking somehow; anyone have clues? Any way to get more info from jackd?
What am I doing wrong? This problem is the same across several distros
(see above), and occurs whether or not I use the -R realtime flag.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
--
**** Listen to my CD at http://www.mellowood.ca/music/cedars ****
Bob van der Poel ** Wynndel, British Columbia, CANADA **
EMAIL: bob(a)mellowood.ca
WWW: http://www.mellowood.ca
1:44 a.m.
No, Rotter splits the files at one hour. Average file size is about 52
megabytes. This is, incidentally, the only way I can tell that the signal
chain is hosed/fubar- it starts writing files of 800k or so.
On Mon, Jul 18, 2011 at 4:27 PM, Bob van der Poel <bob(a)mellowood.ca> wrote:
There is a limit to the size of a wav file. Are you
writing a file too
large after the time period?
On Mon, Jul 18, 2011 at 4:02 PM, Eric Steinberg
<eric.steinberg(a)gmail.com> wrote:
Hi all,
I'm trying to build an audio logger- a machine to record audio that can
stay
running all the time. I've been attempting
to use Rotter and Jack for
this,
but I'm getting the same problems across a
variety of distros (Arch,
Ubuntu,
Ubuntu Studio, CentOs &c.). I'm using an
Edirol FA-101, attached to a P4
3ghz machine with a second SATA drive as an audio recording volume. I
launch jack, with a big period so as to avoid xruns:
jackd -v -d firewire -r 44100 -n 3 -p 4096 2> jack_stderr.log >
jack_stdout.log &
Then I launch multiple instances of Rotter, one per channel, six total:
rotter -f flac -c 1 -n channelone -d 500 -L flat -v
/secondsatadrive/channelone
.
.
.
rotter -f flac -c 1 -n channelsix -d 500 -L flat -v
/secondsatadrive/channelsix
Then I launch esjit and patch physical input one to rotter instance
channelone, etc.
This works, for a while. Jackd still gets xruns, but the recording
happens
and files are written properly. However, after
several hours and a whole
lot of xruns, it stops working. Files are still written and named
properly,
but they are only a few hundred k in size and
contain no audio. The
error
logs show the xruns, but nothing else (no error
messages, except for
xruns).
The output of Rotter also does not show anything
wrong. Clearly, jackd
is
breaking somehow; anyone have clues? Any way to
get more info from
jackd?
What am I doing wrong? This problem is the same
across several distros
(see above), and occurs whether or not I use the -R realtime flag.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
--
**** Listen to my CD at http://www.mellowood.ca/music/cedars ****
Bob van der Poel ** Wynndel, British Columbia, CANADA **
EMAIL: bob(a)mellowood.ca
WWW: http://www.mellowood.ca
4:48 a.m.
....and also, I'm writing flac not wav.
On Mon, Jul 18, 2011 at 4:27 PM, Bob van der Poel <bob(a)mellowood.ca> wrote:
There is a limit to the size of a wav file. Are you
writing a file too
large after the time period?
On Mon, Jul 18, 2011 at 4:02 PM, Eric Steinberg
<eric.steinberg(a)gmail.com> wrote:
Hi all,
I'm trying to build an audio logger- a machine to record audio that can
stay
running all the time. I've been attempting
to use Rotter and Jack for
this,
but I'm getting the same problems across a
variety of distros (Arch,
Ubuntu,
Ubuntu Studio, CentOs &c.). I'm using an
Edirol FA-101, attached to a P4
3ghz machine with a second SATA drive as an audio recording volume. I
launch jack, with a big period so as to avoid xruns:
jackd -v -d firewire -r 44100 -n 3 -p 4096 2> jack_stderr.log >
jack_stdout.log &
Then I launch multiple instances of Rotter, one per channel, six total:
rotter -f flac -c 1 -n channelone -d 500 -L flat -v
/secondsatadrive/channelone
.
.
.
rotter -f flac -c 1 -n channelsix -d 500 -L flat -v
/secondsatadrive/channelsix
Then I launch esjit and patch physical input one to rotter instance
channelone, etc.
This works, for a while. Jackd still gets xruns, but the recording
happens
and files are written properly. However, after
several hours and a whole
lot of xruns, it stops working. Files are still written and named
properly,
but they are only a few hundred k in size and
contain no audio. The
error
logs show the xruns, but nothing else (no error
messages, except for
xruns).
The output of Rotter also does not show anything
wrong. Clearly, jackd
is
breaking somehow; anyone have clues? Any way to
get more info from
jackd?
What am I doing wrong? This problem is the same
across several distros
(see above), and occurs whether or not I use the -R realtime flag.
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
--
**** Listen to my CD at http://www.mellowood.ca/music/cedars ****
Bob van der Poel ** Wynndel, British Columbia, CANADA **
EMAIL: bob(a)mellowood.ca
WWW: http://www.mellowood.ca
2:49 a.m.
On 07/18/2011 06:02 PM, Eric Steinberg wrote:
jackd -v -d firewire -r 44100 -n 3 -p 4096 2>
jack_stderr.log >
jack_stdout.log &
Snip...
This works, for a while. Jackd still gets xruns, but
the recording
happens and files are written properly. However, after several hours
and a whole lot of xruns, it stops working. Files are still written and
named properly, but they are only a few hundred k in size and contain no
audio. The error logs show the xruns, but nothing else (no error
How many hours? Is it a regular number?
What happens when jack_frame_time() overflows? (jack_nframes_t, 32-bit
unsigned) At 44.1k this would be about 27 hours.
-gabriel
3:06 a.m.
Interesting, but it's more like 48-72 hours, and sometimes as much as 100
hours before recordings start to break. Others (Roberto, Lieven) on this
list have mentioned integer overflow as a possible culprit, but I've been
pretty consistent in the way I've set up recording, and the time-to-break is
not consistent, i.e. it will sometimes go 4 or 5 days before breaking. What
is consistent is that there will be many xruns and then things will break.
I have not done hdparm on the disk used for recording, and I am not
currently using an RT kernel (although I have used the Ubuntu Studio distro,
with the same results as above).
On Mon, Jul 18, 2011 at 5:49 PM, Gabriel Beddingfield <gabrbedd(a)gmail.com>wrote;wrote:
On 07/18/2011 06:02 PM, Eric Steinberg wrote:
jackd -v -d firewire -r 44100 -n 3 -p 4096 2>
jack_stderr.log >
jack_stdout.log &
Snip...
This works, for a while. Jackd still gets xruns, but the recording
happens and files are written properly. However,
after several hours
and a whole lot of xruns, it stops working. Files are still written and
named properly, but they are only a few hundred k in size and contain no
audio. The error logs show the xruns, but nothing else (no error
How many hours? Is it a regular number?
What happens when jack_frame_time() overflows? (jack_nframes_t, 32-bit
unsigned) At 44.1k this would be about 27 hours.
-gabriel
______________________________**_________________
Linux-audio-user mailing list
Linux-audio-user@lists.**linuxaudio.org<Linux-audio-user@lists.linuxaudio.org>
http://lists.linuxaudio.org/**listinfo/linux-audio-user<http://lists.lin…
4:48 a.m.
On Mon, Jul 18, 2011 at 9:06 PM, Eric Steinberg
<eric.steinberg(a)gmail.com> wrote:
Interesting, but it's more like 48-72 hours, and
sometimes as much as 100
hours before recordings start to break.
if you're using WAV and recording 2 channels, this is impossible.
you absolutely cannot record more than 2^31 bytes into a WAV file.
ever, under any circumstances, with any software.
5:01 a.m.
I'm not. As I mentioned above, rotter splits the file each hour, and I'm
recording flac. The files (when they're good) are about 50 MB each.
On Jul 18, 2011 7:48 PM, "Paul Davis" <paul(a)linuxaudiosystems.com> wrote:
On Mon, Jul 18, 2011 at 9:06 PM, Eric Steinberg
<eric.steinberg(a)gmail.com> wrote:
Interesting, but it's more like 48-72 hours,
and sometimes as much as 100
hours before recordings start to break.
if you're using WAV and recording 2 channels, this is impossible.
you absolutely cannot record more than 2^31 bytes into a WAV file.
ever, under any circumstances, with any software. 
5:22 a.m.
Paul Davis wrote:
On Mon, Jul 18, 2011 at 9:06 PM, Eric Steinberg
<eric.steinberg(a)gmail.com> wrote:
Err, actually, up to 2^32+7 bytes is ok for WAV [0]. More than
that and you have an invalid file. Many programs will produce
an invalid WAF file if you ask them :-).
Erik
[0] A WAV file consists or four bytes 'RIFF' followed by an uint32_t
with the length of the file following the length field. Biggest
number in a uint32_t is 0xffffffff so the file can be that big
plus 8 bytes for 'RIFF' and the length.
--
----------------------------------------------------------------------
Erik de Castro Lopo
http://www.mega-nerd.com/
Interesting, but it's more like 48-72 hours,
and sometimes as much as 100
hours before recordings start to break.
if you're using WAV and recording 2 channels, this is impossible.
you absolutely cannot record more than 2^31 bytes into a WAV file.
ever, under any circumstances, with any software.
4813
days inactive
4814
days old
linux-audio-user@lists.linuxaudio.org
8 comments
5 participants
participants (5)
-
Bob van der Poel -
Eric Steinberg -
Erik de Castro Lopo -
Gabriel Beddingfield -
Paul Davis