[abcde] Re: Bug#70102: abcde: will not start cdparanoia on /dev/scd1 if cdparanoia running on /dev/scd0

[ferret@phonewave.net]

Heh. Lunar phase probably explains why two cdparanoia processes work when
started manually. Anyway, cdparanoia will halt after the
	Report bugs to paranoia@xiph.org
	http://www.xiph.org/paranoia/
splash until the other running cdparanoia has exited. It's just weird that
I can't duplicate it manually.

On Sun, 27 Aug 2000, Robert Woodcock wrote:

> On Sun, Aug 27, 2000 at 02:06:50PM -0700, idalton@ferret.phonewave.net wrote:
> > Package: abcde
> > Version: 1.0.1.1-1
> > Severity: normal
> > 
> > I'm running two instances of abcde from two shells. My drives are
> > /dev/scd0 and /dev/scd1.
> > 
> > If I start the instances so that the instance on /dev/scd1 begins before
> > the one on /dev/scd0, there will be two instances of cdparanoia ripping
> > two audio discs. However, if the instance on /dev/scd0 starts first, the
> > one on /dev/scd1 will block after displaying the splash screen but
> > before ripping.
> > 
> > If I manually start two cdparanoia sessions from the shell instead of
> > using abcde, both will rip irrespective of which one begins first.
> 
> Abcde doesn't wait for other processes that happen to have the same name to
> finish - it keeps all of that independant using subshells and pipes.
> 
> I've seen this when using abcde on a CD tower (essentially 7 old cdrom
> drives on a scsi bus) - only three or four can get scsi bus access
> simultaneously, and how fast the whole thing goes is determined by lunar
> phasing.
> 
> The command line abcde uses to call cdparanoia is:
> 
> cdparanoia -d $CDROM $UTRACKNUM "$WAVDATA"
> 
> or for example:
> 
> cdparanoia -d /dev/scd0 3 abcde.currenttrack.Uu4GtE.wav
> 
> It's probably a kernel issue. You could ask the cdparanoia people about it but
> they already have enough reason to hate the linux scsi code. :)
> -- 
> Robert Woodcock - rcw@debian.org
> "And yet people take these numbers as if they're written in stone."
> 	-- George Madaus, professor of educational testing and public policy
> 	at Boston College, on test scores
> 


To unsubscribe: echo unsubscribe | mail abcde-request@whimsica.lly.org