On Mon, Apr 25, 2011 at 05:54:24PM -0500, Ivan Kuznetsov wrote:
> > Could you try
> > to run ghostscript manually to convert it to pdf (but omit the `-q'
> > flag)? Like, just run
> >
> > gs -dSAFER -dDEVICEWIDTHPOINTS=612.00 -dDEVICEHEIGHTPOINTS=792.00 \
> > -dCompatibilityLevel=1.4 -dNOPAUSE -dBATCH -r1200 \
> > -sDEVICE=pdfwrite -sOutputFile="./song02.pdf" -c .setpdfwrite \
> > -f "song02.ps"
> >
> > from your shell.
>
> I just ran the above and song02.pdf was created with no errors.
> So only difference between the above command and what lilypond
> executes is the "-q" flag I see. Why does the -q flag now fail?
I don't think it's the `-q' that causes the failure, because this
flag only makes ghostscript quiet (no startup message, no progress
information written to stdout).
My best guess is that the newer ghostscript (when run from within
lilypond) hits some resource limit like open file descriptors or
memory. But in this case it *should* have printed *some* error
message to stderr, and I've no idea why it didn't in the output you
pastet in your original mail.
Assuming that you're running a shell like ksh or bash, you can display
the current limits for file descriptors with
ulimit -n
and the data size limit with
ulimit -d
If those are rather small, try to increase them. For example 512 file
descriptors and 1GB data size:
ulimit -Sn 512
ulimit -Sd 1048576
But please note that this is just a wild guess. It may help or not.
Ciao,
Kili
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user
