[rrd-developers] Re: calling rrd_create directly in librrd.so

[Ashok]

Hi,

This happens with the first rrd_xxx call itself. I made sure that optind and opterr
were set to zero before making this call, but it still doesn't work. The other
solution I implemented was to fork a  rrdtool process with "-" argument (so that it
will not exit after each command, but waits for the next command) and write to the
stdin of this process through a pipe from my application process. This works, but
it is difficult to determine if there were any errors unless I read the stdout and
search for the "ERROR" string.

The solution with calling "rrd_xxx" functions within the librrd.so would be nice if
it worked. The interesting thing is calling rrd_fetch_fn() directly(instead of
rrd_fetch())  works. (which again raises the suspicion on the getopt_long() calls.)

Thanks.
Ashok.


paulw at oninit.com wrote:

> Does this happen on the first rrd call or subsequent ones?  If it's
> subsquent you need to be resetting optind and opterr.  I've just been
> through this loop.  The worrying thing is I've found it faster (10%)
> to build a system call and run it than it was to link in the libraries.
>
> Ashok Mandala wrote:
> >
> > Hi,
> >
> > I am trying to use the "C" API  for the rrd functionality to
> > create/update rrd files. I have created the proper argument list that
> > rrd_create expects, but I get a segmentation fault from rrd_create when
> > I try to use it. I have stepped through the code using gdb and have
> > followed it to the point in rrd_create.c where it calls getopt_long() -
> > there it issues a segmentation fault when I try to step into it. I saw
> > some code for getopt_long() definition in getopt1.c included with
> > rrdtool - so this call should have gone to that getopt_long() and the
> > source should have opened up. Is the call binding to the libc
> > getopt_long() ? Has anyone faced this same problem before ?
> >
> > Thanks.
> > Ashok.
> >
> > --
> > Unsubscribe mailto:rrd-developers-request at list.ee.ethz.ch?subject=unsubscribe
> > Help        mailto:rrd-developers-request at list.ee.ethz.ch?subject=help
> > Archive     http://www.ee.ethz.ch/~slist/rrd-developers
> > WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
>
> --
> Paul Watson             #
> Oninit Ltd              # You are only young once
> Tel: +44 1436 672201    # but you can be immature
> Fax: +44 1436 678693    # for ever
> www.oninit.com          #
>
> --
> Unsubscribe mailto:rrd-developers-request at list.ee.ethz.ch?subject=unsubscribe
> Help        mailto:rrd-developers-request at list.ee.ethz.ch?subject=help
> Archive     http://www.ee.ethz.ch/~slist/rrd-developers
> WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi


--
Unsubscribe mailto:rrd-developers-request at list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request at list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi