Re: getwd: read error in ..

[mooreron@my-deja.com]

Allow me to clarify what I have attempted.

If I create a very long subdirectory name,
i.e. "abcdefghijklmnopqrstuvwxyz", and populate it with my source code
and build my library there (a.mklib), I SUCCESSFULLY compile and link.
If, however, a create a long subdirectory pathname,
i.e. "abcd/efgh/ijkl/mnop", and populate it with my source code and
build my library there (a.mklib), I was UNABLE to compile the source.
The error in question, "getwd: read error in .." results.

That is what led me to believe that the problem lies with the number of
subdirectories in the pathname.  Permissions were checked on the
ancestor directories and appeared acceptable (drwxr-xr-x).  I am not
familiar with the phrase "underlying mount points".  Could you clarify,
please?

Thank you.

Ron Moore


In article <0dMf4.79$%%2.252@burlma1-snr2>,
  Barry Margolin <barmar@bbnplanet.com> wrote:
> In article <85o1js$esk$1@nnrp1.deja.com>,  <mooreron@my-deja.com>
wrote:
> >I need to make a trivial modification to existing configured
simulation
> >software.  It was built under Rational VADS 623.53.3 on IRIX 5.3.  We
> >have since upgraded our OS to IRIX 6.5.5 and installed VADS 623.53.7
to
> >make this change.  However, when I issue the "a.make" command, it
> >results in the error message "a.make: getwd: read error in .." being
> >issued prior to compilation.  I tried to shorten the file pathname of
> >the Ada source code and was able to compile the code.  But, I still
get
> >the message "a.ld fatal: getwd: read error in .." during link time.
It
> >appears to me that some file pathname environment variable may need
to
> >be increased.  Is that the case and, if so, which one(s)?
>
> It appears to me that you have a permissions problem, not an
environment
> variable problem -- the pwd command is failing because it can't read
one of
> the ancestor directories.
>
> If you can't find an ancestor directory that's lacking read and
execute
> permissions, check the permissions on the underlying mount points --
these
> sometimes bleed through (although I thought this was only on Suns).
>
> --
> Barry Margolin, barmar@bbnplanet.com
> GTE Internetworking, Powered by BBN, Burlington, MA
> *** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to
newsgroups.
> Please DON'T copy followups to me -- I'll assume it wasn't posted to
the group.
>


Sent via Deja.com http://www.deja.com/
Before you buy.