Re: [Queue-developers] HP-UX queued.c compile patch for 1.12.9

[Werner K. <wer...@ya...>]

The patch should fix the queued.c compile problem, but there is still the
./configure hang problem on HP-UX.

I give instructions on how to properly fix the ./configure problem in the
support information below by tweaking `configure.in' and running GNU 'autoconf'
on HP-UX, (or on RedHat Linux where it is usually already installed --- output
is platform independent.)

Alternatively, you can just type Control-] to kill telnet in the ./configure;
this will get around the problem as well (but may misdiagnose identd support.)

"Werner G. Krebs" wrote:

> The following patch should fix the HP-UX queued.c compile time problem.
>
> The patch is for 1.12.9.
>
> The patch can be retrieved at:
> http://sourceforge.net/patch/?func=detailpatch&patch_id=100428&group_id=5605
>
> _______________________________________________
> Queue-developers mailing list
> Que...@li...
> http://lists.sourceforge.net/mailman/listinfo/queue-developers

> Support Request #101868, which you submitted on 05/18/00 14:21
> has been updated. You can respond by visiting:
> http://sourceforge.net/support/?func=detailsupport&support_id=101868&group_id=5605
>
> Category: None
> Status: Open
> Priority: 5
> Summary: Configure & Compile HPUX 10.20
>
> By: mbrown
> Date: 05/19/00 09:15
>
> Message:
> Logged In: YES
> user_id=33778
> Browser: Mozilla/4.7 [en] (X11; I; HP-UX B.10.20 9000/785)
>
> I did the autoconf, and the configure seemed to go okay.
> This time, configure said that I did not have a working
> identd on port 113 (non-root can use netstat on my system).
> After my original modification to the configure script, it
> said I had a working identd on port 113.
>
> When I ran gmake, it gave an error since line 3860 of the
> configure was missing an if statement:
> then cat >> confdefs.h <<\EOF
>
> Replacing this with the corresponding line from the original
> configure script, gmake got to the same point as originally.
>
> Mike
>
> ----------------------------------------------------------------------
>
> By: wkrebs
> Date: 05/19/00 06:51
>
> Message:
> Logged In: YES
> user_id=32209
> Browser: Mozilla/4.51 [en] (X11; I; Linux 2.2.5-15smp i686)
>
> OK, the problem with telnet in ./configure is known on some
> systems. If you look in the CVS repository, you'll see that
> 1.12.7 used netstat instead of telnet to try to determine
> this. However, netstat didn't work on some systems because
> ordinary users lacked the privileges to run it.
>
> Find out if non-root users can run netstat under HP-UX. If
> so, the patch is
> to change the following TWO lines in configure.in:
>
> --Cut here
> dnl AC_CACHE_CHECK(for working identd service on port 113,
> queue_cv_service_identd, if netstat -nat | grep ':113.*:'
> >/dev/null 2>&1 ; then queue_cv_service_identd=yup; else
> queue_cv_service_identd=nope; fi)
>
> AC_CACHE_CHECK(for working identd service on port 113,
> queue_cv_service_identd, if eval "echo |  telnet localhost
> 113 2>/dev/null | grep -i Connected >/dev/null 2>&1" ; then
> queue_cv_service_identd=yup; else
> queue_cv_service_identd=nope; fi)
> if eval "test $queue_cv_service_identd = yup"; then
> AC_DEFINE(HAVE_IDENTD)
> fi
> ---Cut here
> TO:
> ---Cut here
> AC_CACHE_CHECK(for working identd service on port 113,
> queue_cv_service_identd, if netstat -nat | grep ':113.*:'
> >/dev/null 2>&1 ; then queue_cv_service_identd=yup; else
> queue_cv_service_identd=nope; fi)
>
> dnl AC_CACHE_CHECK(for working identd service on port 113,
> queue_cv_service_identd, if eval "echo |  telnet localhost
> 113 2>/dev/null | grep -i Connected >/dev/null 2>&1" ; then
> queue_cv_service_identd=yup; else
> queue_cv_service_identd=nope; fi)
> if eval "test $queue_cv_service_identd = yup"; then
> AC_DEFINE(HAVE_IDENTD)
> fi
> --- Cut here
>
> Of course, then you need to re-run GNU autoconf to
> regenerate ./configure on your system. (autoconf is now
> installed by default on many Redhat systems, and the output
> is machine-independent; you can also get it from the FSF if
> you don't have RedHat lying around.
>
> This should solve the first problem; once this is fixed,
> we'll take it from there.
>
>