Re: RH 6.0 (fwd)

===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================

---------- Forwarded message ----------
Date: Wed, 14 Jul 1999 11:15:01 -0400 (EDT)
From: Rob Cermak <address@hidden>
To: Robb Kambic <address@hidden>
Subject: Re: RH 6.0 

Found it.

I'm sending you the RH6.0 /usr/include/rpc/auth.h, looks like
the header file (line 180) directly sets AUTH_KERB to 4 -- the
source of the problem.

/usr/include/rpc/auth.h:180:#define AUTH_KERB       4               /* kerberos 
style */

The older Slackware versions and others have a clean auth.h
file.

That should solve our answers... good luck at the seminar.
I'll be out on vacation July 17 to July 30th.  

Rob

On Wed, 14 Jul 1999, Robb Kambic wrote:

> Rob,
> 
> Before you compile the LDM is the environment variable AUTH_KERB set? If
> it's set, unset it and compile the LDM. Also, could you try running a RH
> 5.2 compiled LDM on a RH 6.0 system?  I don't have access to RH 6.0 and I
> need to know because of an upcoming workshop in a week.
> 
> Thanks,
> Robb...
> 
> 
> 
> 
> On Mon, 12 Jul 1999, Rob Cermak wrote:
> 
> > Hi,
> > 
> > I was searching the archives and there was a bug reported for
> > compilation of ldm 5.0.8, but didn't mention the distribution.
> > 
> > I just compiled ldm 5.0.8 on Redhat 6.0 -- hit the same bug.
> > 
> > It looks like a compile define is set (and shouldn't).  If
> > someone that loaded Kerberos support can let the group know
> > if the compile succeeded?
> > 
> > If you don't have KERBEROS support edit this:  protocol/h_clnt.c
> > 
> > Locate this line:
> > 
> > #ifdef AUTH_KERB
> > 
> > Add these above it:
> > /* LINUX ISM */
> > #undef AUTH_KERB
> > 
> > Re-try make and try it out.
> > 
> > [ldm@scree bin]$ ./ldmping wx.envsci.rutgers.edu
> > Jul 12 19:06:08      State    Elapsed Port   Remote_Host
> > rpc_stat
> > Jul 12 19:06:09 SVC_UNAVAIL   0.975858    0   wx.envsci.rutgers.edu  RPC:
> > Unable to receive; errno = Connection reset by peer
> > [ldm@scree protocol]$ uname -a
> > Linux scree.rutgers.edu 2.2.5-22 #1 Wed Jun 2 09:02:27 EDT 1999 i586
> > uknown
> > 
> > Rob
> > 
> > PS> I don't have access to a LDM feed/site.   It compiles, unsure
> >     if it works.   I should be able to test it in August.
> > 
> > PSS> Does the NOAAport feed turn a Linux box into a toaster?
> >      Most likely needs the faster processor, memory and 100MBS NIC?
> > 
> 
> ===============================================================================
> Robb Kambic                              Unidata Program Center
> Software Engineer III                    Univ. Corp for Atmospheric Research
> address@hidden                   WWW: http://www.unidata.ucar.edu/
> ===============================================================================
> 
>