[Pacemaker] Erro compiling PaceMaker for CoroSync and OpenAIS

[Lars Ellenberg]

On Mon, May 03, 2010 at 01:45:28PM -0400, Ruiyuan Jiang wrote:
> >>> I have to admit one more or the main reason why I am compiling
> >>> myself. Linux HA stated that the software runs on Solaris. I might
> >>> use Linux HA for my Solaris (SPARC) system if OpenAIS or Corosync
> >>> does not support Solaris. I don't think there is a pre-compiled
> >>> package for Solaris. If I can successfully compile on Linux, it is
> >>> a lot easier for me to find the problem when I am trying on
> >>> Solaris if there is a problem when I use the same procedure.

> >>
> >> Ah, thats a different story.
> >>
> >> I suggest opening up remote.c and looking for line 74. eg.
> >>
> >>        gnutls_transport_set_ptr(*session,
> >>                                 (gnutls_transport_ptr) GINT_TO_POINTER(csock));
> >>
> >> It looks like your version of gnutls_transport_set_ptr() doesn't like
> >> the call to GINT_TO_POINTER.
> >> Maybe the type for gnutls_transport_ptr is different.
> >> Perhaps google the warning:
> >>  >>> remote.c:74: warning: cast to pointer from integer of different size
> >>
> >> Or, if you're feeling luck, disable fatal warnings when you configure pacemaker.

> What do you need and I can provide you for the platform, Andrew, for
> the problem? Also it did not have problems when I compiled Linux-HA,
> Pacemaker on RHEL v5.4 (x86_64 bit). I also tried OpenAIS, Corosync
> and Pacemaker on RHEL v5.4 (x86_64 bit) and I got problem when
> compiling Pacemaker. I think the problem may have something to do with
> the combination of OpenAIS, Corosync and Pacemaker. 

You could deinstall gnutls-devel, which should disable that particular
piece of code.  You'll likely not need remote tls connections to the cib
anyways.

-- 
: Lars Ellenberg
: LINBIT | Your Way to High Availability
: DRBD/HA support and consulting http://www.linbit.com

DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.