OK, first of all the interpreters_lock mutex only needs to be defined
if we have threads :

Index: src/mod_python.c
 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
3D=3D=3D=3D
--- src/mod_python.c    (revision 280013)
+++ src/mod_python.c    (working copy)
@@ -31,7 +31,9 @@
* (In a Python dictionary) */
static PyObject * interpreters =3D NULL;

+#ifdef WITH_THREAD
static apr_thread_mutex_t* interpreters_lock =3D 0;
+#endif

apr_pool_t *child_init_pool =3D NULL;

Then, it looks like the apr_thread_mutex functions are not defined on
minotaur. Having a look at apr_thread_mutex.h, I saw that on line 35 :

#if APR_HAS_THREADS || defined(DOXYGEN)

Would it be possible that apache was compiled without threads on
minotaur ? What I don't understand is why WITH_THREAD is defined, in
that case...

Regards,
Nicolas




2005/9/10, Jim Gallacher <jg.lists@sympatico.ca>:
> Nicolas Lehuen wrote: 
>=20
> I don't think so, but I ran a simple test on minotaur just to make sure.
>=20
> I copied configure.in and dist/setup.py.in from 3.1.4 to 3.2.1b and
> regenerated configure with autoconf259. Compilation still fails at the
> same point with the same error.
>=20
> If only it had been that simple. :-(
>=20
> Jim
>=20 
with:[vbcol=seagreen] 
r is[vbcol=seagreen] 
e[vbcol=seagreen] 
f[vbcol=seagreen] 
1 at[vbcol=seagreen] 
od_python[vbcol=seagreen] 
be the[vbcol=seagreen] 
Linux[vbcol=seagreen] 
at is[vbcol=seagreen] 
it[vbcol=seagreen] 
umber,[vbcol=seagreen] 
scheme,[vbcol=seagreen] 
>=20
>

[ Post a follow-up to this message ]