Hi all,
"pam_userdb.so" from pam-0.75-48 is linked against threaded libdb-4.0, that
can cause bigger problems. This was the reason to bundle a libdb with pam
(see changelog from pam-0.77-0), but at Red Hat Linux 9 this wasn't done.
So "Very Bad Things" can happen if a unthreaded application uses the
threaded pam_userdb.so...
I reported that as bug (#120558) a week ago, but if I'm realistic, I don't
believe that my issue is fixed before the end of lifetime of Red Hat Linux
9 is reached :-(
So I started my own test to link pam against a unthreaded libdb-4.0, which
always fails with the following error:
--- snipp ---
+ make
/bin/sh ./libtool --mode=compile cc -c -I. -I../dist/../include
-I../dist/../include_auto -D_GNU_SOURCE -D_REENTRANT -fPIC -O2 -g
-march=i386
-mcpu=i686 ../dist/../btree/bt_compare.c
cc -c -I. -I../dist/../include -I../dist/../include_auto -D_GNU_SOURCE
-D_REENTRANT -fPIC -O2 -g -march=i386 -mcpu=i686 ../dist/../btree/bt_com
pare.c -fPIC -DPIC -o bt_compare.o
In file included from db_int.h:435,
from ../btree/bt_compare.c:53:
../include/mutex.h:648: parse error before "tsl_t"
../include/mutex.h:648: warning: no semicolon at end of struct or union
../include/mutex.h:663: parse error before '}' token
In file included from db_int.h:436,
from ../btree/bt_compare.c:53:
../include/region.h:142: field `mutex' has incomplete type
../include/region.h:177: confused by earlier errors, bailing out
make: *** [bt_compare.o] Error 1
--- snapp ---
I did the following (it's more or less a simple hack, that should
theoretically work):
--- snipp ---
--- pam-0.77-40.spec (from Fedora Development)
+++ pam-0.77-40.rh9.spec
@@ -1,11 +1,11 @@
-%define WITH_SELINUX 1
+%define WITH_SELINUX 0
%define build6x 0
%define _sbindir /sbin
%define _sysconfdir /etc
%define pwdb_version 0.62
-%define db_version 4.2.52
+%define db_version 4.0.14
Summary: A security tool which provides authentication for applications.
Name: pam
--- snapp ---
Any ideas? I would be very happy, if someone could help me to solve that
problem :-)
With kind regards,
Robert Scheck
--
Shrike-list mailing list
Shrike-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/shrike-list
