Hi David,
David wrote: > > Hi Alessandro, > > I just realized that my email client was translating some of the test > program source as smileys. :-) > I have saved the code as an attachment and it > compiles fine as you sent it. > > >gcc crypttest.c > crypttest.c: In function `main': > crypttest.c:8: warning: initialization makes pointer from integer without a > cast > > a.out > r=ba4TuD1iozTxw So it doesn't fail in this case. However, truss clearly said it did in authdaemond. Please check "man -s 3c crypt" on your system: it should be different than mine, for I have no /etc/security/crypt.conf either. Try running the crypttest as courier user. It should still not fail, since any user must be able to use crypt for logging in... Another check that can be done is using ldd to make sure crypttest and /usr/local/libexec/courier-authlib/authdaemond link to the same libc.so. > I read through the config.log file for courier-authlib and found this > regarding crypt: > > > configure:25432: checking for crypt() prototype > configure:25462: gcc -c -I/var/vpop/include -g -O2 -Wall -I.. -I./.. > conftest.c >&5 > conftest.c:53: error: conflicting types for 'crypt' > /usr/include/crypt.h:29: error: previous declaration of 'crypt' was here > conftest.c:53: error: conflicting types for 'crypt' > /usr/include/crypt.h:29: error: previous declaration of 'crypt' was here > configure:25468: $? = 1 > [...] Yes, for some arcane reason there are conflicting declarations of crypt. Probably, including crypt.h instead of unistd.h I could avoid that compile time warning. However, this does not change the way the crypt function in libc is invoked. ------------------------------------------------------- This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting Tool for open source databases. Create drag-&-drop reports. Save time by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc. Download a FREE copy at http://www.intelliview.com/go/osdn_nl _______________________________________________ courier-users mailing list [email protected] Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users
