On Thu, Mar 03, 2005 at 01:20:35PM +0100,
Florian Pflug <fgp@phlo.org> wrote
a message of 47 lines which said:
> Seems so.. you could try to start the postmaster via strace -f, and
> capture the log
...
> Then try to connect, and see what happens - you should see the
> postmaster open your pam_ldap.conf, and then try to connect to your
> ldap server.
Well, the easiest way was to 'tail -f' the LDAP server's log. I can
see that PostgreSQL, indeed, calls getpwuid and tries to find LDAP's
"uid" (the user name):
Mar 8 10:23:24 eve slapd[586]: conn=38854 op=1 SRCH base="ou=People,dc=eureg,dc=eu" scope=1
filter="(&(objectClass=posixAccount)(uidNumber=1011))"
Mar 8 10:23:24 eve slapd[586]: conn=38854 op=1 SRCH attr=uid userPassword uidNumber gidNumber cn homeDirectory
loginShellgecos description objectClass
Strangely enough, everything works now. The only thing I see is that
the machine rebooted two days ago. :-(
Thanks for those who searched and sorry for leading them on the wrong
track.
