Full_Name: Ramon Escriba
OS: openSUSE 11.4
Submission from: (NULL) (18.104.22.168)
well, is not a bug, maybe an strange interaction between openldap & postfix
I detected those logs:
Apr 13 16:40:42 <myldap> slapd: <= bdb_equality_candidates: (mailhost)
So I thought, indexing may solve it.
"mailhost" user attribute may only have 2 values, mail01 or mail02. Yes, maybe
is not the best thing to index, but I did a try.
So I added index mailhost eq in slapd.conf. Slaptest ok. Rcldap stop;
slapindex; rcldap start.
But then, one mail system began to fail. I comment out the index, restarted
daemons & works fine again.
ldap_filter: ( &(uid=%u) (mailhost=mail02) (mActive=1)
query_filter = ( &(mail=%s) (mailhost=mail02) (mActive=1)
query_filter = ( &(maildrop=%s) (mailhost=mail02) (mActive=1)
Is just for curiosity. What do you thing it may happen? Why indexing mailhost
makes the query fail?? I think I miss something important.