Re: LDAP: bugfix and deprecated OpenLDAP API
| От | Abhijit Menon-Sen |
|---|---|
| Тема | Re: LDAP: bugfix and deprecated OpenLDAP API |
| Дата | |
| Msg-id | 20130923064744.GA18928@toroid.org обсуждение исходный текст |
| Ответ на | Re: LDAP: bugfix and deprecated OpenLDAP API (Albe Laurenz <laurenz.albe@wien.gv.at>) |
| Ответы |
Re: LDAP: bugfix and deprecated OpenLDAP API
|
| Список | pgsql-hackers |
At 2013-08-19 11:47:36 +0000, laurenz.albe@wien.gv.at wrote:
>
> To repeat: this fixes a bug in LDAP connection parameter lookup
Hi.
I read through the patch, and it looks sensible.
I would have preferred the ldap_simple_bind_s() call in the HAVE_LIBLDAP
branch to not be inside an else {} (the if block above returns if there
is an error anyway), but that's a minor point.
I tested the code as follows:
1. Built the patched source --with-ldap
2. Set up ~/.pg_service.conf:
[foo] ldap://localhost:3343/cn=dbserver,cn=hosts?pgconnectinfo?base?(objectclass=*)
ldap://localhost:3443/cn=dbserver,cn=hosts?pgconnectinfo?base?(objectclass=*)
3. iptables -A INPUT -p tcp -d 127.0.0.1 --dport 3343 -j DROP
4. netcat -l 127.0.0.1 3343 ; netcat -l 127.0.0.1 3443
5. PGSERVICE=foo bin/psql
psql did connect to localhost:3443 after a few seconds of trying to
connect to :3343 and failing. (I tried without the iptables rule, so
I know that it does try to connect to both.)
This doesn't seem to handle timeouts in the sense of a server that
doesn't respond after you connect (or perhaps the timeout was long
enough that it outlasted my patience), but that's not the fault of
this patch, anyway.
I can't say anything about the patch on Windows, but since Magnus seemed
to think it was OK, I'm marking this ready for committer.
-- Abhijit
В списке pgsql-hackers по дате отправления: