Re: Syncrepl N-Way multimaster replication - Configuration issue on Rhel6

[coma <coma.inf@gmail.com>]

Hello Ulrich,

thank you very much for your prompt reply.

Sorry, it was a copy/paste error, i've added the port in the URI but it made no difference whatever I do (test other port, test with default port...), error "read_config: no serverID / URL match found." was always present.

But today i've solved the issue by upgrading my servers and reboot them, but for the moment i' don't understand what was exactly the cause, maybe a conflict on name resolution, error messages in debug mode point in this direction (getaddrinfo() failed even the host is existing and resolved by DNS).

About my second issue, it was a stupid mistake from my part, slapd was listening only on localhost due to an entry in /etc/hosts (i added my server name as alias of 127.0.0.1).

Thanks you again,

Regards,

2014-11-28 8:42 GMT+01:00 Ulrich Windl <Ulrich.Windl@rz.uni-regensburg.de>:

>>> coma <coma.inf@gmail.com> schrieb am 27.11.2014 um 17:18 in Nachricht
<CABnSCoknUmvmY_eJPac9mDqsWcja57b8W_1gT09VFZv0=ncFpA@mail.gmail.com>:
> Hello everybody,
> i'm currently tring to configure N-Way multimaster replication, and
> i'm facing two issues with olcServerId and slapd -h options.
>
> For information:
> I'm running on Rhel6.6 with openldap 2.4.39-8.
> I'm running slapd on non-standard ports (11389 for ldap and ldap with
> TLS, and 11390 for ldaps)
> I've tried on standard ports but same issues.
> I've configured the replication following these two procedures:
> https://access.redhat.com/solutions/273533
> http://www.openldap.org/doc/admin24/replication.html#N-Way%20Multi-Master
>
> - First Issue details:
>
> When i'm adding olcServerID's on both servers, using following ldif:
>
> cat <<EOF | ldapmodify -Y EXTERNAL -H ldapi:///
> dn: cn=config
> changetype: modify
> replace: olcServerID
> olcServerID: 1 ldap://server1-test1.test.com
> olcServerID: 2 ldap://server2-test1.test.com
> EOF

Why didn't you add the port to your URI? You should have known that with a
non-default port at least your URI won't match your server's configuration.
Same for ldap: vs ldaps: I guess.

>
> i'm no longer able to restart slapd. Error is: read_config: no
> serverID / URL match found. Check slapd -h arguments.
>
> To resolve it, i've tried to add the URL of my servers in
> correspondant /etc/sysconfig/ldapExample:
>
> SLAPD_LDAP=no
> SLAPD_LDAPI=yes
> SLAPD_LDAPS=no
> SLAPD_URLS="ldap://server1-test1.test.com:11389
> ldaps://server1-test1.test.com:11390"
>
> But issue "Error is: read_config: no serverID / URL match found." is
> always present event after a server reboot and a full openldap
> reinstallationn.
>
> - Second issue détails (replication disabled, serverID's removed):
>
> With /etc/sysconfig/ldap configured as:
> SLAPD_LDAP=no
> SLAPD_LDAPI=yes
> SLAPD_LDAPS=no
> SLAPD_URLS="ldap://:11389 ldaps://:11390"
>
> i'm able to connect on port 11389/11390 with clear, starttls and SSL
> using a ldap browser or ldapsearch,
>
> But with /etc/sysconfig/ldap configured as:
> SLAPD_LDAP=no
> SLAPD_LDAPI=yes
> SLAPD_LDAPS=no
> SLAPD_URLS="ldap://server1-test1.test.com:11389
> ldaps://server1-test1.test.com:11390"
>
> i'm not able to connect anymore.

How do your certificates look like?


Regards,
Ulrich

>
> Can you please help me on this?
>
> Thanks in advance,