ldap server unreachable cause plone/zope server down - ldap

Question

I have a plone installation (4.2.5) with plone.app.ldap add-on. There is a site with plone-ldap enabled and our ldap server was changed to another domain/IP. So on, zope server downs on plone-ldap retrieving ldap information. Nothing more works even root ZMI.
Any request to server doesn't load anything few seconds after plone restart. Therefore I can't reconfigure our new ldap server neither by our site or ZMI interface.
In such case, How can I proceed to reconfigure the new ldap server on plone-ldap component? Is there some script application similar to ZMI to do this? Is it a known bug?
Some logs:
1) Zeoserver.log
2016-06-06T15:52:04 new connection ('127.0.0.1', 40051): <ManagedServerConnection ('127.0.0.1', 40051)>
2016-06-06T15:52:04 (127.0.0.1:40049) received handshake 'Z3101'
2016-06-06T15:52:04 (unconnected) disconnected
2016-06-06T15:52:04 (unconnected) disconnected
2016-06-06T15:52:08 new connection ('127.0.0.1', 40052): <ManagedServerConnection ('127.0.0.1', 40052)>
2016-06-06T15:52:08 new connection ('127.0.0.1', 40053): <ManagedServerConnection ('127.0.0.1', 40053)>
2016-06-06T15:52:08 new connection ('127.0.0.1', 40054): **<ManagedServerConnection ('127.0.0.1', 40054)>
2016-06-06T15:52:08 (127.0.0.1:40052) received handshake 'Z3101'
2016-06-06T15:52:08 (unconnected) disconnected
2016-06-06T15:52:08 (unconnected) disconnected**
2) client1/event.log
2016-06-06T15:53:12 ERROR event.LDAPDelegate {'desc': "Can't contact LDAP server"}
Traceback (most recent call last):
File "/usr/local/Plone/buildout-cache/eggs/Products.LDAPUserFolder-2.26-py2.7.egg/Products/LDAPUserFolder/LDAPDelegate.py", line 366, in search
connection = self.connect(bind_dn=bind_dn, bind_pwd=bind_pwd)
File "/usr/local/Plone/buildout-cache/eggs/Products.LDAPUserFolder-2.26-py2.7.egg/Products/LDAPUserFolder/LDAPDelegate.py", line 265, in connect
raise e
**SERVER_DOWN: {'desc': "Can't contact LDAP server"}**

Answer 1

Backup first
Disclaimer - I never seen before an LDAP configuration that freeze also the root-level admin ZMI access to the Plone site.
What I can quickly suggest you is to delete the ldap plugin from the site's acl_users and starts from scratch.
As ZMI is not usable you must use the console access.
For doing this run a zope instance as follow:
$ bin/instance debug
(where "instance" is one of your instances)
The you can delete the ldap plugin:
del app.Plone.acl_users['ldap-plugin-id']
Where Plone is the is of your site and ldap-plugin-id the is of the LDAP plugin.
If you don't remember it, look for it in this set:
app.Plone.acl_users.objectValues()
Finally you must persist your changes:
import transaction;transaction.commit()
...then exit using CTRL+D
Now you must be able to access ZMI and you must create and reconfigure a new plugin.
Please note: when configuring an LDAP or AD plugin always set a "Connection timout" and an "Operation timeout". This is probably why your access attempt totally freeze the instance.