Univention Bugzilla – Bug 49528
Zimbra4UCS fails regularly with AUTH_EXPIRED
Last modified: 2019-08-26 14:23:32 CEST
A customer uses the Zimbra4UCS connector and runs in the following traceback every few days. The problem is solved immediately by restarting the service but the connector should catch this particular error and act accordingly (renew the authentication or so). Feel free to move to another component. I couldn't find one for Zimbra4UCS. 2019-05-22 08:28:11,884 - INFO - Request 'GetAccountRequest' with {'account': {'by': 'name', '_content': 'john@doe.com'}, 'attrs': 'zimbraId'} 2019-05-22 08:28:11,899 - ERROR - <traceback object at 0x7f0b62a1ae60> Traceback (most recent call last): File "/usr/sbin/zimbra4ucs-connector", line 72, in run opts.dry_run) File "/usr/lib/pymodules/python2.7/zimbra4ucs/tools.py", line 96, in process_files cls.modify(dn, new, old) File "/usr/lib/pymodules/python2.7/zimbra4ucs/models.py", line 158, in ldap_modify return cls(cls.get_connection(), **cls.map_ldap_attributes(old)).modify(cls.map_ldap_attributes(new)) File "/usr/lib/pymodules/python2.7/zimbra4ucs/models.py", line 302, in modify return super(ZimbraAccount, self).modify(values) File "/usr/lib/pymodules/python2.7/zimbra4ucs/models.py", line 77, in modify data['id'] = self.get_id(self.name) File "/usr/lib/pymodules/python2.7/zimbra4ucs/models.py", line 65, in get_id return self.get('zimbraId', name=name)['id'] File "/usr/lib/pymodules/python2.7/zimbra4ucs/models.py", line 61, in get return self.request('%sRequest' % self.method_get, data)['%sResponse' % self.method_get][self.result_key] File "/usr/lib/pymodules/python2.7/zimbra4ucs/models.py", line 96, in request return self.connection.request(method, data).get_response() File "/usr/lib/pymodules/python2.7/zimbra4ucs/__init__.py", line 204, in request raise cls(response.get_fault_message(), code) ZimbraResponseError: Response failed: auth credentials have expired (code: service.AUTH_EXPIRED) 2019-05-22 08:28:15,288 - INFO - started with args: ['start'], opts: {'data_dir_maildomain': '/var/lib/zimbra4ucs-maildomain-listener', 'log_file': '/var/log/univention/zimbra4ucs.log', 'log_level': '3', 'dry_run': False, 'data_dir_user': '/var/lib/zimbra4ucs-user-listener'}
Workaround: ucr set cron/restart-zimbra4ucs/time="0 3 * * *" \ cron/restart-zimbra4ucs/command="systemctl restart zimbra4ucs"