freeipa/ipaserver/install/dsinstance.py

1070 lines
40 KiB
Python
Raw Normal View History

0000-12-31 18:09:24 -05:50
# Authors: Karl MacMillan <kmacmillan@mentalrootkit.com>
2008-02-21 13:39:50 -06:00
# Simo Sorce <ssorce@redhat.com>
0000-12-31 18:09:24 -05:50
#
# Copyright (C) 2007 Red Hat
# see file 'COPYING' for use and warranty information
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
0000-12-31 18:09:24 -05:50
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
0000-12-31 18:09:24 -05:50
#
import shutil
import pwd
import sys
import os
import re
import time
import tempfile
import stat
import ldap
from ipapython.ipa_log_manager import *
from ipapython import ipautil, sysrestore, ipaldap
from ipaserver.install import service
from ipaserver.install import installutils
from ipaserver.install import certs
from ipaserver.install import ldapupdate
from ipaserver.install import replication
from ipaserver.install import sysupgrade
from ipaserver.install import upgradeinstance
Add a KRA to IPA This patch adds the capability of installing a Dogtag KRA to an IPA instance. With this patch, a KRA is NOT configured by default when ipa-server-install is run. Rather, the command ipa-kra-install must be executed on an instance on which a Dogtag CA has already been configured. The KRA shares the same tomcat instance and DS instance as the Dogtag CA. Moreover, the same admin user/agent (and agent cert) can be used for both subsystems. Certmonger is also confgured to monitor the new subsystem certificates. To create a clone KRA, simply execute ipa-kra-install <replica_file> on a replica on which a Dogtag CA has already been replicated. ipa-kra-install will use the security domain to detect whether the system being installed is a replica, and will error out if a needed replica file is not provided. The install scripts have been refactored somewhat to minimize duplication of code. A new base class dogtagintance.py has been introduced containing code that is common to KRA and CA installs. This will become very useful when we add more PKI subsystems. The KRA will install its database as a subtree of o=ipaca, specifically o=ipakra,o=ipaca. This means that replication agreements created to replicate CA data will also replicate KRA data. No new replication agreements are required. Added dogtag plugin for KRA. This is an initial commit providing the basic vault functionality needed for vault. This plugin will likely be modified as we create the code to call some of these functions. Part of the work for: https://fedorahosted.org/freeipa/ticket/3872 The uninstallation option in ipa-kra-install is temporarily disabled. Reviewed-By: Rob Crittenden <rcritten@redhat.com> Reviewed-By: Petr Viktorin <pviktori@redhat.com>
2014-03-18 10:23:30 -05:00
from ipalib import api
from ipalib import certstore
from ipalib import errors
from ipalib import constants
from ipaplatform.tasks import tasks
from ipalib.constants import CACERT
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
from ipapython.dn import DN
from ipaplatform import services
from ipaplatform.paths import paths
DS_USER = 'dirsrv'
DS_GROUP = 'dirsrv'
IPA_SCHEMA_FILES = ("60kerberos.ldif",
"60samba.ldif",
"60ipaconfig.ldif",
"60basev2.ldif",
"60basev3.ldif",
"60ipapk11.ldif",
"60ipadns.ldif",
"60certificate-profiles.ldif",
"61kerberos-ipav3.ldif",
"65ipacertstore.ldif",
"65ipasudo.ldif",
"70ipaotp.ldif",
"70topology.ldif",
"71idviews.ldif",
"72domainlevels.ldif",
"15rfc2307bis.ldif",
"15rfc4876.ldif")
ALL_SCHEMA_FILES = IPA_SCHEMA_FILES + ("05rfc2247.ldif", )
DS_INSTANCE_PREFIX = 'slapd-'
0000-12-31 18:09:24 -05:50
def find_server_root():
if ipautil.dir_exists(paths.USR_LIB_DIRSRV_64):
return paths.USR_LIB_DIRSRV_64
else:
return paths.USR_LIB_DIRSRV
0000-12-31 18:09:24 -05:50
def config_dirname(serverid):
return (paths.ETC_DIRSRV_SLAPD_INSTANCE_TEMPLATE % serverid) + "/"
def schema_dirname(serverid):
return config_dirname(serverid) + "/schema/"
def remove_ds_instance(serverid, force=False):
"""A wrapper around the 'remove-ds.pl' script used by
389ds to remove a single directory server instance. In case of error
additional call with the '-f' flag is performed (forced removal). If this
also fails, then an exception is raised.
"""
instance_name = ''.join([DS_INSTANCE_PREFIX, serverid])
args = [paths.REMOVE_DS_PL, '-i', instance_name]
if force:
args.append('-f')
root_logger.debug("Forcing instance removal")
try:
ipautil.run(args)
except ipautil.CalledProcessError:
if force:
root_logger.error("Instance removal failed.")
raise
root_logger.debug("'%s' failed. "
"Attempting to force removal" % paths.REMOVE_DS_PL)
remove_ds_instance(serverid, force=True)
improve handling of ds instances during uninstall Ticket #2502 * remove the "running" flag from backup_state in cainstance.py and dsinstance.py because it does not provide the correct information. In cainstance the running flag was never referenced because restarting dirsrv instances occurs later in dsinstance. In dsinstance when the running flag is set it incorrectly identifed the PKI ds instance configured earlier by cainstance. The intent was to determine if there were any ds instances other than those owned by IPA which will need to be restarted upon uninstall. Clearly the PKI ds instance does not qualify. We were generating a traceback when at the conclusion of dsinstance.uninstall we tried to start the remaining ds instances as indicated by the running flag, but there were none to restart (because the running flag had been set as a consequence of the PKI ds instance). * We only want to restart ds instances if there are other ds instances besides those owned by IPA. We shouldn't be stopping all ds instances either, but that's going to be covered by another ticket. The fix for restarting other ds instances at the end of uninstall is to check and see if there are other ds instances remaining after we've removed ours, if so we restart them. Also it's irrelevant if those ds instances were not present when we installed, it only matters if they exist after we restore things during uninstall. If they are present we have to start them back up because we shut them down during uninstall. * Add new function get_ds_instances() which returns a list of existing ds instances. * fixed error messages that incorrectly stated it "failed to restart" a ds instance when it should be "failed to create".
2012-04-04 14:19:29 -05:00
def get_ds_instances():
'''
Return a sorted list of all 389ds instances.
If the instance name ends with '.removed' it is ignored. This
matches 389ds behavior.
'''
dirsrv_instance_dir = paths.ETC_DIRSRV
improve handling of ds instances during uninstall Ticket #2502 * remove the "running" flag from backup_state in cainstance.py and dsinstance.py because it does not provide the correct information. In cainstance the running flag was never referenced because restarting dirsrv instances occurs later in dsinstance. In dsinstance when the running flag is set it incorrectly identifed the PKI ds instance configured earlier by cainstance. The intent was to determine if there were any ds instances other than those owned by IPA which will need to be restarted upon uninstall. Clearly the PKI ds instance does not qualify. We were generating a traceback when at the conclusion of dsinstance.uninstall we tried to start the remaining ds instances as indicated by the running flag, but there were none to restart (because the running flag had been set as a consequence of the PKI ds instance). * We only want to restart ds instances if there are other ds instances besides those owned by IPA. We shouldn't be stopping all ds instances either, but that's going to be covered by another ticket. The fix for restarting other ds instances at the end of uninstall is to check and see if there are other ds instances remaining after we've removed ours, if so we restart them. Also it's irrelevant if those ds instances were not present when we installed, it only matters if they exist after we restore things during uninstall. If they are present we have to start them back up because we shut them down during uninstall. * Add new function get_ds_instances() which returns a list of existing ds instances. * fixed error messages that incorrectly stated it "failed to restart" a ds instance when it should be "failed to create".
2012-04-04 14:19:29 -05:00
instances = []
for basename in os.listdir(dirsrv_instance_dir):
pathname = os.path.join(dirsrv_instance_dir, basename)
# Must be a directory
if os.path.isdir(pathname):
# Must start with prefix and not end with .removed
if (basename.startswith(DS_INSTANCE_PREFIX) and
not basename.endswith('.removed')):
improve handling of ds instances during uninstall Ticket #2502 * remove the "running" flag from backup_state in cainstance.py and dsinstance.py because it does not provide the correct information. In cainstance the running flag was never referenced because restarting dirsrv instances occurs later in dsinstance. In dsinstance when the running flag is set it incorrectly identifed the PKI ds instance configured earlier by cainstance. The intent was to determine if there were any ds instances other than those owned by IPA which will need to be restarted upon uninstall. Clearly the PKI ds instance does not qualify. We were generating a traceback when at the conclusion of dsinstance.uninstall we tried to start the remaining ds instances as indicated by the running flag, but there were none to restart (because the running flag had been set as a consequence of the PKI ds instance). * We only want to restart ds instances if there are other ds instances besides those owned by IPA. We shouldn't be stopping all ds instances either, but that's going to be covered by another ticket. The fix for restarting other ds instances at the end of uninstall is to check and see if there are other ds instances remaining after we've removed ours, if so we restart them. Also it's irrelevant if those ds instances were not present when we installed, it only matters if they exist after we restore things during uninstall. If they are present we have to start them back up because we shut them down during uninstall. * Add new function get_ds_instances() which returns a list of existing ds instances. * fixed error messages that incorrectly stated it "failed to restart" a ds instance when it should be "failed to create".
2012-04-04 14:19:29 -05:00
# Strip off prefix
instance = basename[len(DS_INSTANCE_PREFIX):]
improve handling of ds instances during uninstall Ticket #2502 * remove the "running" flag from backup_state in cainstance.py and dsinstance.py because it does not provide the correct information. In cainstance the running flag was never referenced because restarting dirsrv instances occurs later in dsinstance. In dsinstance when the running flag is set it incorrectly identifed the PKI ds instance configured earlier by cainstance. The intent was to determine if there were any ds instances other than those owned by IPA which will need to be restarted upon uninstall. Clearly the PKI ds instance does not qualify. We were generating a traceback when at the conclusion of dsinstance.uninstall we tried to start the remaining ds instances as indicated by the running flag, but there were none to restart (because the running flag had been set as a consequence of the PKI ds instance). * We only want to restart ds instances if there are other ds instances besides those owned by IPA. We shouldn't be stopping all ds instances either, but that's going to be covered by another ticket. The fix for restarting other ds instances at the end of uninstall is to check and see if there are other ds instances remaining after we've removed ours, if so we restart them. Also it's irrelevant if those ds instances were not present when we installed, it only matters if they exist after we restore things during uninstall. If they are present we have to start them back up because we shut them down during uninstall. * Add new function get_ds_instances() which returns a list of existing ds instances. * fixed error messages that incorrectly stated it "failed to restart" a ds instance when it should be "failed to create".
2012-04-04 14:19:29 -05:00
# Must be non-empty
if instance:
instances.append(instance)
instances.sort()
return instances
def check_ports():
"""
Check of Directory server ports are open.
Returns a tuple with two booleans, one for unsecure port 389 and one for
secure port 636. True means that the port is free, False means that the
port is taken.
"""
ds_unsecure = not ipautil.host_port_open(None, 389)
ds_secure = not ipautil.host_port_open(None, 636)
return (ds_unsecure, ds_secure)
def is_ds_running(server_id=''):
return services.knownservices.dirsrv.is_running(instance_name=server_id)
def create_ds_user():
"""Create DS user/group if it doesn't exist yet."""
tasks.create_system_user(
name=DS_USER,
group=DS_USER,
homedir=paths.VAR_LIB_DIRSRV,
shell=paths.NOLOGIN,
)
0000-12-31 18:09:24 -05:50
INF_TEMPLATE = """
[General]
FullMachineName= $FQDN
SuiteSpotUserID= $USER
SuiteSpotGroup= $GROUP
0000-12-31 18:09:24 -05:50
ServerRoot= $SERVER_ROOT
0000-12-31 18:09:24 -05:50
[slapd]
ServerPort= 389
ServerIdentifier= $SERVERID
Suffix= $SUFFIX
RootDN= cn=Directory Manager
RootDNPwd= $PASSWORD
InstallLdifFile= /var/lib/dirsrv/boot.ldif
inst_dir= /var/lib/dirsrv/scripts-$SERVERID
"""
BASE_TEMPLATE = """
dn: $SUFFIX
objectClass: top
objectClass: domain
objectClass: pilotObject
dc: $BASEDC
info: IPA V2.0
0000-12-31 18:09:24 -05:50
"""
class DsInstance(service.Service):
def __init__(self, realm_name=None, domain_name=None, dm_password=None,
fstore=None, domainlevel=None):
service.Service.__init__(self, "dirsrv",
service_desc="directory server",
dm_password=dm_password,
ldapi=False,
autobind=ipaldap.AUTOBIND_DISABLED
)
self.nickname = 'Server-Cert'
self.dm_password = dm_password
self.realm = realm_name
self.sub_dict = None
self.domain = domain_name
0000-12-31 18:09:24 -05:50
self.serverid = None
self.master_fqdn = None
self.pkcs12_info = None
self.cacert_name = None
self.ca_is_configured = True
self.dercert = None
self.idstart = None
self.idmax = None
self.subject_base = None
self.open_ports = []
self.run_init_memberof = True
self.domainlevel = domainlevel
if realm_name:
self.suffix = ipautil.realm_to_suffix(self.realm)
self.__setup_sub_dict()
else:
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
self.suffix = DN()
0000-12-31 18:09:24 -05:50
if fstore:
self.fstore = fstore
else:
self.fstore = sysrestore.FileStore(paths.SYSRESTORE)
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
subject_base = ipautil.dn_attribute_property('_subject_base')
def __common_setup(self, enable_ssl=False):
2008-02-21 13:39:50 -06:00
self.step("creating directory server user", create_ds_user)
self.step("creating directory server instance", self.__create_instance)
self.step("adding default schema", self.__add_default_schemas)
self.step("enabling memberof plugin", self.__add_memberof_module)
self.step("enabling winsync plugin", self.__add_winsync_module)
self.step("configuring replication version plugin", self.__config_version_module)
self.step("enabling IPA enrollment plugin", self.__add_enrollment_module)
self.step("enabling ldapi", self.__enable_ldapi)
self.step("configuring uniqueness plugin", self.__set_unique_attrs)
2010-10-15 16:52:37 -05:00
self.step("configuring uuid plugin", self.__config_uuid_module)
self.step("configuring modrdn plugin", self.__config_modrdn_module)
self.step("configuring DNS plugin", self.__config_dns_module)
self.step("enabling entryUSN plugin", self.__enable_entryusn)
self.step("configuring lockout plugin", self.__config_lockout_module)
self.step("configuring topology plugin", self.__config_topology_module)
2008-01-25 12:29:49 -06:00
self.step("creating indices", self.__create_indices)
self.step("enabling referential integrity plugin", self.__add_referint_module)
if enable_ssl:
self.step("configuring ssl for ds instance", self.__enable_ssl)
self.step("configuring certmap.conf", self.__certmap_conf)
self.step("configure autobind for root", self.__root_autobind)
self.step("configure new location for managed entries", self.__repoint_managed_entries)
self.step("configure dirsrv ccache", self.configure_dirsrv_ccache)
self.step("enable SASL mapping fallback", self.__enable_sasl_mapping_fallback)
self.step("restarting directory server", self.__restart_instance)
def __common_post_setup(self):
self.step("initializing group membership", self.init_memberof)
self.step("adding master entry", self.__add_master_entry)
self.step("initializing domain level", self.__set_domain_level)
self.step("configuring Posix uid/gid generation",
self.__config_uidgid_gen)
self.step("adding replication acis", self.__add_replication_acis)
self.step("enabling compatibility plugin",
self.__enable_compat_plugin)
self.step("activating sidgen plugin", self._add_sidgen_plugin)
self.step("activating extdom plugin", self._add_extdom_plugin)
self.step("tuning directory server", self.__tuning)
self.step("configuring directory to start on boot", self.__enable)
0000-12-31 18:09:24 -05:50
def init_info(self, realm_name, fqdn, domain_name, dm_password,
subject_base, idstart, idmax, pkcs12_info, ca_file=None):
self.realm = realm_name.upper()
self.serverid = installutils.realm_to_serverid(self.realm)
self.suffix = ipautil.realm_to_suffix(self.realm)
self.fqdn = fqdn
self.dm_password = dm_password
self.domain = domain_name
self.principal = "ldap/%s@%s" % (self.fqdn, self.realm)
self.subject_base = subject_base
self.idstart = idstart
self.idmax = idmax
self.pkcs12_info = pkcs12_info
if pkcs12_info:
self.ca_is_configured = False
self.ca_file = ca_file
self.__setup_sub_dict()
def create_instance(self, realm_name, fqdn, domain_name,
dm_password, pkcs12_info=None,
idstart=1100, idmax=999999, subject_base=None,
hbac_allow=True, ca_file=None):
self.init_info(
realm_name, fqdn, domain_name, dm_password,
subject_base, idstart, idmax, pkcs12_info, ca_file=ca_file)
self.__common_setup()
self.step("adding default layout", self.__add_default_layout)
self.step("adding delegation layout", self.__add_delegation_layout)
self.step("creating container for managed entries", self.__managed_entries)
self.step("configuring user private groups", self.__user_private_groups)
self.step("configuring netgroups from hostgroups", self.__host_nis_groups)
self.step("creating default Sudo bind user", self.__add_sudo_binduser)
self.step("creating default Auto Member layout", self.__add_automember_config)
self.step("adding range check plugin", self.__add_range_check_plugin)
if hbac_allow:
self.step("creating default HBAC rule allow_all", self.add_hbac)
self.step("creating default CA ACL rule", self.add_caacl)
self.step("adding entries for topology management", self.__add_topology_entries)
self.__common_post_setup()
self.start_creation(runtime=60)
def enable_ssl(self):
self.steps = []
self.step("configuring ssl for ds instance", self.__enable_ssl)
self.step("restarting directory server", self.__restart_instance)
self.step("adding CA certificate entry", self.__upload_ca_cert)
self.start_creation(runtime=10)
def create_replica(self, realm_name, master_fqdn, fqdn,
domain_name, dm_password, subject_base,
pkcs12_info=None, ca_file=None, ca_is_configured=None):
# idstart and idmax are configured so that the range is seen as
# depleted by the DNA plugin and the replica will go and get a
# new range from the master.
# This way all servers use the initially defined range by default.
idstart = 1101
idmax = 1100
self.init_info(
realm_name=realm_name,
fqdn=fqdn,
domain_name=domain_name,
dm_password=dm_password,
subject_base=subject_base,
idstart=idstart,
idmax=idmax,
pkcs12_info=pkcs12_info,
ca_file=ca_file
)
self.master_fqdn = master_fqdn
if ca_is_configured is not None:
self.ca_is_configured = ca_is_configured
self.__common_setup(True)
self.step("setting up initial replication", self.__setup_replica)
self.step("updating schema", self.__update_schema)
# See LDIFs for automember configuration during replica install
self.step("setting Auto Member configuration", self.__add_replica_automember_config)
self.step("enabling S4U2Proxy delegation", self.__setup_s4u2proxy)
self.step("importing CA certificates from LDAP", self.__import_ca_certs)
self.__common_post_setup()
self.start_creation(runtime=60)
def __setup_replica(self):
replication.enable_replication_version_checking(self.fqdn,
self.realm,
self.dm_password)
repl = replication.ReplicationManager(self.realm,
self.fqdn,
self.dm_password)
repl.setup_replication(self.master_fqdn,
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
r_binddn=DN(('cn', 'Directory Manager')),
r_bindpw=self.dm_password)
self.run_init_memberof = repl.needs_memberof_fixup()
def __update_schema(self):
# FIXME: https://fedorahosted.org/389/ticket/47490
self._ldap_mod("schema-update.ldif")
def __enable(self):
self.backup_state("enabled", self.is_enabled())
# At the end of the installation ipa-server-install will enable the
# 'ipa' service wich takes care of starting/stopping dirsrv
self.disable()
0000-12-31 18:09:24 -05:50
def __setup_sub_dict(self):
0000-12-31 18:09:24 -05:50
server_root = find_server_root()
try:
idrange_size = self.idmax - self.idstart + 1
except TypeError:
idrange_size = None
self.sub_dict = dict(FQDN=self.fqdn, SERVERID=self.serverid,
PASSWORD=self.dm_password,
RANDOM_PASSWORD=self.generate_random(),
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
SUFFIX=self.suffix,
REALM=self.realm, USER=DS_USER,
SERVER_ROOT=server_root, DOMAIN=self.domain,
TIME=int(time.time()), IDSTART=self.idstart,
IDMAX=self.idmax, HOST=self.fqdn,
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
ESCAPED_SUFFIX=str(self.suffix),
GROUP=DS_GROUP,
IDRANGE_SIZE=idrange_size,
DOMAIN_LEVEL=self.domainlevel,
MAX_DOMAIN_LEVEL=constants.MAX_DOMAIN_LEVEL,
MIN_DOMAIN_LEVEL=constants.MIN_DOMAIN_LEVEL,
)
0000-12-31 18:09:24 -05:50
def __create_instance(self):
pent = pwd.getpwnam(DS_USER)
self.backup_state("serverid", self.serverid)
self.fstore.backup_file(paths.SYSCONFIG_DIRSRV)
self.sub_dict['BASEDC'] = self.realm.split('.')[0].lower()
base_txt = ipautil.template_str(BASE_TEMPLATE, self.sub_dict)
root_logger.debug(base_txt)
target_fname = paths.DIRSRV_BOOT_LDIF
base_fd = open(target_fname, "w")
base_fd.write(base_txt)
base_fd.close()
# Must be readable for dirsrv
os.chmod(target_fname, 0o440)
os.chown(target_fname, pent.pw_uid, pent.pw_gid)
inf_txt = ipautil.template_str(INF_TEMPLATE, self.sub_dict)
root_logger.debug("writing inf template")
inf_fd = ipautil.write_tmp_file(inf_txt)
inf_txt = re.sub(r"RootDNPwd=.*\n", "", inf_txt)
root_logger.debug(inf_txt)
if ipautil.file_exists(paths.SETUP_DS_PL):
args = [paths.SETUP_DS_PL, "--silent", "--logfile", "-", "-f", inf_fd.name]
root_logger.debug("calling setup-ds.pl")
else:
args = [paths.DS_NEWINST_PL, inf_fd.name]
root_logger.debug("calling ds_newinst.pl")
try:
ipautil.run(args)
root_logger.debug("completed creating ds instance")
except ipautil.CalledProcessError as e:
improve handling of ds instances during uninstall Ticket #2502 * remove the "running" flag from backup_state in cainstance.py and dsinstance.py because it does not provide the correct information. In cainstance the running flag was never referenced because restarting dirsrv instances occurs later in dsinstance. In dsinstance when the running flag is set it incorrectly identifed the PKI ds instance configured earlier by cainstance. The intent was to determine if there were any ds instances other than those owned by IPA which will need to be restarted upon uninstall. Clearly the PKI ds instance does not qualify. We were generating a traceback when at the conclusion of dsinstance.uninstall we tried to start the remaining ds instances as indicated by the running flag, but there were none to restart (because the running flag had been set as a consequence of the PKI ds instance). * We only want to restart ds instances if there are other ds instances besides those owned by IPA. We shouldn't be stopping all ds instances either, but that's going to be covered by another ticket. The fix for restarting other ds instances at the end of uninstall is to check and see if there are other ds instances remaining after we've removed ours, if so we restart them. Also it's irrelevant if those ds instances were not present when we installed, it only matters if they exist after we restore things during uninstall. If they are present we have to start them back up because we shut them down during uninstall. * Add new function get_ds_instances() which returns a list of existing ds instances. * fixed error messages that incorrectly stated it "failed to restart" a ds instance when it should be "failed to create".
2012-04-04 14:19:29 -05:00
root_logger.critical("failed to create ds instance %s" % e)
# check for open port 389 from now on
self.open_ports.append(389)
root_logger.debug("restarting ds instance")
try:
self.__restart_instance()
root_logger.debug("done restarting ds instance")
except ipautil.CalledProcessError as e:
print "failed to restart ds instance", e
root_logger.debug("failed to restart ds instance %s" % e)
inf_fd.close()
os.remove(paths.DIRSRV_BOOT_LDIF)
0000-12-31 18:09:24 -05:50
def __add_default_schemas(self):
pent = pwd.getpwnam(DS_USER)
for schema_fname in IPA_SCHEMA_FILES:
target_fname = schema_dirname(self.serverid) + schema_fname
shutil.copyfile(ipautil.SHARE_DIR + schema_fname, target_fname)
os.chmod(target_fname, 0o440) # read access for dirsrv user/group
os.chown(target_fname, pent.pw_uid, pent.pw_gid)
try:
shutil.move(schema_dirname(self.serverid) + "05rfc2247.ldif",
schema_dirname(self.serverid) + "05rfc2247.ldif.old")
target_fname = schema_dirname(self.serverid) + "05rfc2247.ldif"
shutil.copyfile(ipautil.SHARE_DIR + "05rfc2247.ldif", target_fname)
os.chmod(target_fname, 0o440)
os.chown(target_fname, pent.pw_uid, pent.pw_gid)
except IOError:
# Does not apply with newer DS releases
pass
0000-12-31 18:09:24 -05:50
def restart(self, instance=''):
try:
super(DsInstance, self).restart(instance)
if not is_ds_running(instance):
root_logger.critical("Failed to restart the directory server. See the installation log for details.")
sys.exit(1)
except SystemExit as e:
raise e
except Exception as e:
# TODO: roll back here?
root_logger.critical("Failed to restart the directory server (%s). See the installation log for details." % e)
def __restart_instance(self):
self.restart(self.serverid)
def __enable_entryusn(self):
self._ldap_mod("entryusn.ldif")
def __add_memberof_module(self):
self._ldap_mod("memberof-conf.ldif")
2007-10-16 09:17:39 -05:00
def init_memberof(self):
if not self.run_init_memberof:
return
self._ldap_mod("memberof-task.ldif", self.sub_dict)
# Note, keep dn in sync with dn in install/share/memberof-task.ldif
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
dn = DN(('cn', 'IPA install %s' % self.sub_dict["TIME"]), ('cn', 'memberof task'),
('cn', 'tasks'), ('cn', 'config'))
root_logger.debug("Waiting for memberof task to complete.")
conn = ipaldap.IPAdmin(self.fqdn)
if self.dm_password:
conn.do_simple_bind(
DN(('cn', 'directory manager')), self.dm_password)
else:
conn.do_sasl_gssapi_bind()
replication.wait_for_task(conn, dn)
conn.unbind()
def apply_updates(self):
data_upgrade = upgradeinstance.IPAUpgrade(self.realm)
try:
data_upgrade.create_instance()
except Exception as e:
# very fatal errors only will raise exception
raise RuntimeError("Update failed: %s" % e)
installutils.store_version()
0000-12-31 18:09:24 -05:50
def __add_referint_module(self):
self._ldap_mod("referint-conf.ldif")
0000-12-31 18:09:24 -05:50
def __set_unique_attrs(self):
self._ldap_mod("unique-attributes.ldif", self.sub_dict)
def __config_uidgid_gen(self):
self._ldap_mod("dna.ldif", self.sub_dict)
def __add_master_entry(self):
self._ldap_mod("master-entry.ldif", self.sub_dict)
def __add_topology_entries(self):
self._ldap_mod("topology-entries.ldif", self.sub_dict)
def __add_winsync_module(self):
self._ldap_mod("ipa-winsync-conf.ldif")
def __enable_compat_plugin(self):
ld = ldapupdate.LDAPUpdate(dm_password=self.dm_password, sub_dict=self.sub_dict)
rv = ld.update([paths.SCHEMA_COMPAT_ULDIF])
if not rv:
raise RuntimeError("Enabling compatibility plugin failed")
def __config_version_module(self):
self._ldap_mod("version-conf.ldif")
2010-10-15 16:52:37 -05:00
def __config_uuid_module(self):
self._ldap_mod("uuid-conf.ldif")
self._ldap_mod("uuid.ldif", self.sub_dict)
2010-10-15 16:52:37 -05:00
def __config_modrdn_module(self):
self._ldap_mod("modrdn-conf.ldif")
self._ldap_mod("modrdn-krbprinc.ldif", self.sub_dict)
def __config_dns_module(self):
# Configure DNS plugin unconditionally as we would otherwise have
# troubles if other replica just configured DNS with ipa-dns-install
self._ldap_mod("ipa-dns-conf.ldif")
def __config_lockout_module(self):
self._ldap_mod("lockout-conf.ldif")
def __config_topology_module(self):
self._ldap_mod("ipa-topology-conf.ldif", self.sub_dict)
def __repoint_managed_entries(self):
self._ldap_mod("repoint-managed-entries.ldif", self.sub_dict)
def configure_dirsrv_ccache(self):
pent = pwd.getpwnam("dirsrv")
ccache = paths.TMP_KRB5CC % pent.pw_uid
filepath = paths.SYSCONFIG_DIRSRV
if not os.path.exists(filepath):
# file doesn't exist; create it with correct ownership & mode
open(filepath, 'a').close()
os.chmod(filepath,
stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP | stat.S_IROTH)
os.chown(filepath, 0, 0)
replacevars = {'KRB5CCNAME': ccache}
old_values = ipautil.backup_config_and_replace_variables(
self.fstore, filepath, replacevars=replacevars)
tasks.restore_context(filepath)
def __managed_entries(self):
self._ldap_mod("managed-entries.ldif", self.sub_dict)
def __user_private_groups(self):
self._ldap_mod("user_private_groups.ldif", self.sub_dict)
def __host_nis_groups(self):
self._ldap_mod("host_nis_groups.ldif", self.sub_dict)
def __add_enrollment_module(self):
self._ldap_mod("enrollment-conf.ldif", self.sub_dict)
def generate_random(self):
return ipautil.ipa_generate_password()
def __enable_ssl(self):
dirname = config_dirname(self.serverid)
dsdb = certs.CertDB(self.realm, nssdir=dirname, subject_base=self.subject_base)
if self.pkcs12_info:
if self.ca_is_configured:
trust_flags = 'CT,C,C'
else:
trust_flags = None
dsdb.create_from_pkcs12(self.pkcs12_info[0], self.pkcs12_info[1],
ca_file=self.ca_file,
trust_flags=trust_flags)
server_certs = dsdb.find_server_certs()
if len(server_certs) == 0:
raise RuntimeError("Could not find a suitable server cert in import in %s" % self.pkcs12_info[0])
# We only handle one server cert
self.nickname = server_certs[0][0]
self.dercert = dsdb.get_cert_from_db(self.nickname, pem=False)
else:
cadb = certs.CertDB(self.realm, host_name=self.fqdn, subject_base=self.subject_base)
# FIXME, need to set this nickname in the RA plugin
cadb.export_ca_cert('ipaCert', False)
dsdb.create_from_cacert(cadb.cacert_fname, passwd=None)
self.dercert = dsdb.create_server_cert(
self.nickname, self.fqdn, cadb)
dsdb.create_pin_file()
self.cacert_name = dsdb.cacert_name
if self.ca_is_configured:
dsdb.track_server_cert(
self.nickname, self.principal, dsdb.passwd_fname,
'restart_dirsrv %s' % self.serverid)
conn = ipaldap.IPAdmin(self.fqdn)
conn.do_simple_bind(DN(('cn', 'directory manager')), self.dm_password)
mod = [(ldap.MOD_REPLACE, "nsSSLClientAuth", "allowed"),
(ldap.MOD_REPLACE, "nsSSL3Ciphers", "+all"),
(ldap.MOD_REPLACE, "allowWeakCipher", "off")]
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
conn.modify_s(DN(('cn', 'encryption'), ('cn', 'config')), mod)
mod = [(ldap.MOD_ADD, "nsslapd-security", "on")]
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
conn.modify_s(DN(('cn', 'config')), mod)
entry = conn.make_entry(
DN(('cn', 'RSA'), ('cn', 'encryption'), ('cn', 'config')),
objectclass=["top", "nsEncryptionModule"],
cn=["RSA"],
nsSSLPersonalitySSL=[self.nickname],
nsSSLToken=["internal (software)"],
nsSSLActivation=["on"],
)
conn.add_entry(entry)
2008-02-21 13:39:50 -06:00
conn.unbind()
0000-12-31 18:09:24 -05:50
# check for open secure port 636 from now on
self.open_ports.append(636)
def __upload_ca_cert(self):
"""
Upload the CA certificate from the NSS database to the LDAP directory.
"""
dirname = config_dirname(self.serverid)
dsdb = certs.CertDB(self.realm, nssdir=dirname,
subject_base=self.subject_base)
trust_flags = dict(reversed(dsdb.list_certs()))
conn = ipaldap.IPAdmin(self.fqdn)
conn.do_simple_bind(DN(('cn', 'directory manager')), self.dm_password)
nicknames = dsdb.find_root_cert(self.cacert_name)[:-1]
for nickname in nicknames:
cert = dsdb.get_cert_from_db(nickname, pem=False)
certstore.put_ca_cert_nss(conn, self.suffix, cert, nickname,
trust_flags[nickname])
nickname = self.cacert_name
cert = dsdb.get_cert_from_db(nickname, pem=False)
certstore.put_ca_cert_nss(conn, self.suffix, cert, nickname,
trust_flags[nickname],
config_ipa=self.ca_is_configured,
config_compat=self.master_fqdn is None)
conn.unbind()
def __import_ca_certs(self):
dirname = config_dirname(self.serverid)
dsdb = certs.CertDB(self.realm, nssdir=dirname,
subject_base=self.subject_base)
conn = ipaldap.IPAdmin(self.fqdn)
conn.do_simple_bind(DN(('cn', 'directory manager')), self.dm_password)
self.import_ca_certs(dsdb, self.ca_is_configured, conn)
conn.unbind()
0000-12-31 18:09:24 -05:50
def __add_default_layout(self):
self._ldap_mod("bootstrap-template.ldif", self.sub_dict)
2008-02-21 13:39:50 -06:00
def __add_delegation_layout(self):
self._ldap_mod("delegation.ldif", self.sub_dict)
def __add_replication_acis(self):
self._ldap_mod("replica-acis.ldif", self.sub_dict)
def __setup_s4u2proxy(self):
self._ldap_mod("replica-s4u2proxy.ldif", self.sub_dict)
2008-01-25 12:29:49 -06:00
def __create_indices(self):
self._ldap_mod("indices.ldif")
def __certmap_conf(self):
shutil.copyfile(ipautil.SHARE_DIR + "certmap.conf.template",
config_dirname(self.serverid) + "certmap.conf")
installutils.update_file(config_dirname(self.serverid) + "certmap.conf",
'$SUBJECT_BASE', str(self.subject_base))
sysupgrade.set_upgrade_state(
'certmap.conf',
'subject_base',
str(self.subject_base)
)
def __enable_ldapi(self):
self._ldap_mod("ldapi.ldif", self.sub_dict)
def __enable_sasl_mapping_fallback(self):
self._ldap_mod("sasl-mapping-fallback.ldif", self.sub_dict)
def add_hbac(self):
self._ldap_mod("default-hbac.ldif", self.sub_dict)
def add_caacl(self):
self._ldap_mod("default-caacl.ldif", self.sub_dict)
def change_admin_password(self, password):
root_logger.debug("Changing admin password")
dirname = config_dirname(self.serverid)
dmpwdfile = ""
admpwdfile = ""
try:
(dmpwdfd, dmpwdfile) = tempfile.mkstemp(dir=paths.VAR_LIB_IPA)
os.write(dmpwdfd, self.dm_password)
os.close(dmpwdfd)
(admpwdfd, admpwdfile) = tempfile.mkstemp(dir=paths.VAR_LIB_IPA)
os.write(admpwdfd, password)
os.close(admpwdfd)
args = [paths.LDAPPASSWD, "-h", self.fqdn,
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
"-ZZ", "-x", "-D", str(DN(('cn', 'Directory Manager'))),
"-y", dmpwdfile, "-T", admpwdfile,
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
str(DN(('uid', 'admin'), ('cn', 'users'), ('cn', 'accounts'), self.suffix))]
try:
env = { 'LDAPTLS_CACERTDIR':os.path.dirname(CACERT),
'LDAPTLS_CACERT':CACERT }
ipautil.run(args, env=env)
root_logger.debug("ldappasswd done")
except ipautil.CalledProcessError as e:
print "Unable to set admin password", e
root_logger.debug("Unable to set admin password %s" % e)
finally:
if os.path.isfile(dmpwdfile):
os.remove(dmpwdfile)
if os.path.isfile(admpwdfile):
os.remove(admpwdfile)
def uninstall(self):
if self.is_configured():
self.print_msg("Unconfiguring directory server")
enabled = self.restore_state("enabled")
# Just eat this state if it exists
running = self.restore_state("running")
try:
self.fstore.restore_file(paths.LIMITS_CONF)
self.fstore.restore_file(paths.SYSCONFIG_DIRSRV)
except ValueError as error:
root_logger.debug(error)
pass
# disabled during IPA installation
if enabled:
self.enable()
serverid = self.restore_state("serverid")
if serverid is not None:
self.stop_tracking_certificates(serverid)
root_logger.debug("Removing DS instance %s" % serverid)
try:
remove_ds_instance(serverid)
root_logger.debug("Removing DS keytab")
installutils.remove_file(paths.DS_KEYTAB)
except ipautil.CalledProcessError:
root_logger.error("Failed to remove DS instance. You may "
"need to remove instance data manually")
# At one time we removed this user on uninstall. That can potentially
# orphan files, or worse, if another useradd runs in the intermim,
# cause files to have a new owner.
user_exists = self.restore_state("user_exists")
# Make sure some upgrade-related state is removed. This could cause
# re-installation problems.
self.restore_state('nsslapd-port')
self.restore_state('nsslapd-security')
self.restore_state('nsslapd-ldapiautobind')
improve handling of ds instances during uninstall Ticket #2502 * remove the "running" flag from backup_state in cainstance.py and dsinstance.py because it does not provide the correct information. In cainstance the running flag was never referenced because restarting dirsrv instances occurs later in dsinstance. In dsinstance when the running flag is set it incorrectly identifed the PKI ds instance configured earlier by cainstance. The intent was to determine if there were any ds instances other than those owned by IPA which will need to be restarted upon uninstall. Clearly the PKI ds instance does not qualify. We were generating a traceback when at the conclusion of dsinstance.uninstall we tried to start the remaining ds instances as indicated by the running flag, but there were none to restart (because the running flag had been set as a consequence of the PKI ds instance). * We only want to restart ds instances if there are other ds instances besides those owned by IPA. We shouldn't be stopping all ds instances either, but that's going to be covered by another ticket. The fix for restarting other ds instances at the end of uninstall is to check and see if there are other ds instances remaining after we've removed ours, if so we restart them. Also it's irrelevant if those ds instances were not present when we installed, it only matters if they exist after we restore things during uninstall. If they are present we have to start them back up because we shut them down during uninstall. * Add new function get_ds_instances() which returns a list of existing ds instances. * fixed error messages that incorrectly stated it "failed to restart" a ds instance when it should be "failed to create".
2012-04-04 14:19:29 -05:00
# If any dirsrv instances remain after we've removed ours then
# (re)start them.
for ds_instance in get_ds_instances():
try:
services.knownservices.dirsrv.restart(ds_instance, wait=False)
except Exception as e:
improve handling of ds instances during uninstall Ticket #2502 * remove the "running" flag from backup_state in cainstance.py and dsinstance.py because it does not provide the correct information. In cainstance the running flag was never referenced because restarting dirsrv instances occurs later in dsinstance. In dsinstance when the running flag is set it incorrectly identifed the PKI ds instance configured earlier by cainstance. The intent was to determine if there were any ds instances other than those owned by IPA which will need to be restarted upon uninstall. Clearly the PKI ds instance does not qualify. We were generating a traceback when at the conclusion of dsinstance.uninstall we tried to start the remaining ds instances as indicated by the running flag, but there were none to restart (because the running flag had been set as a consequence of the PKI ds instance). * We only want to restart ds instances if there are other ds instances besides those owned by IPA. We shouldn't be stopping all ds instances either, but that's going to be covered by another ticket. The fix for restarting other ds instances at the end of uninstall is to check and see if there are other ds instances remaining after we've removed ours, if so we restart them. Also it's irrelevant if those ds instances were not present when we installed, it only matters if they exist after we restore things during uninstall. If they are present we have to start them back up because we shut them down during uninstall. * Add new function get_ds_instances() which returns a list of existing ds instances. * fixed error messages that incorrectly stated it "failed to restart" a ds instance when it should be "failed to create".
2012-04-04 14:19:29 -05:00
root_logger.error('Unable to restart ds instance %s: %s', ds_instance, e)
def stop_tracking_certificates(self, serverid=None):
if serverid is None:
serverid = self.get_state("serverid")
if not serverid is None:
# drop the trailing / off the config_dirname so the directory
# will match what is in certmonger
dirname = config_dirname(serverid)[:-1]
dsdb = certs.CertDB(self.realm, nssdir=dirname)
dsdb.untrack_server_cert(self.nickname)
# we could probably move this function into the service.Service
# class - it's very generic - all we need is a way to get an
# instance of a particular Service
def add_ca_cert(self, cacert_fname, cacert_name=''):
"""Add a CA certificate to the directory server cert db. We
first have to shut down the directory server in case it has
opened the cert db read-only. Then we use the CertDB class
to add the CA cert. We have to provide a nickname, and we
do not use 'IPA CA' since that's the default, so
we use 'Imported CA' if none specified. Then we restart
the server."""
# first make sure we have a valid cacert_fname
try:
if not os.access(cacert_fname, os.R_OK):
root_logger.critical("The given CA cert file named [%s] could not be read" %
cacert_fname)
return False
except OSError as e:
root_logger.critical("The given CA cert file named [%s] could not be read: %s" %
(cacert_fname, str(e)))
return False
# ok - ca cert file can be read
# shutdown the server
self.stop()
dirname = config_dirname(installutils.realm_to_serverid(self.realm))
certdb = certs.CertDB(self.realm, nssdir=dirname, subject_base=self.subject_base)
if not cacert_name or len(cacert_name) == 0:
cacert_name = "Imported CA"
# we can't pass in the nickname, so we set the instance variable
certdb.cacert_name = cacert_name
status = True
try:
certdb.load_cacert(cacert_fname, 'C,,')
except ipautil.CalledProcessError as e:
root_logger.critical("Error importing CA cert file named [%s]: %s" %
(cacert_fname, str(e)))
status = False
# restart the directory server
self.start()
return status
def tune_nofile(self, num=8192):
"""
Increase the number of files descriptors available to directory server
from the default 1024 to 8192. This will allow to support a greater
number of clients out of the box.
"""
# Do the platform-specific changes
proceed = services.knownservices.dirsrv.tune_nofile_platform(
num=num, fstore=self.fstore)
if proceed:
# finally change also DS configuration
# NOTE: dirsrv will not allow you to set max file descriptors unless
# the user limits allow it, so we have to restart dirsrv before
# attempting to change them in cn=config
self.__restart_instance()
nf_sub_dict = dict(NOFILES=str(num))
self._ldap_mod("ds-nfiles.ldif", nf_sub_dict)
def __tuning(self):
self.tune_nofile(8192)
def __root_autobind(self):
self._ldap_mod("root-autobind.ldif")
def __add_sudo_binduser(self):
self._ldap_mod("sudobind.ldif", self.sub_dict)
def __add_automember_config(self):
self._ldap_mod("automember.ldif", self.sub_dict)
def __add_replica_automember_config(self):
self._ldap_mod("replica-automember.ldif", self.sub_dict)
def __add_range_check_plugin(self):
self._ldap_mod("range-check-conf.ldif", self.sub_dict)
def _add_sidgen_plugin(self):
"""
Add sidgen directory server plugin configuration if it does not already exist.
"""
self._ldap_mod('ipa-sidgen-conf.ldif', self.sub_dict)
def add_sidgen_plugin(self):
"""
Add sidgen plugin configuration only if it does not already exist.
"""
dn = DN('cn=IPA SIDGEN,cn=plugins,cn=config')
try:
self.admin_conn.get_entry(dn)
except errors.NotFound:
self._add_sidgen_plugin()
else:
root_logger.debug("sidgen plugin is already configured")
def _add_extdom_plugin(self):
"""
Add directory server configuration for the extdom extended operation.
"""
self._ldap_mod('ipa-extdom-extop-conf.ldif', self.sub_dict)
def add_extdom_plugin(self):
"""
Add extdom configuration if it does not already exist.
"""
dn = DN('cn=ipa_extdom_extop,cn=plugins,cn=config')
try:
self.admin_conn.get_entry(dn)
except errors.NotFound:
self._add_extdom_plugin()
else:
root_logger.debug("extdom plugin is already configured")
def replica_populate(self):
self.ldap_connect()
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
dn = DN(('cn', 'default'), ('ou', 'profile'), self.suffix)
try:
entry = self.admin_conn.get_entry(dn)
srvlist = entry.single_value.get('defaultServerList', '')
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
srvlist = srvlist.split()
if not self.fqdn in srvlist:
srvlist.append(self.fqdn)
attr = ' '.join(srvlist)
mod = [(ldap.MOD_REPLACE, 'defaultServerList', attr)]
self.admin_conn.modify_s(dn, mod)
Use DN objects instead of strings * Convert every string specifying a DN into a DN object * Every place a dn was manipulated in some fashion it was replaced by the use of DN operators * Add new DNParam parameter type for parameters which are DN's * DN objects are used 100% of the time throughout the entire data pipeline whenever something is logically a dn. * Many classes now enforce DN usage for their attributes which are dn's. This is implmented via ipautil.dn_attribute_property(). The only permitted types for a class attribute specified to be a DN are either None or a DN object. * Require that every place a dn is used it must be a DN object. This translates into lot of:: assert isinstance(dn, DN) sprinkled through out the code. Maintaining these asserts is valuable to preserve DN type enforcement. The asserts can be disabled in production. The goal of 100% DN usage 100% of the time has been realized, these asserts are meant to preserve that. The asserts also proved valuable in detecting functions which did not obey their function signatures, such as the baseldap pre and post callbacks. * Moved ipalib.dn to ipapython.dn because DN class is shared with all components, not just the server which uses ipalib. * All API's now accept DN's natively, no need to convert to str (or unicode). * Removed ipalib.encoder and encode/decode decorators. Type conversion is now explicitly performed in each IPASimpleLDAPObject method which emulates a ldap.SimpleLDAPObject method. * Entity & Entry classes now utilize DN's * Removed __getattr__ in Entity & Entity clases. There were two problems with it. It presented synthetic Python object attributes based on the current LDAP data it contained. There is no way to validate synthetic attributes using code checkers, you can't search the code to find LDAP attribute accesses (because synthetic attriutes look like Python attributes instead of LDAP data) and error handling is circumscribed. Secondly __getattr__ was hiding Python internal methods which broke class semantics. * Replace use of methods inherited from ldap.SimpleLDAPObject via IPAdmin class with IPAdmin methods. Directly using inherited methods was causing us to bypass IPA logic. Mostly this meant replacing the use of search_s() with getEntry() or getList(). Similarly direct access of the LDAP data in classes using IPAdmin were replaced with calls to getValue() or getValues(). * Objects returned by ldap2.find_entries() are now compatible with either the python-ldap access methodology or the Entity/Entry access methodology. * All ldap operations now funnel through the common IPASimpleLDAPObject giving us a single location where we interface to python-ldap and perform conversions. * The above 4 modifications means we've greatly reduced the proliferation of multiple inconsistent ways to perform LDAP operations. We are well on the way to having a single API in IPA for doing LDAP (a long range goal). * All certificate subject bases are now DN's * DN objects were enhanced thusly: - find, rfind, index, rindex, replace and insert methods were added - AVA, RDN and DN classes were refactored in immutable and mutable variants, the mutable variants are EditableAVA, EditableRDN and EditableDN. By default we use the immutable variants preserving important semantics. To edit a DN cast it to an EditableDN and cast it back to DN when done editing. These issues are fully described in other documentation. - first_key_match was removed - DN equalty comparison permits comparison to a basestring * Fixed ldapupdate to work with DN's. This work included: - Enhance test_updates.py to do more checking after applying update. Add test for update_from_dict(). Convert code to use unittest classes. - Consolidated duplicate code. - Moved code which should have been in the class into the class. - Fix the handling of the 'deleteentry' update action. It's no longer necessary to supply fake attributes to make it work. Detect case where subsequent update applies a change to entry previously marked for deletetion. General clean-up and simplification of the 'deleteentry' logic. - Rewrote a couple of functions to be clearer and more Pythonic. - Added documentation on the data structure being used. - Simplfy the use of update_from_dict() * Removed all usage of get_schema() which was being called prior to accessing the .schema attribute of an object. If a class is using internal lazy loading as an optimization it's not right to require users of the interface to be aware of internal optimization's. schema is now a property and when the schema property is accessed it calls a private internal method to perform the lazy loading. * Added SchemaCache class to cache the schema's from individual servers. This was done because of the observation we talk to different LDAP servers, each of which may have it's own schema. Previously we globally cached the schema from the first server we connected to and returned that schema in all contexts. The cache includes controls to invalidate it thus forcing a schema refresh. * Schema caching is now senstive to the run time context. During install and upgrade the schema can change leading to errors due to out-of-date cached schema. The schema cache is refreshed in these contexts. * We are aware of the LDAP syntax of all LDAP attributes. Every attribute returned from an LDAP operation is passed through a central table look-up based on it's LDAP syntax. The table key is the LDAP syntax it's value is a Python callable that returns a Python object matching the LDAP syntax. There are a handful of LDAP attributes whose syntax is historically incorrect (e.g. DistguishedNames that are defined as DirectoryStrings). The table driven conversion mechanism is augmented with a table of hard coded exceptions. Currently only the following conversions occur via the table: - dn's are converted to DN objects - binary objects are converted to Python str objects (IPA convention). - everything else is converted to unicode using UTF-8 decoding (IPA convention). However, now that the table driven conversion mechanism is in place it would be trivial to do things such as converting attributes which have LDAP integer syntax into a Python integer, etc. * Expected values in the unit tests which are a DN no longer need to use lambda expressions to promote the returned value to a DN for equality comparison. The return value is automatically promoted to a DN. The lambda expressions have been removed making the code much simpler and easier to read. * Add class level logging to a number of classes which did not support logging, less need for use of root_logger. * Remove ipaserver/conn.py, it was unused. * Consolidated duplicate code wherever it was found. * Fixed many places that used string concatenation to form a new string rather than string formatting operators. This is necessary because string formatting converts it's arguments to a string prior to building the result string. You can't concatenate a string and a non-string. * Simplify logic in rename_managed plugin. Use DN operators to edit dn's. * The live version of ipa-ldap-updater did not generate a log file. The offline version did, now both do. https://fedorahosted.org/freeipa/ticket/1670 https://fedorahosted.org/freeipa/ticket/1671 https://fedorahosted.org/freeipa/ticket/1672 https://fedorahosted.org/freeipa/ticket/1673 https://fedorahosted.org/freeipa/ticket/1674 https://fedorahosted.org/freeipa/ticket/1392 https://fedorahosted.org/freeipa/ticket/2872
2012-05-13 06:36:35 -05:00
except errors.NotFound:
pass
except ldap.TYPE_OR_VALUE_EXISTS:
pass
self.ldap_disconnect()
Add a KRA to IPA This patch adds the capability of installing a Dogtag KRA to an IPA instance. With this patch, a KRA is NOT configured by default when ipa-server-install is run. Rather, the command ipa-kra-install must be executed on an instance on which a Dogtag CA has already been configured. The KRA shares the same tomcat instance and DS instance as the Dogtag CA. Moreover, the same admin user/agent (and agent cert) can be used for both subsystems. Certmonger is also confgured to monitor the new subsystem certificates. To create a clone KRA, simply execute ipa-kra-install <replica_file> on a replica on which a Dogtag CA has already been replicated. ipa-kra-install will use the security domain to detect whether the system being installed is a replica, and will error out if a needed replica file is not provided. The install scripts have been refactored somewhat to minimize duplication of code. A new base class dogtagintance.py has been introduced containing code that is common to KRA and CA installs. This will become very useful when we add more PKI subsystems. The KRA will install its database as a subtree of o=ipaca, specifically o=ipakra,o=ipaca. This means that replication agreements created to replicate CA data will also replicate KRA data. No new replication agreements are required. Added dogtag plugin for KRA. This is an initial commit providing the basic vault functionality needed for vault. This plugin will likely be modified as we create the code to call some of these functions. Part of the work for: https://fedorahosted.org/freeipa/ticket/3872 The uninstallation option in ipa-kra-install is temporarily disabled. Reviewed-By: Rob Crittenden <rcritten@redhat.com> Reviewed-By: Petr Viktorin <pviktori@redhat.com>
2014-03-18 10:23:30 -05:00
def find_subject_base(self):
"""
Try to find the current value of certificate subject base.
1) Look in sysupgrade first
2) If no value is found there, look in DS (start DS if necessary)
3) Last resort, look in the certmap.conf itself
4) If all fails, log loudly and return None
Note that this method can only be executed AFTER the ipa server
is configured, the api is initialized elsewhere and
that a ticket already have been acquired.
"""
root_logger.debug(
'Trying to find certificate subject base in sysupgrade')
subject_base = sysupgrade.get_upgrade_state(
'certmap.conf', 'subject_base')
if subject_base:
root_logger.debug(
'Found certificate subject base in sysupgrade: %s',
subject_base)
return subject_base
root_logger.debug(
'Unable to find certificate subject base in sysupgrade')
root_logger.debug(
'Trying to find certificate subject base in DS')
ds_is_running = is_ds_running()
if not ds_is_running:
try:
self.start()
ds_is_running = True
except ipautil.CalledProcessError as e:
root_logger.error('Cannot start DS to find certificate '
'subject base: %s', e)
if ds_is_running:
try:
api.Backend.ldap2.connect(autobind=True)
ret = api.Command['config_show']()
subject_base = str(
ret['result']['ipacertificatesubjectbase'][0])
root_logger.debug(
'Found certificate subject base in DS: %s', subject_base)
except errors.PublicError as e:
Add a KRA to IPA This patch adds the capability of installing a Dogtag KRA to an IPA instance. With this patch, a KRA is NOT configured by default when ipa-server-install is run. Rather, the command ipa-kra-install must be executed on an instance on which a Dogtag CA has already been configured. The KRA shares the same tomcat instance and DS instance as the Dogtag CA. Moreover, the same admin user/agent (and agent cert) can be used for both subsystems. Certmonger is also confgured to monitor the new subsystem certificates. To create a clone KRA, simply execute ipa-kra-install <replica_file> on a replica on which a Dogtag CA has already been replicated. ipa-kra-install will use the security domain to detect whether the system being installed is a replica, and will error out if a needed replica file is not provided. The install scripts have been refactored somewhat to minimize duplication of code. A new base class dogtagintance.py has been introduced containing code that is common to KRA and CA installs. This will become very useful when we add more PKI subsystems. The KRA will install its database as a subtree of o=ipaca, specifically o=ipakra,o=ipaca. This means that replication agreements created to replicate CA data will also replicate KRA data. No new replication agreements are required. Added dogtag plugin for KRA. This is an initial commit providing the basic vault functionality needed for vault. This plugin will likely be modified as we create the code to call some of these functions. Part of the work for: https://fedorahosted.org/freeipa/ticket/3872 The uninstallation option in ipa-kra-install is temporarily disabled. Reviewed-By: Rob Crittenden <rcritten@redhat.com> Reviewed-By: Petr Viktorin <pviktori@redhat.com>
2014-03-18 10:23:30 -05:00
root_logger.error('Cannot connect to DS to find certificate '
'subject base: %s', e)
finally:
try:
api.Backend.ldap2.disconnect()
except Exception:
pass
if not subject_base:
root_logger.debug('Unable to find certificate subject base in DS')
root_logger.debug('Trying to find certificate subject base in '
'certmap.conf')
certmap_dir = config_dirname(
installutils.realm_to_serverid(api.env.realm)
Add a KRA to IPA This patch adds the capability of installing a Dogtag KRA to an IPA instance. With this patch, a KRA is NOT configured by default when ipa-server-install is run. Rather, the command ipa-kra-install must be executed on an instance on which a Dogtag CA has already been configured. The KRA shares the same tomcat instance and DS instance as the Dogtag CA. Moreover, the same admin user/agent (and agent cert) can be used for both subsystems. Certmonger is also confgured to monitor the new subsystem certificates. To create a clone KRA, simply execute ipa-kra-install <replica_file> on a replica on which a Dogtag CA has already been replicated. ipa-kra-install will use the security domain to detect whether the system being installed is a replica, and will error out if a needed replica file is not provided. The install scripts have been refactored somewhat to minimize duplication of code. A new base class dogtagintance.py has been introduced containing code that is common to KRA and CA installs. This will become very useful when we add more PKI subsystems. The KRA will install its database as a subtree of o=ipaca, specifically o=ipakra,o=ipaca. This means that replication agreements created to replicate CA data will also replicate KRA data. No new replication agreements are required. Added dogtag plugin for KRA. This is an initial commit providing the basic vault functionality needed for vault. This plugin will likely be modified as we create the code to call some of these functions. Part of the work for: https://fedorahosted.org/freeipa/ticket/3872 The uninstallation option in ipa-kra-install is temporarily disabled. Reviewed-By: Rob Crittenden <rcritten@redhat.com> Reviewed-By: Petr Viktorin <pviktori@redhat.com>
2014-03-18 10:23:30 -05:00
)
try:
with open(os.path.join(certmap_dir, 'certmap.conf')) as f:
for line in f:
if line.startswith('certmap ipaca'):
subject_base = line.strip().split(',')[-1]
root_logger.debug(
'Found certificate subject base in certmap.conf: '
'%s', subject_base)
except IOError as e:
root_logger.error('Cannot open certmap.conf to find certificate '
'subject base: %s', e.strerror)
if subject_base:
return subject_base
root_logger.debug('Unable to find certificate subject base in '
'certmap.conf')
return None
def __set_domain_level(self):
# Create global domain level entry and set the domain level
if self.domainlevel is not None:
self._ldap_mod("domainlevel.ldif", self.sub_dict)