If ipaserver_domain is not given, the domain name is generated from the
host fqdn.
This generated value was so far not returned, but the empty given value
instead.
As FreeIPA now requires MS-PAC to be set in ipaKrbAuthzData to trigger
PAC generation, there's a timing issue that causes API malfunction which
is long enough to cause the client part insallation to fail.
By restarting KDC after DS password is set, we force cached values to be
refreshed, allowing the API to work correctly.
Resolves: https://github.com/freeipa/ansible-freeipa/issues/1200
This is "Fix ipa-client-automount install/uninstall with new install
states" https://github.com/freeipa/freeipa/pull/7100 for ansible-freeipa:
Issue 8384 introduced a new installation state for the statestore
to identify when client/server installation is completely finished
rather than relying on has_files().
The problem is that ipa-client-automount may be called during
ipa-client-install and since installation is not complete at that
point the automount install was failing with "IPA client not
configured".
Add a new state, 'automount', to designate that automount installation
is in process. If check_client_configuration() fails it checks to
see if [installation] automount is True. If so it continues with the
installation.
This also addresses an issue where the filestore and statestore are
shared between the client and automount installers but the client
wasn't refreshing state after automount completed. This resulted in
an incomplete state and index file of backed-up files which caused
files to not be restored on uninstall and the state file to be
orphaned.
Fixes: https://pagure.io/freeipa/issue/9487
This is "ipa-client-install: enable SELinux for SSSD"
https://github.com/freeipa/freeipa/pull/6978 for ansible-freeipa:
For passkeys (FIDO2) support, SSSD uses libfido2 library which needs
access to USB devices. Add SELinux booleans handling to ipa-client-install
so that correct SELinux booleans can be enabled and disabled during
install and uninstall. Ignore and record a warning when SELinux policy
does not support the boolean.
Fixes: https://pagure.io/freeipa/issue/9434
When deploying an IPA client with ipaclient, if an error occured while
getting an OTP, no error message is logged, as the task that logs the
error is not excuted due to the previous taks failure.
By adding a 'rescue' section to the code block and moving the error
reporting to this new section, we ensure that the proper error messages
will be reported.
The ipaclient_automount_location variable was badly named as
ipaautomount_location. Additionally it was not documented in the role
README file.
Fixes: #1166 (.. automount-location to the ipa-client role)
Altough most of ansible-freeipa documentation and playbooks use
'ipaserver' as the group for the first server deployed for a realm, the
ipareplica role only supported the use of groups["ipaservers"] as an
alternative to set ipareplica_servers.
Also supporting groups.ipaserver, as already supported by the ipaclient
role, make ansible-freeipa playbooks more consistent and current
documentation and examples easier to follow when deploying a cluster
with a server and a replica.
The use of del os.environ assumes that the environment variable exists.
If the variable does not exist, this call will result in a traceback.
The solution is to use os.environ.pop(VARIABLE, None) instead.
This is the ansible-freeipa fix for https://pagure.io/freeipa/issue/9446
(Nightly test failure for replica installation with --setup-ca)
Changing the use of 'Exception' to 'RuntimeError' has the benefits of
making the error more specific and meaningful for what is being reported
and to remove warnings from linters (pylint).
The same change is applied to all deployment roles.
Currently, the minimum supported Ansible version is 2.13, and
ansible-freeipa roles does not work with any version less than 2.9,
altough ansible-freeipa documentation states that the minimum version to
use is 2.8.
This patch fixes documentation and roles metadata to require that the
minimum Ansible version used is 2.13.
Ubuntu does not have a FreeIPA server package since version 20.04. As
versions 16.04 (Xenial Xerus) and 18.04 (Bionic Beaver) will be
supported by Canonical until 2026 and 2028, repectively, we should keep
existing support for both versions in the ipaserver, ipareplica and
ipabackup roles until them.
This patch changes documentation to reflect that only those versions are
supported.
If server FQDN matches the domain name, the installation will succeed,
but DNS records will not work. If 'setup_dns: true' is used, there will
be no A record for the host, only a NS record, and the PTR record will
point to the domain name.
Based on: https://github.com/freeipa/freeipa/pull/6853
Related to: https://pagure.io/freeipa/issue/9003
ipaserver_random_serial_numbers was enabled by default in
roles/ipaserver/defaults/main.yml. This should not be the default and
also resulted in issues in all IPA versions that do not support RSN.
The parameter now defaults to false.
Since FreeIPA version 4.10 it is possible to deploy servers that use
Random Serial Number v3 support for certificates.
This patch exposes the 'random_serial_numbers' parameter, as
'ipaserver_random_serial_numbers', allowing a user to have random serial
numbers enabled for the domain.
The use of random serial numbers is allowed on new installations only.
New variables have been added to ipareplica and ipaserver role to enable
the removal from the domein with the undeployment.
`ipaserver_remove_from_domain`
This enables the removal of the server from the domain additionally to the
undeployment.
`ipaserver_remove_on_server`
The value defines the server/replica in the domain that will to be used to
remove the server/replica from the domain if
`ipaserver_ignore_topology_disconnect` and `ipaserver_remove_from_domain`
are enabled. Without the need to enable
`ipaserver_ignore_topology_disconnect`, the value will be automatically
detected using the replication agreements of the server/replica.
For the replica role it is possible to use the server variables, but
also the replica versions: `ipareplica_remove_from_domain` and
`ipareplica_remove_on_server`.
The already existing parameters `ipaserver_ignore_topology_disconnect` and
`ipaserver_ignore_last_of_role` have been added to the README files for
server and replica with descriptions. The same for the replica versions
of the parameters.
The ipareplica role is not calling the `ipa-server-install` anymore, it
is instead using (including) the server role for the task.
The new module `ipaserver_get_connected_server` has been added to the
server role to be able to get a connected server using the replication
agreements. This module is only used if
`ipaserver_ignore_topology_disconnect` is not needed.
The cleanup of the root IPA cache was depending on the result of the
ipaserver_enable_ipa and ipareplica_enable_ipa tasks. Instead of
"when: something.changed" a handler should be used instead. As
"/root/.ipa_cache" should be removed always (same in command line) the
removal of the file has been moded into the always section and does not
need a when anymore.
With the fix to defer creating the final krb5.conf on clients a bug has
been introduced with ipaclient_fix_ca: The krb_name parameter that
points to the temporary krb5 configuration was not added to the module
Without this the server affinity is broken for allow_repair and additionally
ipaclient_fix_ca could fail if krb5 configuration needs to be repraied
and also CA needs to be fixed.
The krb_name parameter has been added to ipaclient_fix_ca and is also
properly set in tasks/install.yml.
With the fix to defer creating the final krb5.conf on clients a bug has
been introduced with ipaclient_setup_nss: The krb_name parameter that
points to the temporary krb5 configuration was not added to the module.
With a properly configured DNS (like for example IPA DNS) the krb TXT
records have been present in the DNS configuration. These have been used
automatically as a fallback and broke server affinity for the client.
Without the TXT records creating the IPA NSS database failed with
"Cannot find KDC for realm ..".
The krb_name parameter has been added to ipaclient_setup_nss and is also
properly set in tasks/install.yml.
Some ipareplica role had a few module calls with parameters set like
'some_argument | default(omit)' that were not actually available in such
modules. If a user provided 'some_argument', the paramater would then
be passed to the module and ipareplica deployment would fail.
By removing the parameters from the 'install' task, ipareplica
deployment works even if the variables are set by the user.
This is an ansible-freeipa update for the freeipa RFE:
https://pagure.io/freeipa/issue/9159
"`ipa-client-install` should provide option to enable `subid: sss`
in `/etc/nsswitch.conf`".
This option allows to configure authselect with the sssd
profile + with-subid feature, in order to have SSSD setup as
a datasource for subid in /etc/nsswitch.conf.
The default behavior remains unchanged: without the option,
/etc/nsswitch.conf keeps the line subid: files
Signed-off-by: Denis Karpelevich <dkarpele@redhat.com>
A temporary krb5 configuration was used to join the domain in
ipaclient_join. After that the final krkb5 configuration was created
with enabled DNS discovery and used for the remainaing tasks, where also
a connection to the IPA API was done.
With several servers the DNS discovery could have picked up a different
server. If the client deployment was faster than the replication this
could have lead to an unknown host error.
The issue was seen in performance testing where many simultaneous client
enrollments have been done..
The goal is to keep server affinity as long as possible within the
deployment process:
The temporary krb5.conf that was used before in ipaclient_join was
pulled out into an own module. The generated temporary krb5.conf is now
used in ipaclient_join and also ipaclient_api.
The generation of the final krb5.conf is moved to the end of the
deployment process.
Same as: https://pagure.io/freeipa/issue/9228
The setup of certmonger has been pulled out of ipaclient_setup_nss and moved
to the end of the process after generating the final krb5.conf as it will
use t will only use /etc/krb5.conf.
Certificate issuance may fail during deployment due to using the final
krb5.conf, but certmonger will re-try the request in this case.
Same as: https://pagure.io/freeipa/issue/9246
The test in ipaclient_test_keytab is at first trying to use an existing
krb5.conf to test if the host keytab can be used. With working DNS lookup
an absent krb5.conf is not reported as an error as DNS lookup is
silently used instead.
A temporary krb5.conf is now used in this test that forces to deactivate
DNS lookups and also to load /etc/krb5.conf. A missing krb5.conf is now
detected properly as the kinit call fails now properly. Thanks to Julien
Rische for this proposal.
ipaclient_test_keytab is now properly returning the state of usable or
not usable krb5.conf in krb5_conf_ok. This fixes the handling of this
case later on in the role.
Use Fully Qualified Collection Name (FQCN) for ansible builtins. This is
ansible.builtin.set_fact instead of set_fact for example and aplies for
all actions that are part of ansible.builtin.
All the replaced ansible.builtins:
assert, command, copy, debug, fail, fetch, file, import_playbook,
import_tasks, include_role, include_tasks, include_vars, package,
set_fact, shell, slurp, stat, systemd
Use Fully Qualified Collection Name (FQCN) for ansible builtins. This is
ansible.builtin.set_fact instead of set_fact for example and aplies for
all actions that are part of ansible.builtin.
All the replaced ansible.builtins:
assert, command, copy, debug, fail, fetch, file, import_playbook,
import_tasks, include_role, include_tasks, include_vars, package,
set_fact, shell, slurp, stat, systemd
Use Fully Qualified Collection Name (FQCN) for ansible builtins. This is
ansible.builtin.set_fact instead of set_fact for example and aplies for
all actions that are part of ansible.builtin.
All the replaced ansible.builtins:
assert, command, copy, debug, fail, fetch, file, import_playbook,
import_tasks, include_role, include_tasks, include_vars, package,
set_fact, shell, slurp, stat, systemd
Use Fully Qualified Collection Name (FQCN) for ansible builtins. This is
ansible.builtin.set_fact instead of set_fact for example and aplies for
all actions that are part of ansible.builtin.
All the replaced ansible.builtins:
assert, command, copy, debug, fail, fetch, file, import_playbook,
import_tasks, include_role, include_tasks, include_vars, package,
set_fact, shell, slurp, stat, systemd
