The server-del command passes str instance instead of bytes to
ldap.modify_s which results in the target server not being
removed properly.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Felipe Volpone <fbarreto@redhat.com>
Python 3 does not allow comparing SetuptoolsVersion and str
instances.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Stanislav Laznicka <slaznick@redhat.com>
There were several cases in ipaserver.install.certs where bytes
would be read/written as normal strings, this commit fixes that.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Felipe Volpone <fbarreto@redhat.com>
This commit enables ipa-client-install to be installable in
Python 3 and makes it run in Python 3 by default.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Martin Basti <mbasti@redhat.com>
Reviewed-By: Felipe Volpone <fbarreto@redhat.com>
We were trying to load a certificate as a string even though it
was an object already.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Martin Basti <mbasti@redhat.com>
Since OptionParser behaves differently in Python 2/3,
zonemgr_callback now handles value as str in both version.
https://pagure.io/freeipa/issue/5990
Reviewed-By: Stanislav Laznicka <slaznick@redhat.com>
It's nearly impossible to find out what happened when doing
replica connection check and it fails during the RPC phase.
The error is now logged.
Reviewed-By: Martin Basti <mbasti@redhat.com>
When version is mismatched and ipa-server-upgrade is required,
log the version mismatch properly in journal.
Signed-off-by: Tomas Krizek <tkrizek@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
certmonger returns PEM certificates with an additional newline
after the base64 encoded cert, remove it
https://pagure.io/freeipa/issue/4985
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
Certificates, both in PEM and DER format, should be handled as bytes
in Python 3.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
Up until now, Bytes parameter was used for certificate parameters
throughout the framework. However, the Bytes parameter does nothing
special for certificates, like validation, so this had to be done
for each of the parameters which were supposed to represent a
certificate.
This commit introduces a special Certificate parameter which takes
care of certificate validation so this does not have to be done
separately. It also makes sure that the certificates represented by
this parameter are always converted to DER format so that we can work
with them in a unified manner throughout the framework.
This commit also makes it possible to pass bytes directly during
instantiation of the Certificate parameter and they are still
represented correctly after their conversion in the _convert_scalar()
method.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
Previously, the type check of the Param class did only allow
the parameters to only have a value that's of a direct type.
However, that's nonsensically restrictive. For example, if
there's an interface implemented as an `ABCMeta` class then
the check for type fails since the interface's type is
`ABCMeta` instead of directly a `type`. Among others,
this is the case for cryptography.x509.Certificate.
Being a type is a transitive property of a Python object and we
should respect that in our framework.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
Recent certificate refactoring probably unclogged some failure
in handling certificates which causes test_changepw to correctly
fail since it is trying to connect using an HTTPS connection
without the CA certificate. This patch adds the CA cert to the
connection.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
load_unknown_x509_certificate() serves for the cases where we
can't be sure what the format of its input certificate is. This
is the case for installers, it should not be used anywhere else.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
Splitting the load_certificate() function into two separate helps
us word the requirements for the input explicitly. It also makes
our backend similar to the one of python-cryptography so eventually
we can swap python-cryptography for IPA x509 module.
https://pagure.io/freeipa/issue/4985
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
As we are now generating proper wildcard certificates, remove xfail
in wildcard certificate tests.
https://pagure.io/freeipa/issue/5603
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
During py.test initialization, the value 'debug' is passed instead
of logging.DEBUG.
Signed-off-by: Tomas Krizek <tkrizek@redhat.com>
Reviewed-By: David Kupka <dkupka@redhat.com>
jslint warned about parsing string to integer without explicit radix.
This error was introduced in commit 3cac851 .
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
KDC crashes on quite random interval while trying to reach LDAP
https://pagure.io/freeipa/issue/7017
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
Change makepki.sh for new makepki.py which should be more
readable, maintainable and extendable than the old script.
In this test we use it as a module and import create_pki().
The new makepki adds SKI and AKI extensions for correct
cert validation.
Other minor changes needed as we do not use NSS to store our
certificates on the test controller.
https://pagure.io/freeipa/issue/7030
Signed-off-by: Michal Reznik <mreznik@redhat.com>
Reviewed-By: Stanislav Laznicka <slaznick@redhat.com>
Previously, this configuration field was validated by integer_validator
which only checks that the input is number.
Now new positive_integer_validator can also check that
the inputed number positive.
https://pagure.io/freeipa/issue/6980
Reviewed-By: Felipe Volpone <felipevolpone@gmail.com>
Add new validator which inherits from integer validator
and checks whether the integer is positive.
https://pagure.io/freeipa/issue/6980
Reviewed-By: Felipe Volpone <felipevolpone@gmail.com>
Replace all `ipa_log_manager.log_mgr.get_logger` calls to create
module-level loggers with `logging.getLogger` calls and deprecate
`ipa_log_manager.log_mgr.get_logger`.
Reviewed-By: Martin Basti <mbasti@redhat.com>
Remove logger arguments in all functions and logger attributes in all
objects, with the exception of API object logger, which is now deprecated.
Replace affected logger calls with module-level logger calls.
Reviewed-By: Martin Basti <mbasti@redhat.com>
Remove all object-specific loggers, with the exception of `Plugin.log`,
which is now deprecated. Replace affected logger calls with module-level
logger calls.
Deprecate object-specific loggers in `ipa_log_manager.get_logger`.
Reviewed-By: Martin Basti <mbasti@redhat.com>
Use the actual root logger (`logging.getLogger()`) rather than the `ipa`
logger (or the `root` logger in case of ipa-ods-exporter) as the root
logger.
Always configure logging on the root logger.
Reviewed-By: Martin Basti <mbasti@redhat.com>
Use the standard `logging` module to configure logging instead of the
in-house `ipapython.log_manager` module and remove `ipapython.log_manager`.
Disable the logging-not-lazy and logging-format-interpolation pylint
checks.
Reviewed-By: Martin Basti <mbasti@redhat.com>
Do not configure stderr handler by default and let the application do the
configuration.
Fix ipa-dnskeysync-replica and ipa-dnskeysyncd not to add stderr handler
twice.
Reviewed-By: Martin Basti <mbasti@redhat.com>
Do not use custom Env instance to determine the debug level to use for the
IPA API object - the IPA API object can properly determine the configured
debug level on its own.
Reviewed-By: Martin Basti <mbasti@redhat.com>
Currently the cert-request execution is complicated and cannot
handle aliases in the --principal argument.
Implement the following simplifications:
- Search all user/host/service accounts at once, by krbPrincipalName
(error if no account found). Use principal canonical name to
determine the type of the principal.
- Update subject principals userCertificate attribute uniformly,
instead of dispatching to user/host/service-mod based on type of
principal.
Fixes: https://fedorahosted.org/freeipa/ticket/6531
Reviewed-By: Felipe Volpone <felipevolpone@gmail.com>
IPA will not print error message header when maximum
number of agreements per replica exceeded in topology.
Resolves: https://pagure.io/freeipa/issue/6533
Reviewed-By: David Kupka <dkupka@redhat.com>
Fedora system upgrades don't necessary require the network to be up.
However, we do, so we have to make sure the network is online before
performing an upgrade.
https://pagure.io/freeipa/issue/7039
Reviewed-By: Tomas Krizek <tkrizek@redhat.com>
Changed deletion of edges in remove_vertex method because there's no
need to store redundant variable in memory.
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
Hi!
I've just read the code and I saw that graph bfs uses not optimal
for Python solution. So I've edited it with more optimal one.
https://pagure.io/freeipa/issue/7051
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
Pytest can be executed from a machine that doesn't have IPA configured.
In this case, api can't be finalized because values such as basedn are
unknown and missing.
Fixes https://pagure.io/freeipa/issue/7046
Signed-off-by: Tomas Krizek <tkrizek@redhat.com>
Reviewed-By: Martin Basti <mbasti@redhat.com>
When pulling the DM password we may have the same issues reported in
ticket #6838 for CA keys.
This commit makes sure we always check the peer has keys before any
client operation.
Ticket #6838
Signed-off-by: Simo Sorce <simo@redhat.com>
Reviewed-By: Stanislav Laznicka <slaznick@redhat.com>
Reviewed-By: Michal Reznik <mreznik@redhat.com>
