I added a comment in @d0b420f6d, later, on refactoring in
@c6769ad12 I forgot to remove it. So, it is just a clean up.
Related: https://pagure.io/freeipa/issue/8116
Signed-off-by: Stanislav Levin <slev@altlinux.org>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Commit 49cf5ec64b fixed a bug that
prevented migration from externally-signed to self-signed IPA CA.
But it introduced a subtle new issue: certmonger-initiated renewal
renews an externally-signed IPA CA as a self-signed CA.
To resolve this issue, introduce the `--force-self-signed' flag for
the dogtag-ipa-ca-renew-agent script. Add another certmonger CA
definition that calls this script with the `--force-self-signed'
flag. Update dogtag-ipa-ca-renew-agent to only issue a self-signed
CA certificate if the existing certificate is self-signed or if
`--force-self-signed' was given. Update `ipa-cacert-manage renew'
to supply `--force-self-signed' when appropriate.
As a result of these changes, certmonger-initiated renewal of an
externally-signed IPA CA certificate will not issue a self-signed
certificate.
Fixes: https://pagure.io/freeipa/issue/8176
Reviewed-By: Florence Blanc-Renaud <frenaud@redhat.com>
The helper function ipalib.util.check_client_configuration() now
considers a client configured when either:
* confdir is overridden (e.g. with IPA_CONFDIR) and the conf_default
file exists.
* confdir is /etc/ipa, /etc/ipa/default.conf exists and client
sysrestore state exists.
The check for sysrestore state is faster than checking for the presence
of the directory and presence of files in the directory. The sysrestore
state is always presence. sysrestore.index may be missing if no files
were backed up.
Fixes: https://pagure.io/freeipa/issue/8133
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Client connections no longer override TLS version range and ciphers by
default. Instead clients use the default settings from the system's
crypto policy.
Minimum TLS version is now TLS 1.2. The default crypto policy on
RHEL 8 sets TLS 1.2 as minimum version, while Fedora 31 sets TLS 1.0 as
minimum version. The minimum version is configured with OpenSSL 1.1.1
APIs. Python 3.6 lacks the setters to override the system policy.
The effective minimum version is always TLS 1.2, because FreeIPA
reconfigures Apache HTTPd on Fedora.
Fixes: https://pagure.io/freeipa/issue/8125
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Thomas Woerner <twoerner@redhat.com>
urllib3 now supports post-handshake authentication with TLS 1.3. Enable
TLS 1.3 support for Apache HTTPd.
The update depends on bug fixes for TLS 1.3 PHA support in urllib3 and
Apache HTTPd. New builds are available in freeipa-master COPR and in
F30/F31.
Overwrite crypto-policy on Fedora only. Fedora 31 and earlier have TLS
1.0 and 1.1 still enabled by default.
Fixes: https://pagure.io/freeipa/issue/8125
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Thomas Woerner <twoerner@redhat.com>
When ipa user-add-certmapdata is called with multiple --subject or
multiple --issuer, the DNParam's _convert_scalar method is called with
a tuple containing all the params and should raise an exception as the
--subject and --issuer are single-value params.
The DNParam _convert_scalar method internally calls the DN init method,
and the DN init method is able to create a DN from a tuple of RDNs.
As such, it won't raise exception if a tuple/list is provided.
Check that _convert_scalar is only provided a single element.
Fixes: https://pagure.io/freeipa/issue/8097
Signed-off-by: Florence Blanc-Renaud <flo@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
New Pylint (2.4.3) catches several new 'true problems'. At the same
time, it warns about things that are massively and reasonably
employed in FreeIPA.
list of fixed:
- no-else-continue
- redeclared-assigned-name
- no-else-break
- unnecessary-comprehension
- using-constant-test (false positive)
list of ignored (responsibility of contributors and reviewers):
- import-outside-toplevel
Fixes: https://pagure.io/freeipa/issue/8102
Signed-off-by: Stanislav Levin <slev@altlinux.org>
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
This patche removes 93 pylint deprecation warnings due to invalid escape
sequences (mostly 'invalid escape sequence \d') on unicode strings.
Signed-off-by: Rafael Guterres Jeffman <rjeffman@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
If a certmonger CA is not defined but is referenced within
a request (so was removed sometime after a request was
created) then anything that pulls all certmonger requests would
fail with the cryptic error:
"Failed to get request: bus, object_path and dbus_interface
must not be None."
This was often seen during upgrades.
Catch this specific condition and report a more specific error
so the user will have some bread crumb to know how to address
the issue.
https://pagure.io/freeipa/issue/7870
Signed-off-by: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Florence Blanc-Renaud <frenaud@redhat.com>
The variable is intended to control the timeout for replication
events. If someone had significantly reduced it via configuration
then it could have caused certmogner requests to fail due to timeouts.
Add replication_wait_timeout, certmonger_wait_timeout and
http_timeout to the default.conf man page.
Related: https://pagure.io/freeipa/issue/7971
Reviewed-By: Florence Blanc-Renaud <flo@redhat.com>
Currently, RA agent certificate is issued using caServerCert profile.
This has unfortunate side effect of asserting id-pk-serverAuth EKU which
is not really needed for RA agent. If IPA CA certificate adds SAN DNS
constraints into issued certificates, presence of id-pk-serverAuth EKU
forces NSS (and other crypto libraries) to validate CN value with
regards to SAN DNS constraints, due to historical use of CN bearing DNS
name.
Since RA agent certificate has 'CN=IPA RA', it is guaranteed to fail
the check.
Default IPA CA configuration does *not* add SAN DNS constraints into RA
agent certificate. However, it is better to be prepared to such
behavior.
Related: https://bugzilla.redhat.com/1670239
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Profile-based renewal means we should always explicitly specify the
profile in tracking requests that use the dogtag-ipa-ca-renew-agent
renewal helper. This includes the IPA RA agent certificate. Update
CAInstance.configure_agent_renewal() to add the profile to the
tracking request. This also covers the upgrade scenario (because
the same method gets invoked).
Part of: https://pagure.io/freeipa/issue/7991
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
As we expand the integration tests for external CA functionality, it
is helpful (and avoids duplication) to use the MSCSTemplate*
classes. These currently live in ipaserver.install.cainstance, but
ipatests is no longer permitted to import from ipaserver (see commit
81714976e5e13131654c78eb734746a20237c933). So move these classes to
ipalib.
Part of: https://pagure.io/freeipa/issue/7548
Reviewed-By: Florence Blanc-Renaud <flo@redhat.com>
IPA now uses the system-wide crypto policy for TLS ciphers on RHEL. It's
also now possible to keep the default policy by setting TLS_HIGH_CIPHERS
to None.
Fixes: https://pagure.io/freeipa/issue/7998
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
TLS 1.3 is causing some trouble with client cert authentication.
Conditional client cert authentication requires post-handshake
authentication extension on TLS 1.3. The new feature is not fully
implemented yet.
TLS 1.0 and 1.1 are no longer state of the art and now disabled by
default.
TLS 1.2 works everywhere and supports PFS.
Related: https://pagure.io/freeipa/issue/7667
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
The command
ipa dnsrecord-mod <zone> <record> --ttl
requires to provide at least one record to modify. When none
is specified, it prompts by proposing each of the existing records,
for instance:
ipa dnsrecord-mod ZZZZZ.org ns11 --ttl=86400
No option to modify specific record provided.
Current DNS record contents:
A record: xxx.xxx.xxx.xxx
AAAA record: xxxx:xx
Modify A record 'xxxx.xxxx.xxxx.xxxx'? Yes/No (default No):
Modify AAAA record 'xxxx:xx'? Yes/No (default No):
ipa: ERROR: No options to modify a specific record provided.
The admin should be able to modify the TTL value without
re-entering the record information. The issue happens because of an
internal check that forgot to consider 'dnsttl' as a valid standalone
modification.
Fixes: https://pagure.io/freeipa/issue/7982
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
If lightweight CA key replication has not completed, requests for
the certificate or chain will return 404**. This can occur in
normal operation, and should be a temporary condition. Detect this
case and handle it by simply omitting the 'certificate' and/or
'certificate_out' fields in the response, and add a warning message
to the response.
Also update the client-side plugin that handles the
--certificate-out option. Because the CLI will automatically print
the warning message, if the expected field is missing from the
response, just ignore it and continue processing.
** after the Dogtag NullPointerException gets fixed!
Part of: https://pagure.io/freeipa/issue/7964
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
On Linux systems the length limit for hostnames is hardcoded
at 64 in MAXHOSTNAMELEN
Solaris, for example, allows 255 characters, and DNS allows the
total length to be up to 255 (with each label < 64).
Add a knob to allow configuring the maximum hostname length (FQDN)
The same validators are used between hosts and DNS to apply
the knob only when dealing with a FQDN as a hostname.
The maxlen option is included so installers can limit the length
of allowed hostnames when the --hostname option is used.
https://pagure.io/freeipa/issue/2018
Signed-off-by: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Florence Blanc-Renaud <frenaud@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
The modern PKCS#8 private key format supports better encryption standard
and is preferable over traditional, weak PKCS#1 key format.
Fixes: https://pagure.io/freeipa/issue/7943
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Oleg Kozlov <okozlov@redhat.com>
This reverts commit 3509545897.
We cannot force increase in minimum SASL security factor until our
consumers are ready to deal with it. Unfortunately, realmd uses
anonymous connection for discovery and validation of IPA LDAP server.
The way it is done is fragile (it doesn't take into account an
advertised IPA version, only checks that 'IPA' string exists in the info
field) but since bumping of minimum SSF prevents reading IPA info field
using anonymous connection, client enrollment fails.
We should get back to bumping minimum SSF after realmd and other
potential consumers are fixed.
Reviewed-By: François Cami <fcami@redhat.com>
For HSM support, IPA has to pass the token name for CA and subsystem
certificates to certmonger. For now, only the default 'internal' token is
supported.
Related: https://pagure.io/freeipa/issue/5608
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
man(5) default.conf says that startup_timeout has a default value of 120
seconds. Even 120 seconds are not effective unless systemd is also
reconfigured to have a larger DefaultTimeoutStartSec.
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
The object logger methods been deprecated for about two years since release
4.6.0. The log manager used to moneky-patch additional log methods like
info(), warning(), and error() into API plugin objects. The methods have
been replaced by calls to module logger objects in 4.6.0.
Remove monkey-patch logger methods, log manager, and its root logger from
ipapython.ipa_log_manager.
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Thomas Woerner <twoerner@redhat.com>
Certmonger has hard-coded defaults for key size and key type. In case a
request does not contain these values, certmonger uses 2048 RSA keys.
Since the CA now has 3072, it will also rekey the CA to 2048 instead of
resubmitting with the existing 2048 bit key.
Use key-size and key-type from the existing request when resubmitting.
Related: https://pagure.io/freeipa/issue/6790
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
SSF_MINX 56 level ensures data integrity and confidentiality for SASL
GSSAPI and SASL GSS SPNEGO connections.
Although at least AES128 is enforced pretty much everywhere, 56 is required
for backwards compatibility with systems that announce wrong SSF.
Related: https://pagure.io/freeipa/issue/7140
Related: https://pagure.io/freeipa/issue/4580
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Tibor Dudlak <tdudlak@redhat.com>
The HOME directory may not be available in containers. It's also the
wrong variable on some platforms. Use os.path.expanduser() instead of
HOME.
Fixes: https://pagure.io/freeipa/issue/7837
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
Fixes a test ipatests/test_cmdline/test_cli.py:test_cli_fs_encoding()
which sets IPA_CONFDIR and attempts to interpret the resulting error
message. However, if the test is run on an enrolled machine (a
developer's laptop, for example), check_client_configuration() will
succeed because it ignores IPA_CONFDIR and, as result, api.finalize()
will fail later with a stacktrace.
Pass an environment object and test an overridden config file existence
in this case to fail with a proper and expected message.
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Get the value of `PAGER` environment variable in case it's defined, check the executable, if it exists - use a pager, otherwise - print function.
Fixes: https://pagure.io/freeipa/issue/7746
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Rob Crittenden <rcritten@redhat.com>
pylint 2.2.0 has a new checker for unnecessary pass statements. There is
no need to have a pass statement in functions or classes with a doc
string.
Fixes: https://pagure.io/freeipa/issue/7772
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Thomas Woerner <twoerner@redhat.com>
If the server responds 401 and the response body is empty, the
client raises ResponseNotReady. This occurs because:
1. For a non-200 response, the response read only if the
Content-Length header occurs.
2. The response must be read before another request (e.g. the
follow-up request with WWW-Authenticate header set), and this
condition was not met. For details see
https://github.com/python/cpython/blob/v3.6.7/Lib/http/client.py#L1305-L1321.
This situation should not arise in regular use, because the client
either has a session cookie, or, knowing the details of the server
it is contacting, it establishes the GSS-API context and includes
the WWW-Authenticate header in the initial request.
Nevertheless, this problem has been observed in the wild. I do not
know its ordinary cause(s), but one can force the issue by removing
an authenticated user's session cache from /run/ipa/ccaches, then
performing a request.
Resolve the issue by always reading the response. It is safe to
call response.read() regardless of whether the Content-Length header
appears, or whether the body is empty.
Fixes: https://pagure.io/freeipa/issue/7752
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Some incorrect errors are possible if a client installation
fails and a configuration rollback is required.
These include:
1. Unconfigured automount client failed: CalledProcessError(Command
['/usr/sbin/ipa-client-automount', '--uninstall', '--debug']
returned non-zero exit status 1: '')
Caused by check_client_configuration() not returning the correct
return value (2).
2. WARNING: Unable to revert to the pre-installation state ('authconfig'
tool has been deprecated in favor of 'authselect'). The default sssd
profile will be used instead.
The authconfig arguments would have been: authconfig --disableldap
--disablekrb5 --disablesssdauth --disablemkhomedir
If installation fails before SSSD is configured there is no state
to roll back to. Detect this condition.
3. An error occurred while removing SSSD's cache.Please remove the
cache manually by executing sssctl cache-remove -o.
Again, if SSSD is not configured yet then there is no cache to
remove. Also correct the missing space after the period.
https://pagure.io/freeipa/issue/7729
Signed-off-by: Rob Crittenden <rcritten@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Replace six.moves and six.StringIO/BytesIO imports with cannonical
Python 3 packages.
Note: six.moves.input behaves differently than builtin input function.
Therefore I left six.moves.input for now.
See: https://pagure.io/freeipa/issue/7715
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
In Python 3, six.integer_types is (int,). In most places, the alias can
be simply replaced with int. In other places, it was possible to
simplify the code by unpacking the tuple.
See: https://pagure.io/freeipa/issue/7715
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Florence Blanc-Renaud <frenaud@redhat.com>
In Python 3, six.string_types is just an alias for str.
See: https://pagure.io/freeipa/issue/7715
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Florence Blanc-Renaud <frenaud@redhat.com>
Python 2 had old style and new style classes. Python 3 has only new
style classes. There is no point to subclass from object any more.
See: https://pagure.io/freeipa/issue/7715
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Florence Blanc-Renaud <frenaud@redhat.com>
tox / pytest is complaining about lots and lots of invalid escape
sequences in our code base. Sprinkle raw strings or backslash escapes
across the code base to fix most occurences of:
DeprecationWarning: invalid escape sequence
There is still one warning that keeps repeating, though:
source:264: DeprecationWarning: invalid escape sequence \d
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Florence Blanc-Renaud <frenaud@redhat.com>
pyasn1 0.4 changed handling of ANY containers in a backwards
incompatible way. For 0.3.x, keep explicit wrap and unwrap in octet
strings for ANY container members. For >= 0.4, let pyasn1 do the job.
This patch also makes sorting of extended_key_usage_bytes() stable and
adds tests.
Tested with pyasn1 0.3.7 and 0.4.4.
Fixes: https://pagure.io/freeipa/issue/7685
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Fraser Tweedale <ftweedal@redhat.com>
The replica_file option is not needed anymore. Threfore the option can
be removed.
See: https://pagure.io/freeipa/issue/7689
Signed-off-by: Thomas Woerner <twoerner@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
The replica_file option is only supported for DL0. The option will be
marked deprecated for now.
See: https://pagure.io/freeipa/issue/7669
Signed-off-by: Thomas Woerner <twoerner@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
With increasing the minimal domain level to 1 ipa-replica-install will
refuse to install if the domain has domain level 0.
See: https://pagure.io/freeipa/issue/7669
Signed-off-by: Thomas Woerner <twoerner@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Fix the following violations aiming to support Pylint 2.0
- `unneeded-not` (C0113):
Consider changing "not item in items" to "item not in items" used
when a boolean expression contains an unneeded negation.
- `useless-import-alias` (C0414):
Import alias does not rename original package Used when an import
alias is same as original package.e.g using import numpy as numpy
instead of import numpy as np
- `raising-format-tuple` (W0715):
Exception arguments suggest string formatting might be intended Used
when passing multiple arguments to an exception constructor, the
first of them a string literal containing what appears to be
placeholders intended for formatting
- `bad-continuation` (C0330):
This was already included on the disable list, although with current
version of pylint (2.0.0.dev2) violations at the end of the files
are not being ignored.
See: https://github.com/PyCQA/pylint/issues/2278
- `try-except-raise` (E0705):
The except handler raises immediately Used when an except handler
uses raise as its first or only operator. This is useless because it
raises back the exception immediately. Remove the raise operator or
the entire try-except-raise block!
- `consider-using-set-comprehension` (R1718):
Consider using a set comprehension Although there is nothing
syntactically wrong with this code, it is hard to read and can be
simplified to a set comprehension.Also it is faster since you don't
need to create another transient list
- `dict-keys-not-iterating` (W1655):
dict.keys referenced when not iterating Used when dict.keys is
referenced in a non-iterating context (returns an iterator in
Python 3)
- `comprehension-escape` (W1662):
Using a variable that was bound inside a comprehension Emitted when
using a variable, that was bound in a comprehension handler, outside
of the comprehension itself. On Python 3 these variables will be
deleted outside of the comprehension.
Issue: https://pagure.io/freeipa/issue/7614
Signed-off-by: Armando Neto <abiagion@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
pylinti 2 is having a hard time to handle name mangled, magic attributes
correctly. Double under attributes like __d are internally renamed to
_Env__d. After multiple failed attempts, it was easier to just add more
pylint disable to the implementation.
pylint 2 also thinkgs that Env.server is defined much later or the env
doesn't have that member at all. Ignore the false warnings, too.
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Armando Neto <abiagion@redhat.com>
THe ipa console command takes an optional filename argument. The
filename argument was broken, because the implementation passed a file
object to exec() instead of a string or compiled object.
ipa console now uses compile() to compile the code with print_function
__future__ feature.
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
ipa console is a useful tool to use FreeIPA's API in an interactive
Python console. The patch adds readline tab completion and history
support.
Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-By: Alexander Bokovoy <abokovoy@redhat.com>
In order to support pylint 2.0 the following violations must be fixed:
- `chained-comparison` (R1716):
Simplify chained comparison between the operands This message is
emitted when pylint encounters boolean operation like
"a < b and b < c", suggesting instead to refactor it to "a < b < c".
- `consider-using-in` (R1714):
Consider merging these comparisons with "in" to %r To check if a
variable is equal to one of many values,combine the values into a
tuple and check if the variable is contained "in" it instead of
checking for equality against each of the values.This is faster
and less verbose.
Issue: https://pagure.io/freeipa/issue/7614
Signed-off-by: Armando Neto <abiagion@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
Aiming to support pylint 2.0 some functions and methods must have their
return statements updated in order to fix two new violations:
- `useless-return` (R1711):
Useless return at end of function or method Emitted when a single
"return" or "return None" statement is found at the end of function
or method definition. This statement can safely be removed because
Python will implicitly return None
- `inconsistent-return-statements` (R1710):
Either all return statements in a function should return an
expression, or none of them should. According to PEP8, if any return
statement returns an expression, any return statements where no value
is returned should explicitly state this as return None, and an
explicit return statement should be present at the end of the
function (if reachable)
Issue: https://pagure.io/freeipa/issue/7614
Signed-off-by: Armando Neto <abiagion@redhat.com>
Reviewed-By: Christian Heimes <cheimes@redhat.com>
