]>
git.ipfire.org Git - thirdparty/freeradius-server.git/log
Nick Porter [Mon, 24 Nov 2025 09:46:54 +0000 (09:46 +0000)]
Add client_hello_parse option to enable extraction of Client Hello data
Nick Porter [Mon, 24 Nov 2025 09:32:52 +0000 (09:32 +0000)]
Define callback for extracting data from TLS Client Hello
Nick Porter [Mon, 24 Nov 2025 09:29:55 +0000 (09:29 +0000)]
Add dictionary entries for TLS Client Hello
Nick Porter [Mon, 24 Nov 2025 08:45:02 +0000 (08:45 +0000)]
Use log_request_proto_pair_list when reporting what will be sent
So we don't incorrectly report internal attributes as being sent.
Alan T. DeKok [Mon, 24 Nov 2025 01:36:43 +0000 (20:36 -0500)]
remove migration flag for unions
Alan T. DeKok [Mon, 24 Nov 2025 01:33:02 +0000 (20:33 -0500)]
remove hacks for key fields
Alan T. DeKok [Mon, 24 Nov 2025 01:19:19 +0000 (20:19 -0500)]
remove STRUCT dictionary keyword
it is no longer necessary
Alan T. DeKok [Sun, 23 Nov 2025 19:07:27 +0000 (14:07 -0500)]
start removing the ability for key fields to have children
Alan T. DeKok [Sun, 23 Nov 2025 14:35:45 +0000 (09:35 -0500)]
move final tests to new union syntax
Alan T. DeKok [Sun, 23 Nov 2025 14:34:15 +0000 (09:34 -0500)]
print out the filter pairs, for laziness
Alan T. DeKok [Sun, 23 Nov 2025 14:14:39 +0000 (09:14 -0500)]
move fprintf pair routines to pair_print
Alan T. DeKok [Sun, 23 Nov 2025 14:02:17 +0000 (09:02 -0500)]
take the raw attribute number from the value of the key field
Alan T. DeKok [Sun, 23 Nov 2025 13:58:01 +0000 (08:58 -0500)]
add helper function to turn any integer type into uint64_t
Alan T. DeKok [Sun, 23 Nov 2025 13:41:58 +0000 (08:41 -0500)]
we can't use BEGIN/END on structs
for now, we do auto-push.
Alan T. DeKok [Sun, 23 Nov 2025 12:58:20 +0000 (07:58 -0500)]
no need for BEGIN/END of struct
Alan T. DeKok [Sun, 23 Nov 2025 01:21:55 +0000 (20:21 -0500)]
move the rest of DHCPv6 to the new union syntax
DUID.foo --> foo
Hardware-Type.foo --> foo
Alan T. DeKok [Sun, 23 Nov 2025 01:21:24 +0000 (20:21 -0500)]
token names must be local, and not relative or absolute
Alan T. DeKok [Sun, 23 Nov 2025 01:16:23 +0000 (20:16 -0500)]
give more context for errors
Alan T. DeKok [Sun, 23 Nov 2025 01:12:56 +0000 (20:12 -0500)]
add note that copying ALIAS isn't done yet
Alan T. DeKok [Sat, 22 Nov 2025 17:22:04 +0000 (12:22 -0500)]
Partially move DHCPv6 to new union syntax
Alan T. DeKok [Sat, 22 Nov 2025 17:21:56 +0000 (12:21 -0500)]
tweak assertion again
Alan T. DeKok [Sat, 22 Nov 2025 17:04:25 +0000 (12:04 -0500)]
move DHCPv4 to new union syntax
Alan T. DeKok [Sat, 22 Nov 2025 16:00:00 +0000 (11:00 -0500)]
print the attribute name, and not the enum name
Alan T. DeKok [Sat, 22 Nov 2025 14:26:53 +0000 (09:26 -0500)]
move BFD to new union syntax
which also fixes some encoding errors in the previous code
Alan T. DeKok [Sat, 22 Nov 2025 14:26:32 +0000 (09:26 -0500)]
some fields can be raw
Alan T. DeKok [Sat, 22 Nov 2025 14:17:26 +0000 (09:17 -0500)]
don't check for encode_pair in union encoder
the encode_pair function is needed only by encode_tlv, which already
checks for the existence of encode_pair
Alan T. DeKok [Sat, 22 Nov 2025 14:12:39 +0000 (09:12 -0500)]
don't assert if we fail to instantiate the dictionaries
Alan T. DeKok [Sat, 22 Nov 2025 13:58:42 +0000 (08:58 -0500)]
move to new union syntax
Alan T. DeKok [Sat, 22 Nov 2025 13:58:29 +0000 (08:58 -0500)]
add notes on inter-attribute codes
Alan T. DeKok [Sat, 22 Nov 2025 00:30:26 +0000 (19:30 -0500)]
move DNS to using new UNIONs
Alan T. DeKok [Fri, 21 Nov 2025 21:44:30 +0000 (16:44 -0500)]
glue in copying of key attributes.
including creating enums in the correct order, and adding aliases
for children of a UNION
Alan T. DeKok [Fri, 21 Nov 2025 16:28:07 +0000 (11:28 -0500)]
copy STRUCT children in order
so that we can define KEY attributes before fields which need them.
Alan T. DeKok [Fri, 21 Nov 2025 21:37:51 +0000 (16:37 -0500)]
the new child is always parented by the new dst DA
otherwise the child is marked up as having the old parent, which
is wrong.
Alan T. DeKok [Fri, 21 Nov 2025 16:07:04 +0000 (11:07 -0500)]
refactor inner loop to function, in preparation for future work
Alan T. DeKok [Fri, 21 Nov 2025 14:49:27 +0000 (09:49 -0500)]
simplify and correct enum ext copying
nolade [Thu, 20 Nov 2025 20:57:37 +0000 (15:57 -0500)]
docs-v4: Update path for users.adoc xref link in src/original file
Alan T. DeKok [Fri, 21 Nov 2025 01:05:42 +0000 (20:05 -0500)]
move key_child_ref to extension
Alan T. DeKok [Thu, 20 Nov 2025 20:40:04 +0000 (15:40 -0500)]
use internal print, so we can do %pV and %pP
Alan T. DeKok [Thu, 20 Nov 2025 20:33:11 +0000 (15:33 -0500)]
we only need one begin/end
Alan T. DeKok [Thu, 20 Nov 2025 20:24:06 +0000 (15:24 -0500)]
DNS doesn't have attribute numbers, so use DEFINE
Alan T. DeKok [Wed, 19 Nov 2025 21:41:59 +0000 (16:41 -0500)]
allow FR_TYPE_ATTR to have enums (or not)
so that the code works properly in both cases
Alan T. DeKok [Wed, 19 Nov 2025 15:17:49 +0000 (10:17 -0500)]
limit the scope of 'clone='
and produce better errors, earlier
github-actions[bot] [Wed, 19 Nov 2025 04:25:34 +0000 (04:25 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/der.tar
github-actions[bot] [Wed, 19 Nov 2025 04:24:58 +0000 (04:24 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/cbor.tar
Alan T. DeKok [Wed, 19 Nov 2025 02:19:25 +0000 (21:19 -0500)]
We cannot copy KEY refs.
because they point to a key which is in the same parent as the
union. So copying the union would copy the key ref, which is bad.
the only way to copy a union would be to set a new key ref while
we're copying it.
Alan T. DeKok [Wed, 19 Nov 2025 02:10:35 +0000 (21:10 -0500)]
we cannot re-allocate or move a da which is a target of a ref
otherwise we have DAs which point to the old definitions, and
that's bad
Alan T. DeKok [Tue, 18 Nov 2025 20:37:03 +0000 (15:37 -0500)]
move to using explicit include list when checking for valid types
Alan T. DeKok [Tue, 18 Nov 2025 16:51:08 +0000 (11:51 -0500)]
check parent/child relationships in fr_dict_attr_add() too
Alan T. DeKok [Mon, 17 Nov 2025 19:13:15 +0000 (14:13 -0500)]
limit the places where we allow rcodes
we can't do math, so (-ok) is not allowed. but !ok is allowed.
Add xlat_purify_cond function. update tests to match.
Alan T. DeKok [Mon, 17 Nov 2025 18:25:48 +0000 (13:25 -0500)]
we can't do math on rcodes
so "invalid-packet" is not "invalid - packet"
Nick Porter [Mon, 17 Nov 2025 15:57:54 +0000 (15:57 +0000)]
Enable process tests
Nick Porter [Mon, 17 Nov 2025 14:59:56 +0000 (14:59 +0000)]
Add simple test of Relay-Forward
Nick Porter [Mon, 17 Nov 2025 14:57:36 +0000 (14:57 +0000)]
Correct xlat arguments
Nick Porter [Mon, 17 Nov 2025 14:55:28 +0000 (14:55 +0000)]
Tidy test dictionary
Nick Porter [Mon, 17 Nov 2025 14:15:33 +0000 (14:15 +0000)]
Disable checks on children of structs until UNION data types are done
Nick Porter [Mon, 17 Nov 2025 10:10:55 +0000 (10:10 +0000)]
Add symlinks to dictionaries needed for process tests
Nick Porter [Mon, 17 Nov 2025 10:06:52 +0000 (10:06 +0000)]
Correct enum and edit syntax in process tests
Alexis La Goutte [Wed, 24 Sep 2025 07:54:17 +0000 (09:54 +0200)]
xylan(dictionnary): Add new attribute (Priv-F-R3/R4/W3/W4)
Nick Porter [Mon, 17 Nov 2025 13:44:32 +0000 (13:44 +0000)]
Use correct namespace when resolving children of FR_TYPE_GROUP
Nick Porter [Sat, 15 Nov 2025 20:57:33 +0000 (20:57 +0000)]
Use consistent rctx for all dhcpv6 process sections
Functions defined in process.h expect PROCESS_RCTX to specify the type
in use - which is defined once per process module.
Alan T. DeKok [Sat, 15 Nov 2025 23:59:47 +0000 (18:59 -0500)]
print "::" more consistently for enums
but not for booleans, as true/false look bad with a "::" prefix.
The complicated checks for this condition are wrapped in an inline
function. The various unlang / edit / xlat / etc. print routines
then manually add it.
If instead we add "::" in fr_value_box_print(), then there are
many cases where we _don't_ want to print it. Rather than tweaking
all that through the code, we just instrument each part of the
printing code.
Alan T. DeKok [Sat, 15 Nov 2025 19:47:02 +0000 (14:47 -0500)]
add missing ';'
Alan T. DeKok [Fri, 14 Nov 2025 21:01:24 +0000 (16:01 -0500)]
show the server log when things fail
nolade [Thu, 6 Nov 2025 15:44:47 +0000 (10:44 -0500)]
docs: update path for users.adoc
Alan T. DeKok [Fri, 14 Nov 2025 13:03:44 +0000 (08:03 -0500)]
update last at the last minute.
Alan T. DeKok [Thu, 13 Nov 2025 19:29:26 +0000 (14:29 -0500)]
proposal for upgrade navbar
Alan T. DeKok [Thu, 13 Nov 2025 19:12:44 +0000 (14:12 -0500)]
move upgrade documentation one level up
in preparation for further splitting
Alan T. DeKok [Thu, 13 Nov 2025 15:53:26 +0000 (10:53 -0500)]
add and use 'auto' data type in dictionaries
which can only be used in association with "clone=..."
That way we don't need to duplicate the data type which is in the
cloned (i.e. source) attribute.
Note that we still allow a data type, though perhaps we want to
forbid that?
Alan T. DeKok [Thu, 13 Nov 2025 12:00:42 +0000 (07:00 -0500)]
Override new LLDB terminal window which breaks scrolling
Nick Porter [Fri, 14 Nov 2025 12:04:50 +0000 (12:04 +0000)]
Add test that %rest() can fetch a module call un-supported content type
Nick Porter [Fri, 14 Nov 2025 11:36:48 +0000 (11:36 +0000)]
The %rest() xlat should return the body regardless of content type
Since we're not doing any of the processing of the body which is done
during module method calls.
Nick Porter [Wed, 12 Nov 2025 14:07:49 +0000 (14:07 +0000)]
Add tests of compound keys in rlm_cache
Nick Porter [Thu, 13 Nov 2025 08:26:33 +0000 (08:26 +0000)]
Handle rlm_cache producing multiple values when expanding the key
Since the `key` expansion is set to output FR_TYPE_VOID, so that values
retain their native type, it cannot be set to concatenate values.
However, a key expansion of "%{foo}%{bar}" will produce 2 values which
need to be concatenated as a string before use.
Nick Porter [Wed, 12 Nov 2025 20:52:38 +0000 (20:52 +0000)]
Correct return handling
Wrapping the `ret = module...` assignment in `unlikely()` resulted in
incorrect values in `ret`
Nick Porter [Tue, 11 Nov 2025 14:41:51 +0000 (14:41 +0000)]
Add multi-thread gdb VS code target
Alan T. DeKok [Thu, 13 Nov 2025 02:26:06 +0000 (21:26 -0500)]
enum names might not be safe for SQL
Alan T. DeKok [Wed, 12 Nov 2025 15:49:39 +0000 (10:49 -0500)]
the migration option is gone, and we can delete the code
Alan T. DeKok [Wed, 12 Nov 2025 15:37:15 +0000 (10:37 -0500)]
add CONF_ITEM to attr_rules_t, for complaining with -C
so that when -C is used, and attribute references begin with '&',
it complains.
This should arguably be a parse error in 4.2
Alan T. DeKok [Tue, 11 Nov 2025 18:59:59 +0000 (13:59 -0500)]
catch the case of key fields
for
reply.Server-ID.DUID := ::UUID
Alan T. DeKok [Tue, 11 Nov 2025 16:54:51 +0000 (11:54 -0500)]
rename protocol pair encode/decode xlats
in preparation for adding protocol packet encode/decode xlats
github-actions[bot] [Tue, 11 Nov 2025 04:32:34 +0000 (04:32 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv6.tar
github-actions[bot] [Tue, 11 Nov 2025 04:31:45 +0000 (04:31 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/util.tar
github-actions[bot] [Tue, 11 Nov 2025 04:29:52 +0000 (04:29 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dns.tar
github-actions[bot] [Tue, 11 Nov 2025 04:27:50 +0000 (04:27 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/tftp.tar
github-actions[bot] [Tue, 11 Nov 2025 04:27:47 +0000 (04:27 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/vmps.tar
github-actions[bot] [Tue, 11 Nov 2025 04:27:43 +0000 (04:27 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/tacacs.tar
github-actions[bot] [Tue, 11 Nov 2025 04:27:35 +0000 (04:27 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv4.tar
github-actions[bot] [Tue, 11 Nov 2025 04:27:31 +0000 (04:27 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/radius.tar
github-actions[bot] [Tue, 11 Nov 2025 04:26:49 +0000 (04:26 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/bfd.tar
Arran Cudbard-Bell [Sat, 8 Nov 2025 16:47:06 +0000 (08:47 -0800)]
Fix infinite encoder loop encoding Message-Type as a foreign attribute
Arran Cudbard-Bell [Fri, 7 Nov 2025 01:35:09 +0000 (17:35 -0800)]
Don't use strlen in talloced buffers
Arran Cudbard-Bell [Fri, 10 Oct 2025 20:59:48 +0000 (14:59 -0600)]
Whitespace
Nick Porter [Thu, 6 Nov 2025 11:42:47 +0000 (11:42 +0000)]
Document %uuid.v4() and %uuid.v7()
Nick Porter [Thu, 6 Nov 2025 11:28:55 +0000 (11:28 +0000)]
Add test of %uuid.v4() and %uuid.v7()
Nick Porter [Thu, 6 Nov 2025 10:41:13 +0000 (10:41 +0000)]
Add %uuid.v4() and %uuid.v7() for generating UUIDs
Alan T. DeKok [Tue, 4 Nov 2025 21:23:13 +0000 (16:23 -0500)]
use the correct link
Alan T. DeKok [Sun, 2 Nov 2025 19:01:14 +0000 (14:01 -0500)]
handle corner cases of Protocol-Error
if there's no response configured, and the client supports
Protocol-Error, then synthesize a Protocol-Error with an
appropriate Error-Cause.
If there is a Protocol-Error response configured and the client
does not support it, then synthesize an appropriate response.
Nick Porter [Mon, 3 Nov 2025 12:52:47 +0000 (12:52 +0000)]
Add read / write hexdump option to TACACS listener
Nick Porter [Mon, 3 Nov 2025 11:55:47 +0000 (11:55 +0000)]
Add read / write hexdump options to RADIUS app_io
Nick Porter [Mon, 3 Nov 2025 11:52:58 +0000 (11:52 +0000)]
Add an app IO callback to set the read / write hexdump option