]>
git.ipfire.org Git - thirdparty/freeradius-server.git/log
Alexander Clouter [Mon, 13 Feb 2017 17:10:16 +0000 (17:10 +0000)]
fix radrelay
Mark Johnson [Thu, 9 Feb 2017 20:15:37 +0000 (12:15 -0800)]
Updating Nomadix dictionary with missing attributes.
Arran Cudbard-Bell [Fri, 10 Feb 2017 14:25:33 +0000 (14:25 +0000)]
Fix dictionary test
Arran Cudbard-Bell [Fri, 10 Feb 2017 13:23:28 +0000 (13:23 +0000)]
Fix typo in alcatel dictionaries
Arran Cudbard-Bell [Fri, 10 Feb 2017 13:43:32 +0000 (13:43 +0000)]
Better error message
Arran Cudbard-Bell [Fri, 10 Feb 2017 13:43:22 +0000 (13:43 +0000)]
When checking to see if the attribute is a duplicate we need to check the parent too
This used to work because the TLV nesting was part of the attribute number.
Arran Cudbard-Bell [Fri, 10 Feb 2017 13:42:34 +0000 (13:42 +0000)]
Shouldn’t free attributes when the name is freed
It’s OK to have attributes in the dictionary where the name has been subsumed by a later definition
Alan T. DeKok [Sun, 5 Feb 2017 22:43:34 +0000 (17:43 -0500)]
use fr_strerror_printf() instead of ERROR
Alan T. DeKok [Sun, 5 Feb 2017 18:43:59 +0000 (13:43 -0500)]
more removal of pair_make
Alan T. DeKok [Sun, 5 Feb 2017 18:30:34 +0000 (13:30 -0500)]
use numbers, not names
Alan T. DeKok [Sun, 5 Feb 2017 16:17:46 +0000 (11:17 -0500)]
check ptr
Alan T. DeKok [Sun, 5 Feb 2017 16:16:35 +0000 (11:16 -0500)]
close FD on one more path, too
Alan T. DeKok [Sun, 5 Feb 2017 16:15:37 +0000 (11:15 -0500)]
remove unnecessary check
Alan T. DeKok [Sun, 5 Feb 2017 16:13:35 +0000 (11:13 -0500)]
comment-out generic code
Arran Cudbard-Bell [Thu, 2 Feb 2017 10:21:41 +0000 (10:21 +0000)]
Update copyright year
Alan T. DeKok [Sun, 5 Feb 2017 15:20:57 +0000 (10:20 -0500)]
more ignore discard
Alan T. DeKok [Sun, 5 Feb 2017 15:20:26 +0000 (10:20 -0500)]
set FALL-THROUGH
Alan T. DeKok [Sun, 5 Feb 2017 15:14:30 +0000 (10:14 -0500)]
more "ignore discard"
Alan T. DeKok [Sun, 5 Feb 2017 15:13:29 +0000 (10:13 -0500)]
check data length both ways
Alan T. DeKok [Sun, 5 Feb 2017 15:10:47 +0000 (10:10 -0500)]
check pointer before dereferencing it
Alan T. DeKok [Sun, 5 Feb 2017 15:09:44 +0000 (10:09 -0500)]
note fall-through
Alan T. DeKok [Sun, 5 Feb 2017 15:08:48 +0000 (10:08 -0500)]
ignore discard return code
Alan T. DeKok [Sun, 5 Feb 2017 15:07:28 +0000 (10:07 -0500)]
ignore return code from discarding a packet
Alan T. DeKok [Sun, 5 Feb 2017 15:07:00 +0000 (10:07 -0500)]
ignore return code we can't use anyways
Alan T. DeKok [Sun, 5 Feb 2017 15:04:00 +0000 (10:04 -0500)]
ignore return code
Alan T. DeKok [Sun, 5 Feb 2017 15:02:44 +0000 (10:02 -0500)]
we don't care about this return value
Alan T. DeKok [Sun, 5 Feb 2017 15:02:11 +0000 (10:02 -0500)]
don't assign in an assert
Alan T. DeKok [Sun, 5 Feb 2017 15:01:16 +0000 (10:01 -0500)]
cast to "uint8_t *" before adding byte offsets
Alan T. DeKok [Sun, 5 Feb 2017 14:59:41 +0000 (09:59 -0500)]
return on error
Alan T. DeKok [Sun, 5 Feb 2017 14:56:05 +0000 (09:56 -0500)]
remove unused variable
Alan T. DeKok [Sun, 5 Feb 2017 14:53:33 +0000 (09:53 -0500)]
it's good to close files we've opened..
Alan T. DeKok [Sun, 5 Feb 2017 14:38:51 +0000 (09:38 -0500)]
update hash based on client port, too
Alan T. DeKok [Sun, 5 Feb 2017 14:37:31 +0000 (09:37 -0500)]
a few more checks on length
Arran Cudbard-Bell [Sat, 4 Feb 2017 19:54:01 +0000 (19:54 +0000)]
Split TLS caching operation
OpenSSL now receives tls_cache_serialize as the callback
for storing session resumption data, and the EAP-TLS code calls tls_cache_write separately when sending an EAP-TLS success, to cache the session data.
Alan T. DeKok [Sat, 4 Feb 2017 15:30:11 +0000 (10:30 -0500)]
remove unused macro
Alan T. DeKok [Sat, 4 Feb 2017 15:14:44 +0000 (10:14 -0500)]
remove unused variable
Alan T. DeKok [Sat, 4 Feb 2017 14:40:23 +0000 (09:40 -0500)]
remove extraneous check
Alan T. DeKok [Sat, 4 Feb 2017 14:39:17 +0000 (09:39 -0500)]
add missing 'break'
Alan T. DeKok [Sat, 4 Feb 2017 14:38:35 +0000 (09:38 -0500)]
initialize salen
Alan T. DeKok [Sat, 4 Feb 2017 14:34:25 +0000 (09:34 -0500)]
we don't care about the return value when discarding packets
Alan T. DeKok [Sat, 4 Feb 2017 14:33:53 +0000 (09:33 -0500)]
check status of pthread_create
Alan T. DeKok [Sat, 4 Feb 2017 14:32:36 +0000 (09:32 -0500)]
lower bound on maximum stack depth
Alan T. DeKok [Sat, 4 Feb 2017 14:31:48 +0000 (09:31 -0500)]
reference correct field
Alan T. DeKok [Sat, 4 Feb 2017 14:30:11 +0000 (09:30 -0500)]
add missing 'break'
Alan T. DeKok [Sat, 4 Feb 2017 14:28:53 +0000 (09:28 -0500)]
add missing va_end
Alan T. DeKok [Sat, 4 Feb 2017 14:28:03 +0000 (09:28 -0500)]
add missing 'break'
Arran Cudbard-Bell [Sat, 4 Feb 2017 13:37:49 +0000 (13:37 +0000)]
Avoid memcpy when obtaining session ID
Alan T. DeKok [Fri, 3 Feb 2017 18:58:25 +0000 (13:58 -0500)]
remove unused assignment
Alan T. DeKok [Fri, 3 Feb 2017 18:29:50 +0000 (13:29 -0500)]
initialize variable
Alan T. DeKok [Fri, 3 Feb 2017 18:28:04 +0000 (13:28 -0500)]
use file.data for data
Alan T. DeKok [Fri, 3 Feb 2017 18:25:40 +0000 (13:25 -0500)]
remove unused assignment
Alan T. DeKok [Fri, 3 Feb 2017 15:25:59 +0000 (10:25 -0500)]
move dict_tacacs_root to libfreeradius-tacacs
Alan T. DeKok [Fri, 3 Feb 2017 15:24:13 +0000 (10:24 -0500)]
initialize variable
Alan T. DeKok [Fri, 3 Feb 2017 14:56:47 +0000 (09:56 -0500)]
Add TACACS+
Matthew Newton [Thu, 2 Feb 2017 21:10:43 +0000 (21:10 +0000)]
systemd syslog.target is obsolete
Arran Cudbard-Bell [Thu, 2 Feb 2017 10:16:43 +0000 (10:16 +0000)]
Hmm, maybe we shouldn't create the database
Arran Cudbard-Bell [Thu, 2 Feb 2017 00:20:03 +0000 (00:20 +0000)]
Removed unused variable
Arran Cudbard-Bell [Wed, 1 Feb 2017 23:53:49 +0000 (23:53 +0000)]
Update rlm_sql.c
Arran Cudbard-Bell [Wed, 1 Feb 2017 20:51:06 +0000 (20:51 +0000)]
Use the actual field lengths when creating the result array
# Conflicts:
# src/modules/rlm_sql/drivers/rlm_sql_mysql/rlm_sql_mysql.c
Arran Cudbard-Bell [Wed, 1 Feb 2017 20:50:14 +0000 (20:50 +0000)]
Trim whitespace before searching for operation type
# Conflicts:
# src/modules/rlm_sql/rlm_sql.c
Arran Cudbard-Bell [Wed, 1 Feb 2017 20:44:00 +0000 (20:44 +0000)]
Create the database by default...
Alan T. DeKok [Wed, 1 Feb 2017 19:41:05 +0000 (14:41 -0500)]
made more pretty
Alan T. DeKok [Wed, 1 Feb 2017 19:35:07 +0000 (14:35 -0500)]
update for prettiness
Matthew Newton [Wed, 1 Feb 2017 13:54:55 +0000 (13:54 +0000)]
update kibana dashboard so it doesn't have to be imported twice
define the search before the visualisations
Arran Cudbard-Bell [Sat, 28 Jan 2017 16:10:26 +0000 (16:10 +0000)]
Use fr_dict_oid_component in fr_dict_unknown_afrom_oid_str
Alan T. DeKok [Mon, 30 Jan 2017 18:28:05 +0000 (13:28 -0500)]
one more 'const'
Alan T. DeKok [Mon, 30 Jan 2017 17:42:19 +0000 (12:42 -0500)]
one more 'const'
Alan T. DeKok [Mon, 30 Jan 2017 17:03:53 +0000 (12:03 -0500)]
return "known" unknown attribute if it exists
after creating unknown attribute, look it up by name. If it
exists, return that one. This lets us use unknown attributes
in the config files, and then have the decoder automatically
reference then when they're seen in the packets.
Alan T. DeKok [Mon, 30 Jan 2017 17:03:34 +0000 (12:03 -0500)]
make fr_dict_unknown_afrom_fields() return const
Arran Cudbard-Bell [Sat, 28 Jan 2017 15:24:27 +0000 (15:24 +0000)]
Prefix references
Arran Cudbard-Bell [Sat, 28 Jan 2017 14:09:36 +0000 (14:09 +0000)]
Shut up doxygen?
Arran Cudbard-Bell [Sat, 28 Jan 2017 13:33:11 +0000 (13:33 +0000)]
Doxygen and other minor fixes
Arran Cudbard-Bell [Sat, 28 Jan 2017 12:57:35 +0000 (12:57 +0000)]
Add checks in more places parent should not be NULL
(everwhere)
Arran Cudbard-Bell [Sat, 28 Jan 2017 12:52:56 +0000 (12:52 +0000)]
Fix clang scan issues
Arran Cudbard-Bell [Sat, 28 Jan 2017 12:45:01 +0000 (12:45 +0000)]
Simplify unknown attribute OID parsing/printing
As tmpls are no longer stack allocated, we can hang additional talloced memory off the structure.
This means unknown attributes no longer need to be placed in a buffer which is part of the tmpl, but can be allocated on the heap.
It also means that where an unknown attribute hierarchy contains an unknown vendor, the unknown vendor can form part of the hierarchy too, without needing a special field.
References #1883
Alan T. DeKok [Fri, 27 Jan 2017 01:17:04 +0000 (20:17 -0500)]
UNUSED -> used
Alan T. DeKok [Fri, 27 Jan 2017 00:30:11 +0000 (19:30 -0500)]
more debugging
and print out packet in decode routine, not in recv routine
Alan T. DeKok [Thu, 26 Jan 2017 20:34:30 +0000 (15:34 -0500)]
don't write empty packets to the detail file
Arran Cudbard-Bell [Wed, 25 Jan 2017 11:29:58 +0000 (11:29 +0000)]
Formatting
Arran Cudbard-Bell [Wed, 25 Jan 2017 11:07:20 +0000 (11:07 +0000)]
Write type directly to the tmpl
Use a dynamically allocated buffer for unknown attribute names
Alan T. DeKok [Tue, 24 Jan 2017 21:59:36 +0000 (16:59 -0500)]
allow EVS for unknown attributes, too
Arran Cudbard-Bell [Tue, 24 Jan 2017 17:20:24 +0000 (17:20 +0000)]
Remove references to tmpl_from_attr_substr
Arran Cudbard-Bell [Tue, 24 Jan 2017 14:04:17 +0000 (14:04 +0000)]
Remove stack based variants tmpl_attr functions
A major limitation of using stack based allocations was that nested templates couldn't be used.
This meant syntax like Tmp-Attr-1[&Tmp-Integer-0] wasn't possible, as there was no way to include the structure required for the index reference without allocating it.
Alan T. DeKok [Mon, 23 Jan 2017 21:11:36 +0000 (16:11 -0500)]
tmpl_from_attr_substr is now private
Alan T. DeKok [Mon, 23 Jan 2017 21:10:26 +0000 (16:10 -0500)]
use tmpl_afrom_attr_substr() here, too
Alan T. DeKok [Mon, 23 Jan 2017 21:08:34 +0000 (16:08 -0500)]
arg. missed in previous commit. Stupid git
Alan T. DeKok [Mon, 23 Jan 2017 20:45:58 +0000 (15:45 -0500)]
make node->attr a talloc'd pointer
And move alternate into a "union" with attr.
And add a "regex_index" field to the union, because we can't
just steal node->attr->tmpl_num. Which was always a hack.
Alan T. DeKok [Mon, 23 Jan 2017 20:35:24 +0000 (15:35 -0500)]
use tmpl_afrom_attr_substr, not the stack-based version
Alan T. DeKok [Mon, 23 Jan 2017 20:09:09 +0000 (15:09 -0500)]
s/new/n/ "new" is a bad name for a debugger which knows about C++
Alan T. DeKok [Mon, 23 Jan 2017 19:44:31 +0000 (14:44 -0500)]
da may be NULL
Alan T. DeKok [Mon, 23 Jan 2017 18:52:32 +0000 (13:52 -0500)]
Remove "Vendor-%d-Attr-%d" and "Cisco-Attr-%d" syntax.
v3 never produced those formats, but did parse them. it's better
to just disallow them entirely. That makes parsing much simpler
Arran Cudbard-Bell [Mon, 23 Jan 2017 17:55:55 +0000 (17:55 +0000)]
Fix assert
Arran Cudbard-Bell [Mon, 23 Jan 2017 17:32:19 +0000 (17:32 +0000)]
Doxygen doesn't like @fixme
Arran Cudbard-Bell [Mon, 23 Jan 2017 17:32:08 +0000 (17:32 +0000)]
Fix call to value_box_asprint
Arran Cudbard-Bell [Mon, 23 Jan 2017 17:00:07 +0000 (17:00 +0000)]
Add type field to value_box_t
... and modify all the hundres of places which get pair type from vp->da->type, to get it from the boxed value instead.
Add additional validation functions to check for places where the box type isn't set correctly.
References #1883
Closes #1012
Arran Cudbard-Bell [Mon, 23 Jan 2017 16:58:39 +0000 (16:58 +0000)]
Const to the left
Alan T. DeKok [Fri, 20 Jan 2017 23:13:20 +0000 (18:13 -0500)]
Fixed raw VSA test
Alan T. DeKok [Fri, 20 Jan 2017 23:13:08 +0000 (18:13 -0500)]
comment out failing tests until we can fix them
Alan T. DeKok [Fri, 20 Jan 2017 23:04:11 +0000 (18:04 -0500)]
more unknown vendors
Alan T. DeKok [Fri, 20 Jan 2017 23:02:15 +0000 (18:02 -0500)]
more unknown attributes