]>
git.ipfire.org Git - thirdparty/freeradius-server.git/log
Arran Cudbard-Bell [Thu, 11 Apr 2024 17:00:03 +0000 (11:00 -0600)]
Print sizes with suffixes
Alan T. DeKok [Tue, 9 Apr 2024 20:18:34 +0000 (16:18 -0400)]
remove unused variable
Alan T. DeKok [Tue, 9 Apr 2024 19:53:41 +0000 (15:53 -0400)]
ensure that each Access-Request packet is unique
Alan T. DeKok [Tue, 9 Apr 2024 19:43:33 +0000 (15:43 -0400)]
flush items after write becomes unblocked
Alan T. DeKok [Tue, 9 Apr 2024 15:50:48 +0000 (11:50 -0400)]
simplify and handle more corner cases
remove the difference between write_partial and write_cancelled.
it turns out they're both largely the same.
Which then simplifies the cancel() function, as it just sets
item->cancelled, which is much better.
ensure that if we're writing a partial packet, that the timers
are removed. Because if writes are blocked, there's no reason
to run the retries to write more packets.
There's still an @todo to flush the outbound retry list when
write becomes unblocked.
Alan T. DeKok [Tue, 9 Apr 2024 14:51:49 +0000 (10:51 -0400)]
handle the case of WOULD_BLOCK
Nick Porter [Tue, 9 Apr 2024 13:12:01 +0000 (14:12 +0100)]
Need to track what log destination was cloned
log_dst_by_name() returns a destination from the list of destinations
parsed from the config, which then gets cloned for appending to a
request's list.
Therefore when matching in the request's list we need to know what the
original destination was that was cloned.
Nick Porter [Tue, 9 Apr 2024 13:07:27 +0000 (14:07 +0100)]
Set request debug level when amending list of log destinations
Nick Porter [Mon, 8 Apr 2024 13:23:12 +0000 (14:23 +0100)]
Test SQL cache_groups option
Nick Porter [Mon, 8 Apr 2024 13:18:23 +0000 (14:18 +0100)]
Use distinct prioities in test data to ensure consistent behaviour
Nick Porter [Mon, 8 Apr 2024 13:07:54 +0000 (14:07 +0100)]
Correct handling of cache_groups option
Nick Porter [Wed, 3 Apr 2024 09:49:32 +0000 (10:49 +0100)]
Add testing of Fall-Through to SQL groups
Nick Porter [Fri, 29 Mar 2024 12:16:54 +0000 (12:16 +0000)]
Remove unused attribute refs
Alan T. DeKok [Sat, 6 Apr 2024 15:16:38 +0000 (11:16 -0400)]
update as per documentation. Fixes #5312
Manual changes as a result of looking at the PR, and updating
it based on the v4 format
Alan T. DeKok [Sat, 6 Apr 2024 15:10:38 +0000 (11:10 -0400)]
allow relative references in "clone="
remove code which allowed clone to reference other protocols.
The "reference protocol" code was changed during the refactoring
to support
ATTRIBUTE ... group ref=..DHCPv4
which now makes this code invalid. So we fix up this code to work,
and (at least temporarily) remove the ability to reference other
protocols.
Now that the framework supports relative references, it should be
fairly simply to re-add protocol references as a separate commit.
Nick Porter [Fri, 5 Apr 2024 11:22:18 +0000 (12:22 +0100)]
Correctly handle open() return value
Nick Porter [Fri, 5 Apr 2024 09:51:10 +0000 (10:51 +0100)]
No destination means disable
Nick Porter [Fri, 5 Apr 2024 09:49:55 +0000 (10:49 +0100)]
Use correct free function
github-actions[bot] [Fri, 5 Apr 2024 09:35:08 +0000 (09:35 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/radius.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:55 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/vmps.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:52 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/tacacs.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:49 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dns.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:46 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv6.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:43 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/dhcpv4.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:37 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/bfd.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:34 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/util.tar
github-actions[bot] [Fri, 5 Apr 2024 09:34:31 +0000 (09:34 +0000)]
Scheduled fuzzing: Update src/tests/fuzzer-corpus/tftp.tar
Nick Porter [Thu, 4 Apr 2024 17:32:15 +0000 (18:32 +0100)]
Return the fr_log_t not the fr_log_track_t
Nick Porter [Thu, 4 Apr 2024 17:31:13 +0000 (18:31 +0100)]
Fix up args for %log.destination()
So they're individual value boxes rather than groups
Nick Porter [Thu, 4 Apr 2024 13:58:53 +0000 (14:58 +0100)]
dhcp is no-longer in a separate deb package
Alan T. DeKok [Wed, 3 Apr 2024 23:31:04 +0000 (19:31 -0400)]
randomize the ID array.
There's no reason to let observers know how many packets we've sent.
Alan T. DeKok [Wed, 3 Apr 2024 23:10:15 +0000 (19:10 -0400)]
catch more corner cases and cleanups
Nick Porter [Thu, 4 Apr 2024 09:27:35 +0000 (10:27 +0100)]
Run the correct section
Alan T. DeKok [Tue, 2 Apr 2024 21:10:24 +0000 (17:10 -0400)]
make '-c' work again
we can do do
RADCLIENT="radclient-ng" make test.radclient
and the tests pass. So that's something.
Alan T. DeKok [Tue, 2 Apr 2024 21:01:48 +0000 (17:01 -0400)]
set ifindex, allow -i to work, print filename
Alan T. DeKok [Tue, 2 Apr 2024 20:56:30 +0000 (16:56 -0400)]
add fr_ipaddr_get_scope_id() function and use it in FD bios
Alan T. DeKok [Tue, 2 Apr 2024 20:12:07 +0000 (16:12 -0400)]
rename fr_packet_log() to fr_radius_packet_log()
Alan T. DeKok [Tue, 2 Apr 2024 20:10:34 +0000 (16:10 -0400)]
add API to force a client ID
largely used only for radclient
Alan T. DeKok [Tue, 2 Apr 2024 19:01:58 +0000 (15:01 -0400)]
make -r and -t work again
Alan T. DeKok [Tue, 2 Apr 2024 18:44:07 +0000 (14:44 -0400)]
make -p work again
along with a few other minor cleanups, and an initial stab at
making -i work again
Alan T. DeKok [Tue, 2 Apr 2024 16:37:06 +0000 (12:37 -0400)]
WOULD_BLOCK is not a fatal error when opening connected sockets
Alan T. DeKok [Tue, 2 Apr 2024 16:36:57 +0000 (12:36 -0400)]
comment out unused code
Alan T. DeKok [Tue, 2 Apr 2024 13:44:04 +0000 (09:44 -0400)]
re-enable filters and stats
Alan T. DeKok [Tue, 2 Apr 2024 12:55:33 +0000 (08:55 -0400)]
don't use id_ctx after it's been deallocated
Alan T. DeKok [Mon, 1 Apr 2024 16:34:35 +0000 (12:34 -0400)]
print out times, too
Alan T. DeKok [Mon, 1 Apr 2024 13:07:22 +0000 (09:07 -0400)]
add assertions and fix return code check for reset timer
Alan T. DeKok [Sun, 31 Mar 2024 23:20:54 +0000 (19:20 -0400)]
remove redundant code, and use correct source port
Arran Cudbard-Bell [Tue, 2 Apr 2024 16:36:27 +0000 (11:36 -0500)]
...and lib64
Arran Cudbard-Bell [Tue, 2 Apr 2024 16:28:34 +0000 (11:28 -0500)]
Add alt openldap dir
Arran Cudbard-Bell [Tue, 2 Apr 2024 16:20:22 +0000 (11:20 -0500)]
Needed for linux
Arran Cudbard-Bell [Tue, 2 Apr 2024 15:52:09 +0000 (10:52 -0500)]
Add rlm_rest cancellation test
Arran Cudbard-Bell [Tue, 2 Apr 2024 15:51:21 +0000 (10:51 -0500)]
Simplify dealing with indentations
Nick Porter [Mon, 1 Apr 2024 17:33:20 +0000 (18:33 +0100)]
rlm_sql_freetds: handle returned NULL column values
These don't update the results buffer - so zero it out during
allocation.
Alan T. DeKok [Sun, 31 Mar 2024 21:40:50 +0000 (17:40 -0400)]
cleanups and documentation
Alan T. DeKok [Sun, 31 Mar 2024 21:03:35 +0000 (17:03 -0400)]
foo - 2
Alan T. DeKok [Sun, 31 Mar 2024 15:57:47 +0000 (11:57 -0400)]
return bio errors to the caller
Alan T. DeKok [Sun, 31 Mar 2024 15:57:26 +0000 (11:57 -0400)]
add API documentation
Alan T. DeKok [Sun, 31 Mar 2024 14:53:20 +0000 (10:53 -0400)]
add "ef_out" to fr_event_fd_insert()
Alan T. DeKok [Sun, 31 Mar 2024 01:42:49 +0000 (21:42 -0400)]
add client_bio_info and use it
Alan T. DeKok [Sun, 31 Mar 2024 01:36:42 +0000 (21:36 -0400)]
API to get string descriptions of errors
Alan T. DeKok [Sun, 31 Mar 2024 00:30:05 +0000 (20:30 -0400)]
can't call shutdown() from the bottom bio
Alan T. DeKok [Sun, 31 Mar 2024 00:14:17 +0000 (20:14 -0400)]
update comments and check for async with connect
Alan T. DeKok [Sat, 30 Mar 2024 22:27:36 +0000 (18:27 -0400)]
comment out unused function
Alan T. DeKok [Sat, 30 Mar 2024 22:26:27 +0000 (18:26 -0400)]
only build radclient-ng if we're testing the bio code
Alan T. DeKok [Sat, 30 Mar 2024 21:07:45 +0000 (17:07 -0400)]
typo
Alan T. DeKok [Sat, 30 Mar 2024 21:06:39 +0000 (17:06 -0400)]
RADIUS client bio can now send and receive a packet
if anything happens radclient-ng asserts, but the basic functionality
is now there and tested.
Alan T. DeKok [Sat, 30 Mar 2024 21:05:44 +0000 (17:05 -0400)]
minor fixes, and allow for logging
Alan T. DeKok [Sat, 30 Mar 2024 20:44:41 +0000 (16:44 -0400)]
minor fixes
the uctx for the timer is the bio, not the item.
call rewrite() with our bio, not the next one
check that things are initialized
clean up all outstanding timers when there's nothing left to do
Alan T. DeKok [Sat, 30 Mar 2024 20:31:39 +0000 (16:31 -0400)]
there can't be a previous bio
Alan T. DeKok [Sat, 30 Mar 2024 19:58:45 +0000 (15:58 -0400)]
use request_ctx for reply and reply_pairs
we really need a better way to allocate these. Perhaps the caller
should always just pre-allocate the reply?
Alan T. DeKok [Sat, 30 Mar 2024 19:58:31 +0000 (15:58 -0400)]
add and use outstanding API
Alan T. DeKok [Sat, 30 Mar 2024 17:52:58 +0000 (13:52 -0400)]
add "cancel packet" API
Alan T. DeKok [Sat, 30 Mar 2024 15:42:27 +0000 (11:42 -0400)]
we don't need or use a "cancelled" entry
Alan T. DeKok [Sat, 30 Mar 2024 14:56:58 +0000 (10:56 -0400)]
add error and set outobund bio to the top-level retry
Alan T. DeKok [Sat, 30 Mar 2024 14:56:14 +0000 (10:56 -0400)]
verify needs a read_size
Alan T. DeKok [Sat, 30 Mar 2024 14:55:48 +0000 (10:55 -0400)]
return errors
Alan T. DeKok [Sat, 30 Mar 2024 14:48:06 +0000 (10:48 -0400)]
return allocated ptr
Alan T. DeKok [Sat, 30 Mar 2024 14:09:49 +0000 (10:09 -0400)]
minor tweaks
Alan T. DeKok [Sat, 30 Mar 2024 13:51:55 +0000 (09:51 -0400)]
update format arg
Arran Cudbard-Bell [Fri, 29 Mar 2024 23:50:55 +0000 (19:50 -0400)]
Use the same scheme as the other runtime features
Arran Cudbard-Bell [Fri, 29 Mar 2024 23:45:18 +0000 (19:45 -0400)]
Indicate if we were built with gperftools support and whether we're being profiled
Nick Porter [Fri, 29 Mar 2024 10:21:47 +0000 (10:21 +0000)]
Add MSCHAPv2 to tests
Nick Porter [Fri, 29 Mar 2024 10:14:37 +0000 (10:14 +0000)]
Add tests of local mschap password change
Nick Porter [Fri, 29 Mar 2024 08:35:42 +0000 (08:35 +0000)]
Align docs / examples with attribute name in dictionary
Nick Porter [Wed, 13 Mar 2024 20:27:19 +0000 (20:27 +0000)]
Use dummy scripts to test mschap ntlm_auth backend
Nick Porter [Wed, 13 Mar 2024 19:51:46 +0000 (19:51 +0000)]
Tidy up following false test failure
Nick Porter [Wed, 13 Mar 2024 19:50:17 +0000 (19:50 +0000)]
Use call_env to remove xlat_eval from mschap password change
Nick Porter [Wed, 13 Mar 2024 14:49:32 +0000 (14:49 +0000)]
Add parsing of chap_error to mschap auth call_env
Nick Porter [Wed, 13 Mar 2024 12:32:23 +0000 (12:32 +0000)]
Use tmpl to determine attribute containing MS-CHAP-NT-Enc-PW
Nick Porter [Tue, 12 Mar 2024 17:35:47 +0000 (17:35 +0000)]
Automatically append "\n" to string written to child process
Nick Porter [Fri, 29 Mar 2024 12:30:12 +0000 (12:30 +0000)]
Reduce fuzzer timeout as main CI job is finishing first
Lengthy fuzzing is still running as a daily scheduled job
Nick Porter [Fri, 29 Mar 2024 11:13:30 +0000 (11:13 +0000)]
Report correct config items in warning message
Nick Porter [Fri, 29 Mar 2024 11:12:12 +0000 (11:12 +0000)]
sql_set_user is only used in rlm_sql.c
And the username option is always NULL - so remove it
Nick Porter [Thu, 28 Mar 2024 19:15:16 +0000 (19:15 +0000)]
Update %(xlat:) to %xlat() in docs
Nick Porter [Thu, 28 Mar 2024 17:55:27 +0000 (17:55 +0000)]
Update %{xlat:} to %xlat() in comments / docs
Alan T. DeKok [Wed, 27 Mar 2024 12:50:09 +0000 (08:50 -0400)]
remove typo
Alan T. DeKok [Tue, 26 Mar 2024 15:29:47 +0000 (11:29 -0400)]
':' is no longer special in xlats
Alan T. DeKok [Tue, 26 Mar 2024 13:30:08 +0000 (09:30 -0400)]
remove more tmp-*
Alan T. DeKok [Tue, 26 Mar 2024 11:27:32 +0000 (07:27 -0400)]
remove PAIR_ENCODE_SKIPPED. we just return 0 instead.
PAIR_ENCODE_SKIPPED is redundant with 0. It's also negative,
which means it's an error. And therefore any code which wants to
NOT treat it as an error needs to check for it specifically
Alan T. DeKok [Tue, 26 Mar 2024 11:25:18 +0000 (07:25 -0400)]
note @todo for Acct-Delay-Time