]> jfr.im git - irc/freenode/syn.git/log
irc/freenode/syn.git
3 years agoAdd IRCCloud cloaking type origin/irccloud-cloaking
Nicole Kleinhoff [Tue, 12 Jan 2021 08:16:13 +0000 (08:16 +0000)] 
Add IRCCloud cloaking type

This cloaking type is entirely specific to the IRCCloud service. It
operates in a fashion similar to to the ident cloaking type, however the
assigned gateway cloak will not include the 'u' or 's', only the
letters "id" and the subscriber ID itself.

Both uid1234 and sid1234 will thus result in a gateway/.../id1234 cloak.

3 years agofacilities.c: add help for ident cloaking type
Nicole Kleinhoff [Tue, 12 Jan 2021 08:33:56 +0000 (08:33 +0000)] 
facilities.c: add help for ident cloaking type

3 years agodon't redundantly process identhosts that match u->host
jesopo [Wed, 28 Oct 2020 14:06:51 +0000 (14:06 +0000)] 
don't redundantly process identhosts that match u->host

3 years agoadd licence info production origin/HEAD origin/production
Stephen Bennett [Thu, 10 Sep 2020 12:49:26 +0000 (12:49 +0000)] 
add licence info

5 years agomasks.c: actually set last_matched on the proper mask(s)
Janik Kleinhoff [Fri, 19 Apr 2019 18:59:17 +0000 (18:59 +0000)] 
masks.c: actually set last_matched on the proper mask(s)

Fixes only the last mask processed being updated (whether or not it
actually matched), instead updating all masks that do match (reported or
not).

We also no longer stop processing when handling an exempt mask since
otherwise we'd update last_matched for all *previous* matching masks,
which would be quite counterintuitive.

5 years agomake override_unaff actually work properly
Janik Kleinhoff [Thu, 14 Mar 2019 00:05:37 +0000 (00:05 +0000)] 
make override_unaff actually work properly

alternative commit message: "i'm really dumb 2: electric boogaloo"

5 years agoget_random_host_part: hash uid!user@host, rather than just uid
Janik Kleinhoff [Mon, 4 Mar 2019 13:08:42 +0000 (13:08 +0000)] 
get_random_host_part: hash uid!user@host, rather than just uid

This should significantly reduce the chance for the same x-part to be
reused after an ircd restart (unless, of course, the user is the same).

Alternatively, we could hash just user@host, but then it would no longer
be "random"-type cloaking, plus there's already ident-type cloaking.

5 years agofix "facilty (null) requires hexip"
Janik Kleinhoff [Tue, 19 Feb 2019 19:49:36 +0000 (19:49 +0000)] 
fix "facilty (null) requires hexip"

5 years agoThis is atheme 7.2, not 7.3 yet
Janik Kleinhoff [Tue, 19 Feb 2019 19:13:10 +0000 (19:13 +0000)] 
This is atheme 7.2, not 7.3 yet

5 years agofacilities.c: add override_unaff setting
Janik Kleinhoff [Tue, 19 Feb 2019 02:38:25 +0000 (02:38 +0000)] 
facilities.c: add override_unaff setting

5 years agoMake "random" cloaking deterministic
Janik Kleinhoff [Tue, 19 Feb 2019 01:50:26 +0000 (01:50 +0000)] 
Make "random" cloaking deterministic

This uses SipHash and adds a prf_key config option.
auxiliary/siphash.c is the unmodified SipHash reference implementation,
taken from https://github.com/veorq/SipHash

Set syn::prf_key to a string of 32 random hex digits in the
configuration to use this feature.

5 years agoAllow util.c to include syn.h
Janik Kleinhoff [Tue, 19 Feb 2019 01:46:56 +0000 (01:46 +0000)] 
Allow util.c to include syn.h

5 years agogateways.c: rip out unused reverse lookup code
Janik Kleinhoff [Wed, 13 Feb 2019 04:33:59 +0000 (04:33 +0000)] 
gateways.c: rip out unused reverse lookup code

5 years agokline.c: fix memory leak, improve debug logging
Janik Kleinhoff [Mon, 11 Feb 2019 22:05:57 +0000 (22:05 +0000)] 
kline.c: fix memory leak, improve debug logging

5 years agokline.c: allow unklining global klines
Janik Kleinhoff [Mon, 11 Feb 2019 20:58:07 +0000 (20:58 +0000)] 
kline.c: allow unklining global klines

On a BAN message, a duration of 0 means an unban.

5 years agofacilities.c: store a cloak for all facility users
Janik Kleinhoff [Fri, 25 Jan 2019 00:25:22 +0000 (00:25 +0000)] 
facilities.c: store a cloak for all facility users

This avoids failing to apply a gateway cloak on users who (via SASL)
connected with a project cloak, then had it removed (or changed to
unaffiliated) mid-session.

5 years agofacilities.c: properly restore session cloaks
Janik Kleinhoff [Thu, 24 Jan 2019 23:31:34 +0000 (23:31 +0000)] 
facilities.c: properly restore session cloaks

0124 232639 -!- test (test) [ilbelkyr@gateway/test/x-wvgwoklegibimcqp] has joined #services

Set a project cloak:
0124 233021 -!- test [ilbelkyr@gateway/test/x-wvgwoklegibimcqp] has quit [Changing host]
0124 233021 -!- test (test) [ilbelkyr@project/example] has joined #services

Remove cloak:
0124 233032 -!- test [ilbelkyr@project/example] has quit [Changing host]
0124 233032 -!- test (test) [ilbelkyr@gateway/test/session] has joined #services
0124 233032 -!- test [ilbelkyr@gateway/test/session] has quit [Changing host]
0124 233032 -!- test (test) [ilbelkyr@gateway/test/x-wvgwoklegibimcqp] has joined #services

This is still not perfect but a lot better than the previous behaviour,
where the project cloak would be restored instead.

We store the computed facility cloak in per-user metadata; user metadata
is not actually written to disk, unlike myuser or mychan metadata. This
is necessary due to the "random" cloaking type.

5 years agofacilities.c: add "account" cloaking type
Janik Kleinhoff [Thu, 24 Jan 2019 22:15:45 +0000 (22:15 +0000)] 
facilities.c: add "account" cloaking type

This is designed to work with sasl_usercloak-style auth spoofs,
replacing the tor-sasl and PIA special cases. Fixes #4.

Note that you must set account cloaking on the Tor and PIA facilities to
make sure gateway cloaks will be properly enforced.

5 years agoMerge branch 'ident-cloaking'
Janik Kleinhoff [Thu, 24 Jan 2019 22:01:02 +0000 (22:01 +0000)] 
Merge branch 'ident-cloaking'

5 years agohelp/addmask: give example of non-/ delimiters
Janik Kleinhoff [Thu, 24 Jan 2019 21:54:43 +0000 (21:54 +0000)] 
help/addmask: give example of non-/ delimiters

Being able to use e.g. # is quite useful if your mask includes /
characters for whatever reason, so mention that it's possible.

5 years agomasks.c: add last matched timestamp for LISTMASK
Janik Kleinhoff [Thu, 24 Jan 2019 21:51:53 +0000 (21:51 +0000)] 
masks.c: add last matched timestamp for LISTMASK

Fixes #5.

5 years agomasks.c: add help files origin/devel
Janik Kleinhoff [Wed, 23 Jan 2019 05:29:58 +0000 (05:29 +0000)] 
masks.c: add help files

5 years agohelp: add FACILITY SHOW help file
Janik Kleinhoff [Wed, 23 Jan 2019 05:01:18 +0000 (05:01 +0000)] 
help: add FACILITY SHOW help file

5 years agofacilities.c: add subcommand help
Janik Kleinhoff [Wed, 23 Jan 2019 04:56:36 +0000 (04:56 +0000)] 
facilities.c: add subcommand help

This is a mess because (1) it'd be much nicer to read this from a file
but we can't do that via the Atheme help system if we also need to
handle subcommands, and (2) syn's sourceinfos have a
command_success_nodata that doesn't split on \n

At least "/msg syn help facility add" now does what it should.

5 years agomasks.c: indicate /i regex flag in log output and LISTMASK
Janik Kleinhoff [Sun, 20 Jan 2019 07:43:03 +0000 (07:43 +0000)] 
masks.c: indicate /i regex flag in log output and LISTMASK

Fixes #8.

5 years agokline.c: handle CIDR klines
Janik Kleinhoff [Sun, 20 Jan 2019 05:59:05 +0000 (05:59 +0000)] 
kline.c: handle CIDR klines

We could probably use a radix tree to hold CIDR k-lines and match much
more efficiently, but this will do fine for now.

Fixes #1.

5 years agofacilities.c: use patricias instead of dictionaries
Janik Kleinhoff [Fri, 18 Jan 2019 04:09:33 +0000 (04:09 +0000)] 
facilities.c: use patricias instead of dictionaries

Per #2 I have no idea how mowgli dictionaries work, if at all, but
atheme has moved to patricias anyway without issue, and this change
makes things actually work.

I'm not sure there's anything else using the current mowgli dictionary
implementation. Atheme certainly doesn't, nor do libmowgli internals.

6 years agofacilities: Add "ident" cloaking type origin/ident-cloaking
Janik Kleinhoff [Sun, 18 Mar 2018 01:53:27 +0000 (01:53 +0000)] 
facilities: Add "ident" cloaking type

Fixes #6.

<@syn> FACILITY ADD gateway/web/irccloud.com by ilbelkitty (ilbelkyr)
<@syn> FACILITY SET cloaking->ident for gateway/web/irccloud.com by ilbelkitty (ilbelkyr)

Test cases:

Basic ident
uid123456 -> gateway/web/irccloud.com/uid123456

Maximum-length ident is handled properly
maxlen1234 -> gateway/web/irccloud.com/maxlen1234

Invalid/disallowed cloak characters in ident
(test twice to ensure the x-part remains stable)
test_me -> gateway/web/irccloud.com/testme/x-30596902

Client-sent USER behaves as ident
~test_me -> gateway/web/irccloud.com/testme/x-30596902

Starting digits removed to avoid CIDR lookalikes
01test23 -> gateway/web/irccloud.com/test23/x-30027077

Different idents generate different x-parts
89test23 -> gateway/web/irccloud.com/test23/x-97047605

Entire ident is invalid/disallowed in hostname as-is
1000 -> gateway/web/irccloud.com//x-31162250

The last of these is probably the most concerning, but similar is
already possible with Tor/PIA cloaking right now as far as I can tell.

6 years agofacilities.c: actually display help for FACILITY
Janik Kleinhoff [Fri, 16 Mar 2018 23:00:19 +0000 (23:00 +0000)] 
facilities.c: actually display help for FACILITY

This doesn't handle the subcommand help yet and does not add any help
for those commands that currently lack a help file; cf #3.

6 years agomodules: use newer module header style
Janik Kleinhoff [Fri, 16 Mar 2018 22:34:49 +0000 (22:34 +0000)] 
modules: use newer module header style

This is compatible with both recent and older Atheme.

6 years agosourceinfo vtable: use designated initializers
Janik Kleinhoff [Fri, 16 Mar 2018 22:33:13 +0000 (22:33 +0000)] 
sourceinfo vtable: use designated initializers

This avoids breaking whenever struct sourceinfo_vtable gains fields.

6 years agosyn.h: s/inline/static inline/
Janik Kleinhoff [Fri, 16 Mar 2018 22:30:37 +0000 (22:30 +0000)] 
syn.h: s/inline/static inline/

Refer to the C standard for details, but essentially a bare "inline"
will not do what we want and/or may break at link time.

8 years agoi'm really dumb
Stephen Bennett [Tue, 10 May 2016 18:41:53 +0000 (18:41 +0000)] 
i'm really dumb

8 years agoi'm dumb
Stephen Bennett [Tue, 10 May 2016 18:39:38 +0000 (18:39 +0000)] 
i'm dumb

8 years agoif we expect a hex ip but can't decode one, kill instead of falling back to random
Stephen Bennett [Tue, 10 May 2016 16:11:46 +0000 (16:11 +0000)] 
if we expect a hex ip but can't decode one, kill instead of falling back to random

9 years agoextend tor-sasl special case to cover privateinternetaccess
Stephen Bennett [Fri, 2 Jan 2015 12:33:31 +0000 (12:33 +0000)] 
extend tor-sasl special case to cover privateinternetaccess

9 years agodon't use -Werror all the time
Stephen Bennett [Sun, 29 Jun 2014 20:47:51 +0000 (20:47 +0000)] 
don't use -Werror all the time

9 years agodon't crash when victim->ip is null, which it can be now
Stephen Bennett [Sun, 29 Jun 2014 20:23:21 +0000 (20:23 +0000)] 
don't crash when victim->ip is null, which it can be now

10 years agodisable reverse lookups for now until i can figure out why they crash
Stephen Bennett [Tue, 18 Jun 2013 10:41:09 +0000 (10:41 +0000)] 
disable reverse lookups for now until i can figure out why they crash

10 years agohandle non-encap CHGHOST properly
Stephen Bennett [Mon, 10 Jun 2013 14:52:00 +0000 (14:52 +0000)] 
handle non-encap CHGHOST properly

10 years agoremove no-longer-necessary atheme patch
Stephen Bennett [Sun, 9 Jun 2013 14:08:56 +0000 (14:08 +0000)] 
remove no-longer-necessary atheme patch

10 years agotry to fix case where a user quits while a reverse lookup is pending
Stephen Bennett [Sun, 9 Jun 2013 14:08:33 +0000 (14:08 +0000)] 
try to fix case where a user quits while a reverse lookup is pending

10 years agofix make install on silly shells
Stephen Bennett [Sun, 9 Jun 2013 13:37:40 +0000 (13:37 +0000)] 
fix make install on silly shells

10 years agoMake reverse dns lookups for gateway users asynchronous
Stephen Bennett [Sun, 9 Jun 2013 13:27:29 +0000 (13:27 +0000)] 
Make reverse dns lookups for gateway users asynchronous

10 years agoIf a gateway user has no hostname in gecos, do a reverse lookup on their decoded...
Stephen Bennett [Thu, 6 Jun 2013 20:36:39 +0000 (20:36 +0000)] 
If a gateway user has no hostname in gecos, do a reverse lookup on their decoded ident to look for hostname klines

10 years agofix new klines
Stephen Bennett [Thu, 6 Jun 2013 18:37:50 +0000 (18:37 +0000)] 
fix new klines

10 years agoattempt 2 at supporting propagated bans. also removes the need to patch atheme.
Stephen Bennett [Thu, 6 Jun 2013 18:15:59 +0000 (18:15 +0000)] 
attempt 2 at supporting propagated bans. also removes the need to patch atheme.

10 years agopropagated kline support try 1
Stephen Bennett [Thu, 6 Jun 2013 17:53:01 +0000 (17:53 +0000)] 
propagated kline support try 1

10 years agomake it work on atheme-7
Stephen Bennett [Thu, 6 Jun 2013 17:33:27 +0000 (17:33 +0000)] 
make it work on atheme-7

10 years agoand update atheme patch for 7
Stephen Bennett [Thu, 6 Jun 2013 15:50:28 +0000 (15:50 +0000)] 
and update atheme patch for 7

10 years agoMake it build against atheme-7
Stephen Bennett [Thu, 6 Jun 2013 13:47:05 +0000 (13:47 +0000)] 
Make it build against atheme-7

13 years agoReject an incoming host change if it resets a gateway user's host to his real host...
Stephen Bennett [Fri, 8 Oct 2010 22:24:50 +0000 (23:24 +0100)] 
Reject an incoming host change if it resets a gateway user's host to his real host. This happens when an unaffiliated cloak is removed on an account, and nickserv resets all logged in sessions to their real hostnames, overwriting any gateway session cloak that might be present.

14 years agoDrop the check to avoid sending out duplicate klines -- seven doesn't leak memory...
Stephen Bennett [Thu, 1 Apr 2010 07:48:48 +0000 (08:48 +0100)] 
Drop the check to avoid sending out duplicate klines -- seven doesn't leak memory the way hyperion did, and it solves the case where one server didn't pick up a kline properly

14 years agoDrop *@* klines.
Stephen Bennett [Fri, 19 Mar 2010 15:19:09 +0000 (15:19 +0000)] 
Drop *@* klines.

14 years agoDon't send out connection snotes every time a kline is added.
Stephen Bennett [Thu, 4 Feb 2010 22:58:14 +0000 (22:58 +0000)] 
Don't send out connection snotes every time a kline is added.

14 years agoSend out fake farconnect snotes for gateway users, that contain the public IP address...
Stephen Bennett [Thu, 4 Feb 2010 22:07:02 +0000 (22:07 +0000)] 
Send out fake farconnect snotes for gateway users, that contain the public IP address so that bopm can scan it

14 years agoAdd special case for tor-sasl users: reset their visible hostname to the original...
Stephen Bennett [Wed, 3 Feb 2010 00:24:39 +0000 (00:24 +0000)] 
Add special case for tor-sasl users: reset their visible hostname to the original (account name substituted) host if SASL auth set an unaffiliated cloak. This effectively brings tor-sasl into line with other facilities, where project cloak overrides gateway cloak overrides unaffiliated cloak.

14 years agoRemove hyperion protocol module changes from the atheme host-change-hook patch, as...
Stephen Bennett [Wed, 3 Feb 2010 00:22:58 +0000 (00:22 +0000)] 
Remove hyperion protocol module changes from the atheme host-change-hook patch, as said module has been dropped

14 years agoDon't search for a kline on a null string
Stephen Bennett [Sat, 30 Jan 2010 20:40:15 +0000 (20:40 +0000)] 
Don't search for a kline on a null string

14 years agoMake syn use the same Makefile.config setup that atheme+fn does.
Stephen Bennett [Tue, 22 Dec 2009 00:00:17 +0000 (00:00 +0000)] 
Make syn use the same Makefile.config setup that atheme+fn does.

14 years agoFix compiler warning/error
Stephen Bennett [Sat, 30 Jan 2010 20:30:29 +0000 (20:30 +0000)] 
Fix compiler warning/error

14 years agoAllow debug output to be sent to a different channel
Stephen Bennett [Mon, 7 Dec 2009 10:21:20 +0000 (10:21 +0000)] 
Allow debug output to be sent to a different channel

14 years agoBe more careful about checkout incoming host changes on seven. An earlier hack for...
Stephen Bennett [Sun, 1 Nov 2009 23:46:08 +0000 (23:46 +0000)] 
Be more careful about checkout incoming host changes on seven. An earlier hack for hyperion broke this.

14 years agoDon't use sscanf.
Stephen Bennett [Sun, 1 Nov 2009 23:45:10 +0000 (23:45 +0000)] 
Don't use sscanf.

14 years agoFix logic for wildcard klines
Stephen Bennett [Sun, 1 Nov 2009 20:01:03 +0000 (20:01 +0000)] 
Fix logic for wildcard klines

14 years agoBe more paranoid about decoding hex ips
Stephen Bennett [Sun, 1 Nov 2009 20:00:40 +0000 (20:00 +0000)] 
Be more paranoid about decoding hex ips

14 years agoFix stupid bug
Stephen Bennett [Fri, 30 Oct 2009 20:25:55 +0000 (20:25 +0000)] 
Fix stupid bug

14 years agoStore non-wildcard klines in a patricia tree instead of a linked list. Should perform...
Stephen Bennett [Thu, 29 Oct 2009 18:49:47 +0000 (18:49 +0000)] 
Store non-wildcard klines in a patricia tree instead of a linked list. Should perform rather better.

14 years agoIf we fail to decode a gateway IP for a user, don't keep trying on every new kline etc
Stephen Bennett [Wed, 2 Sep 2009 09:26:44 +0000 (10:26 +0100)] 
If we fail to decode a gateway IP for a user, don't keep trying on every new kline etc

14 years agoFix stupid typo
Stephen Bennett [Sat, 22 Aug 2009 21:20:30 +0000 (22:20 +0100)] 
Fix stupid typo

14 years agoIf we abort gateway checking because a user caused a kline to be added, then skip...
Stephen Bennett [Sat, 22 Aug 2009 21:19:23 +0000 (22:19 +0100)] 
If we abort gateway checking because a user caused a kline to be added, then skip all further checks on that (now probably non-existent) user as well.

14 years agoDon't segfault when a user connects through a gateway from a tor node not already...
Stephen Bennett [Thu, 13 Aug 2009 20:24:08 +0000 (21:24 +0100)] 
Don't segfault when a user connects through a gateway from a tor node not already klined

14 years agoUpdate for the new user_nickchange hook, and the associated hook_user_data_t -> hook_...
Stephen Bennett [Thu, 13 Aug 2009 11:30:33 +0000 (12:30 +0100)] 
Update for the new user_nickchange hook, and the associated hook_user_data_t -> hook_user_nick_t rename

14 years agoIn new-user hooks, blank out the user pointer if we kill it, and don't attempt to...
Stephen Bennett [Wed, 12 Aug 2009 21:12:18 +0000 (22:12 +0100)] 
In new-user hooks, blank out the user pointer if we kill it, and don't attempt to do anything in subsequent hooks.

14 years agoUpdate for atheme hook changes
Stephen Bennett [Wed, 12 Aug 2009 20:56:21 +0000 (21:56 +0100)] 
Update for atheme hook changes

14 years agoSave join rate data more often
Stephen Bennett [Fri, 7 Aug 2009 13:29:23 +0000 (14:29 +0100)] 
Save join rate data more often

14 years agoSave the mask database more often
Stephen Bennett [Thu, 6 Aug 2009 18:29:42 +0000 (19:29 +0100)] 
Save the mask database more often

14 years agoFix unloading facilities module
Stephen Bennett [Sun, 2 Aug 2009 21:37:28 +0000 (22:37 +0100)] 
Fix unloading facilities module

14 years agoAdd the nasty hack to override unaffiliated cloaks with gateway cloaks, and to work...
Stephen Bennett [Sun, 2 Aug 2009 20:44:50 +0000 (21:44 +0100)] 
Add the nasty hack to override unaffiliated cloaks with gateway cloaks, and to work around the race condition when services and syn both rehost a user on connect

14 years agoAdd an atheme patch to hook into incoming host changes
Stephen Bennett [Sun, 2 Aug 2009 18:11:26 +0000 (19:11 +0100)] 
Add an atheme patch to hook into incoming host changes

14 years agoDon't reveal the oper-reason part of a kline comment when killing a user instead...
Stephen Bennett [Thu, 30 Jul 2009 20:27:55 +0000 (21:27 +0100)] 
Don't reveal the oper-reason part of a kline comment when killing a user instead of klining

14 years agoBe a bit smarter with installing
Stephen Bennett [Thu, 30 Jul 2009 19:52:08 +0000 (20:52 +0100)] 
Be a bit smarter with installing

14 years agoSave facilities db when changing a setting, as apparently it's not done on shutdown
Stephen Bennett [Thu, 30 Jul 2009 19:48:20 +0000 (20:48 +0100)] 
Save facilities db when changing a setting, as apparently it's not done on shutdown

14 years agoCheck explicitly for null IP field, to avoid triggering assertions in libmowgli
Stephen Bennett [Thu, 30 Jul 2009 08:41:54 +0000 (09:41 +0100)] 
Check explicitly for null IP field, to avoid triggering assertions in libmowgli

14 years agodon't spam while bursting
Stephen Bennett [Wed, 29 Jul 2009 12:40:36 +0000 (13:40 +0100)] 
don't spam while bursting

14 years agoBe less silly
Stephen Bennett [Wed, 29 Jul 2009 12:34:45 +0000 (13:34 +0100)] 
Be less silly

14 years agoSpam when allowing gateway users
Stephen Bennett [Wed, 29 Jul 2009 12:27:04 +0000 (13:27 +0100)] 
Spam when allowing gateway users

14 years agoAdd a new function syn_report2 which accepts a verbosity level
Stephen Bennett [Wed, 29 Jul 2009 12:26:38 +0000 (13:26 +0100)] 
Add a new function syn_report2 which accepts a verbosity level

14 years agoFix typo, and remember to delete conf items on unload
Stephen Bennett [Wed, 29 Jul 2009 12:05:12 +0000 (13:05 +0100)] 
Fix typo, and remember to delete conf items on unload

14 years agoCheck that setmask has enough arguments
Stephen Bennett [Wed, 29 Jul 2009 11:35:15 +0000 (12:35 +0100)] 
Check that setmask has enough arguments

14 years agoReset a channel's join-rate counter when changing its warning rate
Stephen Bennett [Wed, 29 Jul 2009 10:59:00 +0000 (11:59 +0100)] 
Reset a channel's join-rate counter when changing its warning rate

14 years agoDon't check facility throttles during netburst
Stephen Bennett [Fri, 29 May 2009 17:42:43 +0000 (18:42 +0100)] 
Don't check facility throttles during netburst

15 years agoCheck masks on nick changes as well as new connections
Stephen Bennett [Fri, 15 May 2009 22:49:20 +0000 (23:49 +0100)] 
Check masks on nick changes as well as new connections

15 years agoBe sure to check gateway users against tor nodes, even when no user has connected...
Stephen Bennett [Fri, 8 May 2009 22:32:10 +0000 (23:32 +0100)] 
Be sure to check gateway users against tor nodes, even when no user has connected directly from that node

15 years agoTypo
Stephen Bennett [Fri, 8 May 2009 21:15:44 +0000 (22:15 +0100)] 
Typo

15 years agoDon't segfault when not given enough arguments
Stephen Bennett [Wed, 6 May 2009 18:41:51 +0000 (19:41 +0100)] 
Don't segfault when not given enough arguments

15 years agoDetect duplicate K:lines
Stephen Bennett [Wed, 6 May 2009 18:41:39 +0000 (19:41 +0100)] 
Detect duplicate K:lines

15 years agoSave and load the masks list to disk
Stephen Bennett [Sun, 3 May 2009 22:47:03 +0000 (23:47 +0100)] 
Save and load the masks list to disk

15 years agoFix display of set/expiry times
Stephen Bennett [Sun, 3 May 2009 22:46:46 +0000 (23:46 +0100)] 
Fix display of set/expiry times

15 years agoStore setter and set time for masks
Stephen Bennett [Sun, 3 May 2009 22:06:41 +0000 (23:06 +0100)] 
Store setter and set time for masks