]> git.ipfire.org Git - thirdparty/iptables.git/log
thirdparty/iptables.git
18 years agoFix iptables-save not printing -s !0/0 and -d !0/0 as well as ip6tables
Patrick McHardy [Sat, 2 Dec 2006 17:17:33 +0000 (17:17 +0000)] 
Fix iptables-save not printing -s !0/0 and -d !0/0 as well as ip6tables
unnecessarily printing the address. Base on patch by Daniel De Graaf.

18 years agoFix /etc/network usage (Pablo Neira)
Pablo Neira Ayuso [Wed, 29 Nov 2006 13:32:32 +0000 (13:32 +0000)] 
Fix /etc/network usage (Pablo Neira)

http://bugs.debian.org/398082

iptables 1.3.5 and 1.3.6 appear to read /etc/networks, but the
information is lost somewhere with 1.3.6.

 # cat /etc/networks
 foonet 10.0.0.0

 # strace -s 255 -o /tmp/foo iptables -v -A INPUT -s foonet/8 -j
ACCEPT #1.3.5 [1]
 ACCEPT  all opt -- in * out *  10.0.0.0/8  -> 0.0.0.0/0

 # strace -s 255 -o /tmp/bar iptables -v -A INPUT -s foonet/8 -j
ACCEPT #1.3.6 [2]
 iptables v1.3.6: host/network `foonet.0.0.0' not found
 Try `iptables -h' or 'iptables --help' for more information.

1. http://people.debian.org/~ljlane/stuff/strace-iptables-1.3.5.txt
2. http://people.debian.org/~ljlane/stuff/strace-iptables-1.3.6.txt

18 years agoFix -E (rename) in iptables/ip6tables
Krzysztof Piotr Oledzki [Tue, 14 Nov 2006 07:50:54 +0000 (07:50 +0000)] 
Fix -E (rename) in iptables/ip6tables

Remove ununsed CHECK entry in commands_v_options.

It makes -E (rename) working again - generic_opt_check
expects options for RENAME not for CHECK at that table index.

Signed-off-by: Krzysztof Piotr Oledzki <ole@ans.pl>
Signed-off-by: Patrick McHardy <kaber@trash.net>
18 years agoUse /lib/modules/$(uname -r)/build instead of /usr/src/linux as KERNEL_DIR default
Patrick McHardy [Tue, 14 Nov 2006 06:39:24 +0000 (06:39 +0000)] 
Use /lib/modules/$(uname -r)/build instead of /usr/src/linux as KERNEL_DIR default

18 years agoAdd ip6tables support for hashlimit match
Patrick McHardy [Mon, 13 Nov 2006 20:31:42 +0000 (20:31 +0000)] 
Add ip6tables support for hashlimit match

18 years agoAdd iptables-xml tool (Amin Azez <azez@ufomechanic.net>)
Amin Azez [Mon, 13 Nov 2006 20:23:36 +0000 (20:23 +0000)] 
Add iptables-xml tool (Amin Azez <azez@ufomechanic.net>)

18 years agoAdd ip6tables support for sctp match
Patrick McHardy [Mon, 13 Nov 2006 19:38:44 +0000 (19:38 +0000)] 
Add ip6tables support for sctp match

18 years agoload ip_[6]tables.ko just before checking revision support in kernel.
Yasuyuki KOZAKAI [Mon, 13 Nov 2006 05:09:16 +0000 (05:09 +0000)] 
load ip_[6]tables.ko just before checking revision support in kernel.

18 years agochanges IP6T_SO_GET_REVISION_{MATCH,TARGET} to 68,69
Yasuyuki KOZAKAI [Mon, 13 Nov 2006 04:03:26 +0000 (04:03 +0000)] 
changes IP6T_SO_GET_REVISION_{MATCH,TARGET} to 68,69
66 and 67 is conflicted with IPv6 Advanced API in kernel <= 2.6.18.

18 years ago- Add revision support to ip6tables.
Rémi Denis-Courmont [Fri, 20 Oct 2006 12:24:34 +0000 (12:24 +0000)] 
- Add revision support to ip6tables.
- Add support port range match to libip6t_multiport
(R?mi Denis-Courmont <rdenis@simphalempin.com>)

18 years agoFix spelling error
Patrick McHardy [Wed, 11 Oct 2006 07:37:26 +0000 (07:37 +0000)] 
Fix spelling error

18 years agoiptables segfaults when given "" to --log-prefix (Mike Frysinger <vapier@gentoo.org>)
Mike Frysinger [Tue, 10 Oct 2006 06:18:40 +0000 (06:18 +0000)] 
iptables segfaults when given "" to --log-prefix (Mike Frysinger <vapier@gentoo.org>)

Bugzilla #516

18 years agoAdd endian annotation types to fix compilation for kernels > 2.6.18
Patrick McHardy [Mon, 9 Oct 2006 18:00:11 +0000 (18:00 +0000)] 
Add endian annotation types to fix compilation for kernels > 2.6.18

18 years agoVersion number was not bumped in Makefile in svn v1.3.6
Joszef Kadlecsik [Mon, 9 Oct 2006 12:09:42 +0000 (12:09 +0000)] 
Version number was not bumped in Makefile in svn

18 years agoUse correct types at error reporting (patch sent by H. Nakano)
Joszef Kadlecsik [Fri, 6 Oct 2006 08:30:20 +0000 (08:30 +0000)] 
Use correct types at error reporting (patch sent by H. Nakano)

18 years agoUse negative-list for "weird character in interface" warning instead of warning for...
Patrick McHardy [Wed, 20 Sep 2006 08:32:25 +0000 (08:32 +0000)] 
Use negative-list for "weird character in interface" warning instead of warning for basically every non-alphanumeric character.

18 years agoNamed realm (Simon Lodal <simon@parknet.dk>)
Simon Lodal [Sat, 2 Sep 2006 12:37:48 +0000 (12:37 +0000)] 
Named realm (Simon Lodal <simon@parknet.dk>)

Optionally read realm values from /etc/iproute2/rt_realms

18 years agoAdd statistic match extension
Patrick McHardy [Thu, 31 Aug 2006 14:01:35 +0000 (14:01 +0000)] 
Add statistic match extension

18 years agoiptables: fix ipt_MARK documentation (Eric Leblond)
Eric Leblond [Tue, 29 Aug 2006 09:45:21 +0000 (09:45 +0000)] 
iptables: fix ipt_MARK documentation (Eric Leblond)

This patch documents --or-mask and --and-mask options of the MARK
target. Description is directly taken from the source code.

18 years agoiptables -Z clears the per-rule counters, but not the chain policy counters (Andy...
Andy Gay [Tue, 22 Aug 2006 02:56:41 +0000 (02:56 +0000)] 
iptables -Z clears the per-rule counters, but not the chain policy counters (Andy Gay <andy@andynet.net>)

https://bugzilla.netfilter.org/bugzilla/show_bug.cgi?id=502

18 years agoupdate quota match for xtables + fix -D bug (Phil Oester <kernel@linuxace.com>)
Phil Oester [Tue, 8 Aug 2006 09:59:59 +0000 (09:59 +0000)] 
update quota match for xtables + fix -D bug (Phil Oester <kernel@linuxace.com>)

19 years agoRevert "proto_to_name duplication" patch, as noticed by Yasuyuki it can cause
Patrick McHardyJesper Brouer [Tue, 25 Jul 2006 01:50:48 +0000 (01:50 +0000)] 
Revert "proto_to_name duplication" patch, as noticed by Yasuyuki it can cause
invalid arguments to get accepted.

19 years agoproto_to_name duplication (Phil Oester <kernel@linuxace.com>)
Phil Oester [Sat, 22 Jul 2006 14:10:53 +0000 (14:10 +0000)] 
proto_to_name duplication (Phil Oester <kernel@linuxace.com>)

Update multiport match to use the iptables version of proto_to_name
instead of reinventing the wheel.

19 years agoBUG: libiptc chain references bug (Jesper Brouer <hawk@diku.dk>)
Patrick McHardyJesper Brouer [Tue, 25 Jul 2006 01:50:48 +0000 (01:50 +0000)] 
BUG: libiptc chain references bug (Jesper Brouer <hawk@diku.dk>)

Correcting a chain references increment bug in libiptc.

The bug lies in function iptc_delete_entry() / TC_DELETE_ENTRY.  The
problem is the construction of "r" the rule entry, that is used for
comparison. The problem is that the function iptcc_map_target()
increase the target chains references count.

19 years agoreduce parse_*_port duplication (Phil Oester <kernel@linuxace.com>)
Phil Oester [Thu, 20 Jul 2006 17:01:54 +0000 (17:01 +0000)] 
reduce parse_*_port duplication (Phil Oester <kernel@linuxace.com>)

The below patch (dependent upon my 'reduce service_to_port duplication' patch)
centralizes the parse_*_port functions into parse_port.

19 years agoreduce service_to_port duplication (Phil Oester <kernel@linuxace.com>)
Phil Oester [Thu, 20 Jul 2006 17:00:19 +0000 (17:00 +0000)] 
reduce service_to_port duplication (Phil Oester <kernel@linuxace.com>)

The service_to_port function is used in a number of places, and could
benefit from some centralization instead of being duplicated everywhere.

19 years agoUse gcc to build shared objects (Phil Oester <kernel@linuxace.com>)
Phil Oester [Thu, 20 Jul 2006 16:59:04 +0000 (16:59 +0000)] 
Use gcc to build shared objects (Phil Oester <kernel@linuxace.com>)

As suggested by Dmitry Levin and included in Fedora Core releases,
use gcc instead of ld to link shared objects.  Fedora rpm notes
refer to this fixing a plugin problem, but does not offer specifics.
But in any event, 'gcc -dumpspecs' does show gcc will pass a number
of parameters which in theory it thinks are better.

Compile tested both with and without NO_SHARED_LIBS.

Closes bug #454.

19 years agoiptables: handle cidr notation more sanely (Phil Oester <kernel@linuxace.com>)
Phil Oester [Mon, 10 Jul 2006 04:52:56 +0000 (04:52 +0000)] 
iptables: handle cidr notation more sanely (Phil Oester <kernel@linuxace.com>)

At present, a command such as

iptables -A foo -s 10.10/16

will interpret 10.10/16 as 10.0.0.10/16, and after applying the mask end
up with 10.0.0.0/16, which likely isn't what the user intended.  Yet
some people do expect 10.10 (without the cidr notation) to end up as
10.0.0.10.

The below patch should satisfy all parties.  It zero pads the missing
octets only in the cidr case, leaving the IP untouched otherwise.

This resolves bug #422

19 years agoplease kill santa-claus (Pierre-Yves Ritschard <pierre-yves@spootnik.org>)
Pierre-Yves Ritschard [Wed, 5 Jul 2006 15:58:45 +0000 (15:58 +0000)] 
please kill santa-claus (Pierre-Yves Ritschard <pierre-yves@spootnik.org>)

Remove "hoho" message :)

19 years agolibiptc symbols clash (Phil Oester <kernel@linuxace.com>)
Phil Oester [Wed, 5 Jul 2006 09:31:45 +0000 (09:31 +0000)] 
libiptc symbols clash (Phil Oester <kernel@linuxace.com>)

As reported by Dmitry Levin, the TC_NUM_RULES and TC_GET_RULE exports
clash.  His patch below, resolving bug #456

19 years ago- force user to specify --icmpv6-type if icmpv6 match is required to load
Yasuyuki KOZAKAI [Tue, 4 Jul 2006 10:23:26 +0000 (10:23 +0000)] 
- force user to specify --icmpv6-type if icmpv6 match is required to load
- Don't allow multiple --icmp-type/icmpv6-type

(Closes: #461)

19 years agoip6tables multiport does not support x:y (Phil Oester <kernel@linuxace.com>)
Phil Oester [Mon, 3 Jul 2006 18:35:24 +0000 (18:35 +0000)] 
ip6tables multiport does not support x:y (Phil Oester <kernel@linuxace.com>)

Update the manpage for ip6tables multiport match to reflect
reality -- it does not (yet) support x:y syntax.  I looked at
adding it, but adding revision support to ip6tables seems a
waste at this point, since once xtables support is added to
iptables, this problem will resolve itself.

Closes bug #451.

19 years agoiptables trivial compile warning cleanup (Phil Oester <kernel@linuxace.com>)
Phil Oester [Mon, 3 Jul 2006 18:20:59 +0000 (18:20 +0000)] 
iptables trivial compile warning cleanup (Phil Oester <kernel@linuxace.com>)

Cleanup a few compile warnings in latest snapshot:

extensions/libipt_dscp_helper.c:69: warning: 'dscp_to_name' defined but not used
extensions/libipt_sctp.c: In function 'print_chunks':
extensions/libipt_sctp.c:465: warning: value computed is not used
extensions/libipt_sctp.c:477: warning: value computed is not used

Resolves bug #457.

19 years agosize_t changed to socklen_t in getsockopt call
Joszef Kadlecsik [Fri, 23 Jun 2006 10:01:08 +0000 (10:01 +0000)] 
size_t changed to socklen_t in getsockopt call

19 years agoset match negation bug fixed
Joszef Kadlecsik [Fri, 23 Jun 2006 09:55:12 +0000 (09:55 +0000)] 
set match negation bug fixed

19 years agoREDIRECT does not accept IP (Phil Oester <kernel@linuxace.com>)
Phil Oester [Tue, 20 Jun 2006 13:45:38 +0000 (13:45 +0000)] 
REDIRECT does not accept IP (Phil Oester <kernel@linuxace.com>)

As pointed out by Nicolas Mailhot in bugzilla #483, REDIRECT
does not accept an IP address and when supplied with one,
provides unexpected results.  Patch below fixes this.

19 years agoAdd new exit value to indicate concurrency issues (Jesper Dangaard Brouer <hawk@comx...
Jesper Dangaard Brouer [Mon, 19 Jun 2006 16:53:03 +0000 (16:53 +0000)] 
Add new exit value to indicate concurrency issues (Jesper Dangaard Brouer <hawk@comx.dk>)

19 years agotrivial connlimit manpage fix (Phil Oester <kernel@linuxace.com>)
Phil Oester [Mon, 29 May 2006 22:59:13 +0000 (22:59 +0000)] 
trivial connlimit manpage fix (Phil Oester <kernel@linuxace.com>)

19 years agoUse lowercase letters for match name (Simon Lodal <simonl@parknet.dk>)
Simon Lodal [Wed, 24 May 2006 16:25:09 +0000 (16:25 +0000)] 
Use lowercase letters for match name (Simon Lodal <simonl@parknet.dk>)

19 years agoAdd information about :<port> syntax (Evan Miller <evanm@frap.net>)
Evan Miller [Wed, 24 May 2006 16:21:57 +0000 (16:21 +0000)] 
Add information about :<port> syntax (Evan Miller <evanm@frap.net>)

19 years agosecmark: Add libip6t_CONNSECMARK
James Morris [Wed, 24 May 2006 16:15:27 +0000 (16:15 +0000)] 
secmark: Add libip6t_CONNSECMARK

This patch adds the shared library module for the CONNSECMARK target
(IPv6).

Signed-off-by: James Morris <jmorris@namei.org>
19 years agoD'oh .. I'm not too smart, forgot to add the new files in the previous patches :)
Patrick McHardy [Wed, 24 May 2006 16:15:03 +0000 (16:15 +0000)] 
D'oh .. I'm not too smart, forgot to add the new files in the previous patches :)

19 years agosecmark: Add libipt_CONNSECMARK
James Morris [Wed, 24 May 2006 16:12:55 +0000 (16:12 +0000)] 
secmark: Add libipt_CONNSECMARK

This patch adds the shared library module for the CONNSECMARK target
(IPv4).

Signed-off-by: James Morris <jmorris@namei.org>
19 years agosecmark: Add libip6t_SECMARK
James Morris [Wed, 24 May 2006 16:12:40 +0000 (16:12 +0000)] 
secmark: Add libip6t_SECMARK

This patch adds the shared library module for the SECMARK target (IPv6).

Signed-off-by: James Morris <jmorris@namei.org>
19 years agosecmark: Add libipt_SECMARK
James Morris [Wed, 24 May 2006 16:12:24 +0000 (16:12 +0000)] 
secmark: Add libipt_SECMARK

This patch adds the shared library module for the SECMARK target (IPv4).

Signed-off-by: James Morris <jmorris@namei.org>
19 years agosecmark: Add libselinux support
James Morris [Wed, 24 May 2006 16:11:58 +0000 (16:11 +0000)] 
secmark: Add libselinux support

This patch adds the infrastructure for linking iptables against
libselinux, for use with the SECMARK target.  This is enabled
by setting DO_SELINUX=1 in the build environment.

Signed-off-by: James Morris <jmorris@namei.org>
19 years agoAdd DCCP/SCTP support to multiport. Patch for kernel will go in 2.6.18.
Patrick McHardy [Fri, 28 Apr 2006 08:10:08 +0000 (08:10 +0000)] 
Add DCCP/SCTP support to multiport. Patch for kernel will go in 2.6.18.

19 years agoReplace annoying "Something wrong... deleting dependencies" message by something...
Patrick McHardy [Fri, 28 Apr 2006 08:09:07 +0000 (08:09 +0000)] 
Replace annoying "Something wrong... deleting dependencies" message by something more useful.

19 years agoDon't overwrite errno with return value of setsockopt (which is -1 on error).
Patrick McHardy [Sat, 22 Apr 2006 02:08:56 +0000 (02:08 +0000)] 
Don't overwrite errno with return value of setsockopt (which is -1 on error).
Fixes "Unknown error 4294967295" message (bugzilla #460).

19 years agoRevert incorrect fix for "Unknown error 4294967295" problem
Patrick McHardyHarald Welte [Sat, 22 Apr 2006 02:08:12 +0000 (02:08 +0000)] 
Revert incorrect fix for "Unknown error 4294967295" problem

19 years agoWhen entering an invalid command (such as iptables -A INPUT -j MARK --set-mark
Harald Welte [Fri, 21 Apr 2006 12:31:53 +0000 (12:31 +0000)] 
When entering an invalid command (such as iptables -A INPUT -j MARK --set-mark
1), the error message "Unknown error 4294967295" is displayed; (Closes: #460)

19 years agoIn ip[6]tables.c, NUMBER_OF_OPT was increased to 12 for the OPT_COUNTERS
Patrick McHardyHarald Welte [Sat, 22 Apr 2006 02:08:12 +0000 (02:08 +0000)] 
In ip[6]tables.c, NUMBER_OF_OPT was increased to 12 for the OPT_COUNTERS
option.  However, the new array element is not initialized in either
commands_v_options[NUMBER_OF_CMD][NUMBER_OF_OPT] or
inverse_for_options[NUMBER_OF_OPT]. (Closes: #462)

19 years agocmdflags is used in cmd2char() to return the option for a command. It uses the
Harald Welte [Fri, 21 Apr 2006 11:56:30 +0000 (11:56 +0000)] 
cmdflags is used in cmd2char() to return the option for a command.  It uses the
bit position of the command mask as an index in the array.  There's no entry for
CMD_CHECK (0x0800U), so lookups for CMD_RENAME_CHAIN (0x1000U) index outside the
array. (Closes: #463)

19 years ago[IPTABLES,IP6TABLES]: check invalid esp spi range
Yasuyuki KOZAKAI [Sat, 15 Apr 2006 03:11:15 +0000 (03:11 +0000)] 
[IPTABLES,IP6TABLES]: check invalid esp spi range

19 years ago[IP6TABLES] kill manual comparing protocol name with "ipv6-icmp".
Yasuyuki KOZAKAI [Sat, 15 Apr 2006 03:09:37 +0000 (03:09 +0000)] 
[IP6TABLES] kill manual comparing protocol name with "ipv6-icmp".

19 years agofix loading shared library of ICMPv6 match.
Yasuyuki KOZAKAI [Sat, 15 Apr 2006 03:05:41 +0000 (03:05 +0000)] 
fix loading shared library of ICMPv6 match.

The current ip6tables tries to load libip6t_icmp6.so when user types
'ip6tables -p icmpv6 ...' or 'ip6tables ... -m icmpv6' ...', and it fails.

This patch renames libip6t_icmpv6.c to libip6t_icmp6.c so that ip6tables
can load it. Now kernel module and user library has same name 'icmp6'.
It can reduce confusion about name mismatch. That's why I renamed it
instead of reverting change in find_match() which brought this bug.

This patch keeps compatibiity and we can use '-p icmpv6', '-p ipv6-icmpv6',
'-m icmpv6', '-m ipv6-icmpv6', and '-m icmp6', as ever.

19 years ago[IPTABLES,IP6TABLES]: fix the path to detect esp/connbytes support in kernel
Harald Welte [Wed, 12 Apr 2006 10:58:41 +0000 (10:58 +0000)] 
[IPTABLES,IP6TABLES]: fix the path to detect esp/connbytes support in kernel

The recent kernels don't have ipt_connbytes.c and ip6t_esp.c.

19 years agoCorrect iptables-save output of osf module (Daniel De Graaf)
Daniel De Graaf [Fri, 31 Mar 2006 01:31:29 +0000 (01:31 +0000)] 
Correct iptables-save output of osf module (Daniel De Graaf)

19 years agodon't allow to specify protocol of IPv6 extension header (Yasuyuki Kozakai)
Yasuyuki KOZAKAI [Wed, 29 Mar 2006 09:24:43 +0000 (09:24 +0000)] 
don't allow to specify protocol of IPv6 extension header (Yasuyuki Kozakai)

Sometimes I hear that people do 'ip6tables -p ah ...' which never matches
any packet. IPv6 extension headers except of ESP are skipped and invalid
as argument of '-p'. Then I propose that ip6tables exits with error in such
case.

19 years agoMultiple matches of the same type can be specified on the commandline.
Joszef Kadlecsik [Fri, 3 Mar 2006 09:36:50 +0000 (09:36 +0000)] 
Multiple matches of the same type can be specified on the commandline.

If two or more matches of the same type are detected then the options
are assumed to be grouped in order to tell which option belongs
to which match:

... -m foo ... <options0> ... -m foo ... <options1> ...

Otherwise the commandline parsing is unmodified.

19 years agoMake '-p all' a special case that is handled before calling getprotoent() (Closes...
Harald Welte [Sat, 11 Feb 2006 09:50:11 +0000 (09:50 +0000)] 
Make '-p all' a special case that is handled before calling getprotoent() (Closes: #446)

19 years agofix double-free if a single match is used multiple times within a signle rule
Harald Welte [Sat, 11 Feb 2006 09:34:16 +0000 (09:34 +0000)] 
fix double-free if a single match is used multiple times within a signle rule
(Closes: #440).  However, while this fixes the double-free, it still doesn't make iptables
support two of the same matches within one rule.  Apparently the last matchinfo is copied into all the previous
matchinfo instances.

19 years agodon't install libiptc.a
Harald Welte [Thu, 9 Feb 2006 14:53:40 +0000 (14:53 +0000)] 
don't install libiptc.a

19 years agofix segfault or loading of invalid counters in ip[6]tables-restore (Olaf Rempel)...
Harald Welte [Thu, 9 Feb 2006 14:35:38 +0000 (14:35 +0000)] 
fix segfault or loading of invalid counters in ip[6]tables-restore (Olaf Rempel) (Closes: #437)

19 years agomake policy match compile independant of kernel headers v1.3.5
Harald Welte [Wed, 1 Feb 2006 13:05:33 +0000 (13:05 +0000)] 
make policy match compile independant of kernel headers

19 years agoSome !%$!*##$@ has modified the kernel include/linux/netfilter_ipv4/ipt_sctp.h
Harald Welte [Wed, 1 Feb 2006 12:58:28 +0000 (12:58 +0000)] 
Some !%$!*##$@ has modified the kernel include/linux/netfilter_ipv4/ipt_sctp.h
file in a way that breaks userspace :(

19 years agofix ipt_conntrack compilation against very early (2.4.0) kernel releases
Harald Welte [Wed, 1 Feb 2006 12:57:34 +0000 (12:57 +0000)] 
fix ipt_conntrack compilation against very early (2.4.0) kernel releases

19 years agoremove other bits of old ip pool code, people should use ipset (ipset.netfilter.org...
Harald Welte [Wed, 1 Feb 2006 12:42:52 +0000 (12:42 +0000)] 
remove other bits of old ip pool code, people should use ipset (ipset.netfilter.org) these days

19 years agoremove ippool
Harald Welte [Wed, 1 Feb 2006 12:41:24 +0000 (12:41 +0000)] 
remove ippool

19 years agoPrepare policy match for x_tables unification by making sure both
Patrick McHardy [Tue, 31 Jan 2006 18:24:14 +0000 (18:24 +0000)] 
Prepare policy match for x_tables unification by making sure both
ipt_policy and ip6t_policy use the same data structure.

19 years agofix 'save' (Michael Rash)
Michael Rash [Mon, 30 Jan 2006 09:02:45 +0000 (09:02 +0000)] 
fix 'save' (Michael Rash)

19 years agomajor manpage update (Yasuyuki Kozakai)
Yasuyuki KOZAKAI [Mon, 30 Jan 2006 08:50:09 +0000 (08:50 +0000)] 
major manpage update (Yasuyuki Kozakai)

19 years agoAdd 'copy+paste' support for 'state' and 'connmark' match, as well as
Harald Welte [Thu, 26 Jan 2006 14:43:52 +0000 (14:43 +0000)] 
Add 'copy+paste' support for 'state' and 'connmark' match, as well as
'CONNMARK' target for ip6tables / nf_conntrack_l3proto_ipv6.  This is a temporary solution for the iptables-1.3.x branch, since the 1.4.x branch will have proper support.

19 years agoadd note about deprecated state
Harald Welte [Thu, 26 Jan 2006 14:43:01 +0000 (14:43 +0000)] 
add note about deprecated state

19 years agofix spelling 'adress' -> 'address' (Closes: #431) (MJ Anthony)
Harald Welte [Sun, 22 Jan 2006 16:02:32 +0000 (16:02 +0000)] 
fix spelling 'adress' -> 'address' (Closes: #431) (MJ Anthony)

19 years agoFix "empty policy element" complaining in non-strict mode.
Noticed by Tom Eastep [Sun, 22 Jan 2006 13:47:07 +0000 (13:47 +0000)] 
Fix "empty policy element" complaining in non-strict mode.
Noticed by Tom Eastep <teastep@shorewall.net>.

19 years agoClarify --tunnel-src/--tunnel-dst options
Patrick McHardy [Thu, 12 Jan 2006 16:14:41 +0000 (16:14 +0000)] 
Clarify --tunnel-src/--tunnel-dst options

19 years agoMove empty policy element check to also catch last element
Patrick McHardy [Thu, 12 Jan 2006 09:43:18 +0000 (09:43 +0000)] 
Move empty policy element check to also catch last element

19 years agoDon't allow using --next option without specifying a policy element
Patrick McHardy [Thu, 12 Jan 2006 09:12:47 +0000 (09:12 +0000)] 
Don't allow using --next option without specifying a policy element

19 years agoFix invalid assignment of tunnel-src to dest address (Patrick McHardy)
Patrick McHardy [Mon, 9 Jan 2006 10:15:04 +0000 (10:15 +0000)] 
Fix invalid assignment of tunnel-src to dest address (Patrick McHardy)

19 years agoAdd documentation for string match (Pablo Neira)
Pablo Neira [Tue, 3 Jan 2006 11:50:16 +0000 (11:50 +0000)] 
Add documentation for string match (Pablo Neira)

19 years agoFix probing for supported revisions (Jones Desougi <jones@ingate.com>)
Jones Desougi [Thu, 22 Dec 2005 03:33:50 +0000 (03:33 +0000)] 
Fix probing for supported revisions (Jones Desougi <jones@ingate.com>)

Bugzilla #413

19 years agofix iptables-save of 'goto' target (Closes: #410)
Harald Welte [Mon, 5 Dec 2005 12:08:03 +0000 (12:08 +0000)] 
fix iptables-save of 'goto' target (Closes: #410)

19 years agoAdd note that TCPMSS is only valid in the mangle table (not true today, but maybe...
Patrick McHardy [Mon, 5 Dec 2005 01:22:50 +0000 (01:22 +0000)] 
Add note that TCPMSS is only valid in the mangle table (not true today, but maybe someday)

19 years agofix compilation of iptables on [old] systems that don't have IPT_F_GOTO
Harald Welte [Thu, 24 Nov 2005 17:04:05 +0000 (17:04 +0000)] 
fix compilation of iptables on [old] systems that don't have IPT_F_GOTO

19 years agonote that we can only delete chains that are empty
Harald Welte [Tue, 22 Nov 2005 22:22:28 +0000 (22:22 +0000)] 
note that we can only delete chains that are empty

19 years agotcp-rst is the alias, not tcp-reset (Torsten Hilbrich)
Harald Welte [Tue, 22 Nov 2005 08:54:28 +0000 (08:54 +0000)] 
tcp-rst is the alias, not tcp-reset (Torsten Hilbrich)

19 years agoAdd policy match extensions from patch-o-matic
Patrick McHardy [Sat, 19 Nov 2005 09:00:03 +0000 (09:00 +0000)] 
Add policy match extensions from patch-o-matic

19 years agoFix some gcc-4 warnings
Patrick McHardy [Fri, 18 Nov 2005 18:00:25 +0000 (18:00 +0000)] 
Fix some gcc-4 warnings

19 years agoDon't eat numeric arguments for other extensions
Patrick McHardy [Fri, 18 Nov 2005 17:59:56 +0000 (17:59 +0000)] 
Don't eat numeric arguments for other extensions

19 years agoThe conntrack match does not print any info for --ctproto, thus
Phil Oester [Thu, 17 Nov 2005 13:34:51 +0000 (13:34 +0000)] 
The conntrack match does not print any info for --ctproto, thus
breaking iptables-restore of any rules using this option.  Below
patch adds output and closes bug #398. (Phil Oester)

19 years agoonly set revisions on real targets, not on jumps. (Pablo Neira)
Pablo Neira [Thu, 17 Nov 2005 13:04:49 +0000 (13:04 +0000)] 
only set revisions on real targets, not on jumps. (Pablo Neira)

19 years ago- Fix memory leak in TC_COMMIT() (Markus Sundberg)
Harald Welte [Sat, 12 Nov 2005 10:39:40 +0000 (10:39 +0000)] 
- Fix memory leak in TC_COMMIT()  (Markus Sundberg)
- Cleanup error path of TC_COMMIT()
- Correctly propagate errors of setsockopt to calling function

19 years agoadd 'goto' support (Henrik Nordstrom <hno@marasystems.com>)
Henrik Nordstrom [Sat, 5 Nov 2005 09:26:40 +0000 (09:26 +0000)] 
add 'goto' support (Henrik Nordstrom <hno@marasystems.com>)

19 years agofix connmark, it's now only 32bits (Deti Fliegl <deti@fliegl.de) v1.3.4
Deti Fliegl [Thu, 3 Nov 2005 18:43:14 +0000 (18:43 +0000)] 
fix connmark, it's now only 32bits (Deti Fliegl <deti@fliegl.de)

We'ver screwed this up with the 2.6.14 release.  It refuses any mask that
extends 32bits.  We should have fixed this by adding a new target/match
revision, but now it's too late anyway :(

19 years agoabout to release 1.3.4
Harald Welte [Mon, 31 Oct 2005 19:22:43 +0000 (19:22 +0000)] 
about to release 1.3.4

19 years agoThe conntrack match extension doesn't handle address inversion correctly. (Tom Eastep)
Tom Eastep [Mon, 19 Sep 2005 15:14:04 +0000 (15:14 +0000)] 
The conntrack match extension doesn't handle address inversion correctly. (Tom Eastep)

19 years agoKernels higher than 2.6.10 don't support multiple --to arguments in
Phil Oester [Mon, 19 Sep 2005 15:00:33 +0000 (15:00 +0000)] 
Kernels higher than 2.6.10 don't support multiple --to arguments in
DNAT and SNAT targets.  At present, the error is somewhat vague:

# iptables -t nat -A foo -j SNAT --to 1.2.3.4 --to 2.3.4.5
iptables: Invalid argument

But if we want current iptables to work with kernels <= 2.6.10, we
cannot simply disallow this in all cases.

So the below patch adds kernel version checking to iptables, and
utilizes it in [DS]NAT.  Now, users will see a more informative error:

# iptables -t nat -A foo -j SNAT --to 1.2.3.4 --to 2.3.4.5
iptables v1.3.3: Multiple --to-source not supported

This generic infrastructure (shamelessly lifted from procps btw) may
come in handy in the future for other changes.

This fixes bugzilla #367. (Phil Oester)

19 years ago* specifying random seed for the Jenkins hash works as documented
KOVACS Krisztian [Mon, 19 Sep 2005 14:50:06 +0000 (14:50 +0000)] 
* specifying random seed for the Jenkins hash works as documented
* iptables-save seems to work now

Signed-off-by: KOVACS Krisztian <hidden@balabit.hu>
Signed-off-by: Harald Welte <laforge@netfilter.org>
19 years agoAdd the aligned_u64 typedef, it's defined in linux/types.h in the kernel.
Martin Josefsson [Sun, 11 Sep 2005 17:32:39 +0000 (17:32 +0000)] 
Add the aligned_u64 typedef, it's defined in linux/types.h in the kernel.
We can't include that header since it conflicts with sys/types.h