Packets which are merely forwarded by the router and which are neither
involved in any DNAT/SNAT nor originate locally, are considered INVALID
from a conntrack point of view, causing them to get dropped in the
zone_*_dest_ACCEPT chains, since those only allow stream with state NEW
or UNTRACKED.
Remove the ctstate restriction on dest accept chains to properly pass-
through unrelated 3rd party traffic.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This commit modifies the /lib/netifd/proto/gre.sh script so that, when
GRE-TAP tunnels are created, either IPv4 or IPv6, the prefix before the chosen
interface name contains the "tap" substring, to differentiate them from non-TAP
GRE tunnels.
Right now, both GRE and GRE-TAP tunnel (either IPv4 or IPv6) interfaces defined
in /etc/config/network are named equally ("gre-"+$ifname or "grev6"+$ifname)
upon creation. For instance, the following tunnels:
config interface 'tuna'
option peeraddr '172.30.22.1'
option proto 'gre'
config interface 'tunb'
option peeraddr '192.168.233.4'
option proto 'gretap'
config interface 'tunc'
option peer6addr 'fdc5:7c9e:e93d:45af::1'
option proto 'grev6'
config interface 'tund'
option peer6addr 'fdc0:6071:1348:31ff::2'
option proto 'grev6tap'
are named, respectively, "gre-tuna", "gre-tunb", "grev6-tunc" and "grev6-tund".
The current change makes that each GRE tunnel interface of the four different
types available (gre, gretap, grev6 and grev6tap) gets a different prefix.
Therefore, the abovementioned tunnels will be named, respectively:
"gre4-tuna", "gre4t-tunb", "gre6-tunc" and "gre6t-tund".
This is coherent with other types of virtual interfaces (i.e. PPP, PPPoE, PPPoA)
where the whole protocol name is used. For instance, a PPPoA interface named
"p1" and a PPPoE interface named "p2" will respectively appear as "pppoa-p1"
and "pppoe-p2", not as "ppp-p1" and "ppp-p2").
Since Linux interfaces names are limited to 15 characters, these prefixes leave,
for the worst case (TAP tunnels), 9 characters for the actual name.
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
64a655d proto: allow configuring deprecated static IPv6 addresses
c99182e remove obsolete /opt/local prefix on Mac OS X
0249d5f system-linux: Don't set gre tunnel ttl by default to 64 (#FS312)
edc15ca ubus: Display the IPv6 prefix assigned address
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Build seems to fail with:
```
Collected errors:
* satisfy_dependencies_for: Cannot satisfy the following dependencies for X:
* grev4 *
* opkg_install_cmd: Cannot install package X
```
After adding an empty install rule, the failure goes away.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
Update to current HEAD in order to drop automatic generation of per-zone
NOTRACK rules.
The NOTRACK rules used to provide a little performance improvement but the
later introduction of the netfilter conntrack cache made those rules largely
unnecessary. Additionally, those rules caused various issues which broke
stateful firewalling in some scenarios.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Update to latest Git head in order to import several fixes and enhancements.
- Disable drop invalid by default (FS#73, FS#154)
Instead of dropping packets with conntrack state INVALID, only allow streams
with explicit NEW or UNTRACKED conntrack state.
This change gives user defined rules the chance to accept traffic like ICMPv6
multicast which would be filtered away by the very early ctstate INVALID drop
rule otherwise.
The old behaviour can be restored by explicitely setting "drop_invalid" to 1
in the global firewall config section.
- Fix re-initialization of loadable iptables extensions on musl (FS#31)
Since musl does not implement actual dlclose() semantics, it is impossible to
re-run initializers on subsequent dlopen() calls.
The firewall3 executable now intercepts the extension registration calls
instead in order to be able to re-call them when needed.
This also allowed us to switch to libxtables' builtin extension loader as a
positive side-effect.
- Fix masquerade rules for multiple negated IP addresses (FS#248)
When building MASQUERADE rules for zones which specify multiple negated
addresses in masq_src or masq_dest, emit -j RETURN rules which jump out of
the masquerading chain instead of creating multiple rules with inverted "-s"
arguments.
- Tag own rules using comments
Instead of relying on the nonstandard xt_id match, use the xt_comment match
to mark own rules. Existing comments are prefixed with "!fw3: " while
uncommented rules are marked with a sole "!fw3" string.
This allows removing the xt_id match entirely in a later commit.
- Make missing ubus connection nonfatal
Technically, firewall3 is able to operate without ubus just fine as long as
the zones are declared using "option device" or "option subnet" instead of
"option network" so do not abort execution if ubus could not be connected or
of no network namespace is exported in ubus.
This allows running firewall3 on ordinary Linux systems.
- Fix conntrack requirement detection for indirectly connected zones
The current code fails to apply the conntrack requirement flag recursively to
zones, leading to stray NOTRACK rules which break conntrack based traffic
policing.
Change the implementation to iteratively reapply the conntrack fixup logic
until no more zones had been changed in order to ensure that all directly and
indirectly connected zones receive the conntrack requirement flag.
- Add support for iptables 1.6.x
Adds support for the xtables version 11 api in order to allow building
against iptables 1.6.x
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This option, defined by RFC3442, allows a DHCP server to send static
routes to a client. But the client has to request this option
explicitely.
Static routes are useful when the gateway configured by DHCP cannot be
in the same subnet as the client. This happens, for instance, when
using DHCP to hand out addresses in /32 subnets.
A new configuration option "classlessroute" is available, allowing
users to disable this feature (the option defaults to true).
Other DHCP clients already request this option by default (dhcpcd, for
instance, and possibly Windows). If a DHCP server does not support
this option, it will simply ignore it.
Signed-off-by: Baptiste Jonglez <git@bitsofnetworks.org>
* Change git packages to xz
* Update mirror checksums in packages where they are used
* Change a few source tarballs to xz if available upstream
* Remove unused lines in packages we're touching, requested by jow- and blogic
* We're relying more on xz-utils so add official mirror as primary source, master site as secondary.
* Add SHA256 checksums to multiple git tarball packages
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
Emits an initial event after the first link-up of a force_link
interface. This is needed for making the dnsmasq dhcp check more
reliable
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Passing the hostname is currently broken in since the shipped busybox includes this commit:
https://git.busybox.net/busybox/commit/networking/udhcp/dhcpc.c?id=2017d48c0d70bef8768efb42909e605ea8eb5a21
Before:
Sun Jan 31 18:11:32 2016 daemon.notice netifd: Interface 'wan' is now down
Sun Jan 31 18:11:32 2016 daemon.notice netifd: Interface 'wan' is setting up now
Sun Jan 31 18:11:32 2016 daemon.notice netifd: wan (18158): udhcpc: option -h NAME is deprecated, use -x hostname:NAME
Sun Jan 31 18:11:32 2016 daemon.notice netifd: wan (18158): udhcpc: malformed hex string 'WR150'
After:
Sun Jan 31 18:11:33 2016 daemon.notice netifd: wan (18169): udhcpc (v1.23.2) started
Sun Jan 31 18:11:33 2016 daemon.notice netifd: wan (18169): Sending discover...
Sun Jan 31 18:11:33 2016 daemon.notice netifd: wan (18169): Sending select for xxx.yyy.zzz.xyz...
Sun Jan 31 18:11:33 2016 daemon.notice netifd: wan (18169): Lease of xxx.yyy.zzz.xyz obtained, lease time 600
Signed-off-by: Merlijn Wajer <merlijn@wizzup.org>
puts br2684ctl init after ADSL init instead of before, so that the ESI
is set at the right time, and for consistency with the PTM driver.
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
Remove the udhcpc -R release option as sending a DHCP release
is configurable via the uci option release.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Update to latest HEAD in order to fix MARK rule generation for local traffic,
also fix a possible race condition during firewall start.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Make sending a DHCP release configurable when the client exits allowing to clean up
IP/mac state info in intermediate devices.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Add nonshared flag to package depending on specific targets or subtargets as
there's no guarantee otherwise that they'll be available in the shared repo.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Let DHCP client send a release when it exists so the DHCP server is
informed the IP address is released and allowing to clean up IP/mac
state info in intermediate devices.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Update to latest HEAD in order to remove the faulty "prelocal" ip rule leading
to unexpected policy rule precedence.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
I do not know if this causes any problems now, but we should not set
it, because it is reserved. Some more recent versions of the Lantiq DSL
API driver and Control is checking if only valid bits are set.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 48948
There are some cases where ISPs are running ATM over VDSL or PTM over
ADSL, this is not the common case, but these cases exist. Make it
possible to configure OpenWrt for such cases by adding a new config
option line_mode.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 48947
The detailed annex option were only available in the danube DSL app
including the activation of G.992.2 Annex A (ADSL Lite). This is now
also added to the vdsl app for the vrx200.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 48946
The adsl control app missed the activation of annex M and annex L in
the Annex A part, this now activates everything the firmware supports.
In Annex L type only the wide US (Mask1) was activated, now also the
narrow US (Mask2) version gets activated.
In addition annex J was also added.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 48945
I am not calling dsl_cmd because I want to ignore the lock, quit
should also be send when someone else is accessing it. I saw that some
other call was stuck here and all following calls were stuck in the
dsl_cmd lock.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 48943
Ensure that ikey and okey are sent in network byte order to the kernel.
Also don't mangle external IP addrs and routes when reconfiguring iinterfaces.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 48919
Before r47933 Bit 1 (first bit) of xTSE Octet 1 (first octet) defaulted
to 1, which allowed T1.413 to operate.
Signed-off-by: Jonathan A. Kollasch <jakllsch@kollasch.net>
SVN-Revision: 48763
Supported syntax is inspired by ethtool. Example usages:
swconfig dev switch0 port 2 set link "duplex half speed 100"
swconfig dev switch0 port 2 set link "autoneg on"
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 48624
UCI paramater multicast is added which allows to toggle multicast support on gre interfaces.
By default multicast support is enabled as gre tunnels are often used in combination with
routing protocols using multicast.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Signed-off-by: Nick Podolak <nicholas.podolak@dtechlabs.com>
SVN-Revision: 48596
* Enable drop_invalid by default to catch unnatted packets (#21738)
* Fix processing of inversions for -i, -o, -s, -d and -p flags
* Remove delegate_* chain indirection but rely on xt_id to identify own rules
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 48551
This adds IPv6 support to qos-scripts for both tc/qdisc and the
iptables classification rules. The tc/qdisc part is accomplished
by removing "protocol ip" from the tc command line, causing the
rule to be applied to all protocols. The iptables part is
accomplished by adding each rule using both iptables and ip6tables.
This patch is based on previous work by Ilkka Ollakka and
Dominique Martinet.
Signed-off-by: Michael Marley <michael@michaelmarley.com>
SVN-Revision: 48452
This adds a "srciface" option that can be used on classification
rules in /etc/config/qos. This is useful to allow prioritization
based on the local network from which the traffic originates, for
example to deprioritize traffic from a guest network.
Signed-off-by: Michael Marley <michael@michaelmarley.com>
SVN-Revision: 48446
Set the save-mark mask for the qos_${cg} chain to 0xff instead of
0xf0. With the old value, the nibble that was saved would be
masked during the restore, preventing ingress traffic from being
classified. Thanks to nbd for recommending the fix.
Signed-off-by: Michael Marley <michael@michaelmarley.com>
SVN-Revision: 48388
This adds support for configuring VTI interfaces within /etc/config/network.
VTI interfaces are used to create IPsec tunnel interfaces. These interfaces
may be used for routing and other purposes.
Example config:
config interface 'vti1'
option proto 'vti'
option mtu '1500'
option tunlink 'wan'
option peeraddr '192.168.5.16'
option zone 'VPN'
option ikey 2
option okey 2
config interface 'vti1_static'
option proto 'static'
option ifname '@vti1'
option ipaddr '192.168.7.2/24'
The options ikey and okey correspond to the fwmark value of a ipsec policy.
The may be null if you do not want fwmarks.
Also peeraddr may be 0.0.0 if you want all ESP packets go through the
interface.
Example strongswan config:
conn vti
left=%any
leftcert=peer2.test.der
leftid=@peer2.test
right=192.168.5.16
rightid=@peer3.test
leftsubnet=0.0.0.0/0
rightsubnet=0.0.0.0/0
mark=2
auto=route
Signed-off-by: André Valentin <avalentin@marcant.net>
SVN-Revision: 48274
Source package gre was depending on kmod-ip6-gre, however the actual
kernel module package that is created is kmod-gre6. Therefore
update (source) package gre for ipv6 gre support.
Signed-off-by: Daniel Dickinson <openwrt@daniel.thecshore.com>
SVN-Revision: 48100
Add back a slightly modified version of the lowlevel settings which
where removed with r46920.
In compare to the old lowlevel settings, the B43c tone is added to
tone_adsl_b and tone_adsl_bv.
If an unsupported tone value is used, the auto probing mode is used, in
compare to the fallback to tone_adsl_av and tone_vdsl_av with the old
lowlevel settings.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 48054
According to ITU-T G.997.1 Amendment 2 (04/2013) section 2.1, bit 3 of
XTSE octet 8 either allow or denies the initialization of G.993.5.
Even if the current redistributable xDSL firmware doesn't include
G.993.5 vectoring support, enable this bit by default to allow people to
get their G.993.5 line working using a custom xDSL firmware.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 48053
r47933 revealed that the driver/app in combination with the chosen
firmware does a good job in selecting a working xtse.
Use this probing mode if no annex is specified.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 48052
This patch adds the missing VDSL2 bits to the annex specific XTSE (like
it should be according to the comments above the XTSE bits).
Since r47933 it's mandatory to remove the annex option to switch to
VDSL2 (only) operation mode.
As shown by ticket #21436 and a few mails I received personally, even
experienced users are not aware that they have to remove the annex
option to get their VDSL2 line working and as shown by this patch it
doesn't need to be that "complicated".
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 48051
When using cli, print link state the same way kernel used to do it.
This will allow kernel switching PORT_LINK from SWITCH_TYPE_STRING.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47998
- CONFIG_IFX_CLI is unused, couldn't find any reference to this config variable
- use disable-feature instead of enable-feature=no
- reorder configure args to have depending args together
- remove configure args which set the default value
- group enable-model and configure args which enable or disable features that
are covered by the feature set
The config.log contains the same values as before. The vdsl_cpe_control binary
has the same checksum as before.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 47888
The typicial feature set doesn't include "DSL PM showtime counters support"
(INCLUDE_DSL_CPE_PM_SHOWTIME_COUNTERS). This feature provides the
vdsl_cpe_control command 'pmccsg', which is used by 'dsl_control status' to get
the line uptime.
The binary size increases to 103912 byte (+4256 byte) uncompressed.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 47887
this error was not visible until recent bump to
busybox 1.24.1 stable which introduced a warning message
when keyword 'local' is not used with a shell-function.
this does not change behavior and is a cosmetic cleanup.
fixes the following output:
root@box:~ ifup <interface>
/sbin/ifup: local: line 362: not in a function
/sbin/ifup: local: line 362: not in a function
/sbin/ifup: local: line 1: not in a function
Signed-off-by: Bastian Bittorf <bittorf@bluebottle.com>
SVN-Revision: 47828
Use the 'typical' compile configuration instead of 'full', which most
notably excludes the soap support.
/sbin/vdsl_cpe_control shrinks down to ~50%, from 178kb(!) to 90kb.
Signed-off-by: Andre Heider <a.heider@gmail.com>
SVN-Revision: 47769
In this upstream dsl driver app version the autoboot is deactivated activate
it again.
In addition to the update this also fixes some build warnings and makes it
use the same configure option as used in Lantiq UGW.
Signed-off-by: Hauke Mehrtens <hauke.mehrtens@lantiq.com>
SVN-Revision: 47637
There is no RFC requirement that DHCPv6 servers must reply with a link local
address and some ISP servers in the wild appear to using addresses in the ULA
range to send DHCPv6 offers.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47048
Since r46834, IPv6 support is builtin if selected. Therefor, dependencies
on kmod-ipv6 can no longer be fulfilled, since it is not a module anymore.
Signed-off-by: Arjen de Korte <arjen+openwrt@de-korte.org>
SVN-Revision: 47022
Our ruleset requires kernel support for conntrack state matching, therfore
depend on the require kmod. Fixes#20542.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 46990
Thanks to Sylwester Petela for testing my patch (successfully on an
ADSL connection) and for pointing out some configuration mistakes.
Others (including me) have also successfully tested this extensively
on VDSL connections.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 46920
Move the pattern setting from netifd's service script to
/etc/sysctl.conf. Put the timestamp component '%t' just after
executable name '%e' for more natural order from output of ls command.
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 46867
Seems like my second try was again whitespace broken. Sorry for the noise.
Remove src_port from firewall.config to receive dhcpv6 replies. Fixes#20295.
Signed-off-by: Anselm Eberhardt <a.eberhardt@cygnusnetworks.de>
SVN-Revision: 46842
otherwise it's not picked up by toolchain:
staging_dir/toolchain-mipsel_24kec+dsp_gcc-4.8-linaro_musl-1.1.10/lib/gcc/mipsel-openwrt-linux-musl/4.8.3/../../../../mipsel-openwrt-linux-musl/bin/ld: cannot find -lsw
Signed-off-by: Roman Yeryomin <roman@advem.lv>
SVN-Revision: 46406
Use shared libipt{,4,6}ext.so libraries instead of statically linking
the userspace matches into the fw3 executable.
As a side effect the match initialization is extremely simplified
compared to the weak function pointer juggling performed before.
This also fixes the initialization of the multiport match.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 45764
hand over parameters to user-script e.g. $1=deconfig
Signed-off-by: Leon George <leon@georgemail.de>
Signed-off-by: Christian Mehlis <christian@m3hlis.de>
SVN-Revision: 45626
The WAN port should at least respond to IGMP and MLD queries as
otherwise a snooping bridge/switch might drop traffic.
RFC4890 recommends to leave IGMP and MLD unfiltered as they are always
link-scoped anyways.
Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue>
SVN-Revision: 45613
Certain DHCP servers push a gateway outside of the assigned interface subnet,
to support those situations install a host route towards the gateway.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 44789
