This work adds the possibility of deriving the zone id from the skb->mark
field in a scalable manner. This allows for having only a single template
serving hundreds/thousands of different zones, for example, instead of the
need to have one match for each zone as an extra CT jump target.
Note that we'd need to have this information attached to the template as at
the time when we're trying to lookup a possible ct object, we already need
to know zone information for a possible match when going into
__nf_conntrack_find_get(). This work provides a minimal implementation for
a possible mapping.
In order to not add/expose an extra ct->status bit, the zone structure has
been extended to carry a flag for deriving the mark.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This work adds a direction parameter to netfilter zones, so identity
separation can be performed only in original/reply or both directions
(default). This basically opens up the possibility of doing NAT with
conflicting IP address/port tuples from multiple, isolated tenants
on a host (e.g. from a netns) without requiring each tenant to NAT
twice resp. to use its own dedicated IP address to SNAT to, meaning
overlapping tuples can be made unique with the zone identifier in
original direction, where the NAT engine will then allocate a unique
tuple in the commonly shared default zone for the reply direction.
In some restricted, local DNAT cases, also port redirection could be
used for making the reply traffic unique w/o requiring SNAT.
The consensus we've reached and discussed at NFWS and since the initial
implementation [1] was to directly integrate the direction meta data
into the existing zones infrastructure, as opposed to the ct->mark
approach we proposed initially.
As we pass the nf_conntrack_zone object directly around, we don't have
to touch all call-sites, but only those, that contain equality checks
of zones. Thus, based on the current direction (original or reply),
we either return the actual id, or the default NF_CT_DEFAULT_ZONE_ID.
CT expectations are direction-agnostic entities when expectations are
being compared among themselves, so we can only use the identifier
in this case.
Note that zone identifiers can not be included into the hash mix
anymore as they don't contain a "stable" value that would be equal
for both directions at all times, f.e. if only zone->id would
unconditionally be xor'ed into the table slot hash, then replies won't
find the corresponding conntracking entry anymore.
If no particular direction is specified when configuring zones, the
behaviour is exactly as we expect currently (both directions).
Support has been added for the CT netlink interface as well as the
x_tables raw CT target, which both already offer existing interfaces
to user space for the configuration of zones.
Below a minimal, simplified collision example (script in [2]) with
netperf sessions:
+--- tenant-1 ---+ mark := 1
| netperf |--+
+----------------+ | CT zone := mark [ORIGINAL]
[ip,sport] := X +--------------+ +--- gateway ---+
| mark routing |--| SNAT |-- ... +
+--------------+ +---------------+ |
+--- tenant-2 ---+ | ~~~|~~~
| netperf |--+ +-----------+ |
+----------------+ mark := 2 | netserver |------ ... +
[ip,sport] := X +-----------+
[ip,port] := Y
On the gateway netns, example:
iptables -t raw -A PREROUTING -j CT --zone mark --zone-dir ORIGINAL
iptables -t nat -A POSTROUTING -o <dev> -j SNAT --to-source <ip> --random-fully
iptables -t mangle -A PREROUTING -m conntrack --ctdir ORIGINAL -j CONNMARK --save-mark
iptables -t mangle -A POSTROUTING -m conntrack --ctdir REPLY -j CONNMARK --restore-mark
conntrack dump from gateway netns:
netperf -H 10.1.1.2 -t TCP_STREAM -l60 -p12865,5555 from each tenant netns
tcp 6 431995 ESTABLISHED src=40.1.1.1 dst=10.1.1.2 sport=5555 dport=12865 zone-orig=1
src=10.1.1.2 dst=10.1.1.1 sport=12865 dport=1024
[ASSURED] mark=1 secctx=system_u:object_r:unlabeled_t:s0 use=1
tcp 6 431994 ESTABLISHED src=40.1.1.1 dst=10.1.1.2 sport=5555 dport=12865 zone-orig=2
src=10.1.1.2 dst=10.1.1.1 sport=12865 dport=5555
[ASSURED] mark=2 secctx=system_u:object_r:unlabeled_t:s0 use=1
tcp 6 299 ESTABLISHED src=40.1.1.1 dst=10.1.1.2 sport=39438 dport=33768 zone-orig=1
src=10.1.1.2 dst=10.1.1.1 sport=33768 dport=39438
[ASSURED] mark=1 secctx=system_u:object_r:unlabeled_t:s0 use=1
tcp 6 300 ESTABLISHED src=40.1.1.1 dst=10.1.1.2 sport=32889 dport=40206 zone-orig=2
src=10.1.1.2 dst=10.1.1.1 sport=40206 dport=32889
[ASSURED] mark=2 secctx=system_u:object_r:unlabeled_t:s0 use=2
Taking this further, test script in [2] creates 200 tenants and runs
original-tuple colliding netperf sessions each. A conntrack -L dump in
the gateway netns also confirms 200 overlapping entries, all in ESTABLISHED
state as expected.
I also did run various other tests with some permutations of the script,
to mention some: SNAT in random/random-fully/persistent mode, no zones (no
overlaps), static zones (original, reply, both directions), etc.
[1] http://thread.gmane.org/gmane.comp.security.firewalls.netfilter.devel/57412/
[2] https://paste.fedoraproject.org/242835/65657871/
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch replaces the zone id which is pushed down into functions
with the actual zone object. It's a bigger one-time change, but
needed for later on extending zones with a direction parameter, and
thus decoupling this additional information from all call-sites.
No functional changes in this patch.
The default zone becomes a global const object, namely nf_ct_zone_dflt
and will be returned directly in various cases, one being, when there's
f.e. no zoning support.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This new expression uses the nf_dup engine to clone packets to a given gateway.
Unlike xt_TEE, we use an index to indicate output interface which should be
fine at this stage.
Moreover, change to the preemtion-safe this_cpu_read(nf_skb_duplicated) from
nf_dup_ipv{4,6} to silence a lockdep splat.
Based on the original tee expression from Arturo Borrero Gonzalez, although
this patch has diverted quite a bit from this initial effort due to the
change to support maps.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Extracted from the xtables TEE target. This creates two new modules for IPv4
and IPv6 that are shared between the TEE target and the new nf_tables dup
expressions.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Quoting Daniel Borkmann:
"When adding connection tracking template rules to a netns, f.e. to
configure netfilter zones, the kernel will endlessly busy-loop as soon
as we try to delete the given netns in case there's at least one
template present, which is problematic i.e. if there is such bravery that
the priviledged user inside the netns is assumed untrusted.
Minimal example:
ip netns add foo
ip netns exec foo iptables -t raw -A PREROUTING -d 1.2.3.4 -j CT --zone 1
ip netns del foo
What happens is that when nf_ct_iterate_cleanup() is being called from
nf_conntrack_cleanup_net_list() for a provided netns, we always end up
with a net->ct.count > 0 and thus jump back to i_see_dead_people. We
don't get a soft-lockup as we still have a schedule() point, but the
serving CPU spins on 100% from that point onwards.
Since templates are normally allocated with nf_conntrack_alloc(), we
also bump net->ct.count. The issue why they are not yet nf_ct_put() is
because the per netns .exit() handler from x_tables (which would eventually
invoke xt_CT's xt_ct_tg_destroy() that drops reference on info->ct) is
called in the dependency chain at a *later* point in time than the per
netns .exit() handler for the connection tracker.
This is clearly a chicken'n'egg problem: after the connection tracker
.exit() handler, we've teared down all the connection tracking
infrastructure already, so rightfully, xt_ct_tg_destroy() cannot be
invoked at a later point in time during the netns cleanup, as that would
lead to a use-after-free. At the same time, we cannot make x_tables depend
on the connection tracker module, so that the xt_ct_tg_destroy() would
be invoked earlier in the cleanup chain."
Daniel confirms this has to do with the order in which modules are loaded or
having compiled nf_conntrack as modules while x_tables built-in. So we have no
guarantees regarding the order in which netns callbacks are executed.
Fix this by allocating the templates through kmalloc() from the respective
SYNPROXY and CT targets, so they don't depend on the conntrack kmem cache.
Then, release then via nf_ct_tmpl_free() from destroy_conntrack(). This branch
is marked as unlikely since conntrack templates are rarely allocated and only
from the configuration plane path.
Note that templates are not kept in any list to avoid further dependencies with
nf_conntrack anymore, thus, the tmpl larval list is removed.
Reported-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Tested-by: Daniel Borkmann <daniel@iogearbox.net>
Resolve compilation breakage when CONFIG_IPV6 is not set by moving the IPv6
code into a separated br_netfilter_ipv6.c file.
Fixes: efb6de9b4b ("netfilter: bridge: forward IPv6 fragmented packets")
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
In case the net_device is gone, we have to unregister the hooks and put back
the reference on the net_device object. Once it comes back, register them
again. This also covers the device rename case.
This patch also adds a new flag to indicate that the basechain is disabled, so
their hooks are not registered. This flag is used by the netdev family to
handle the case where the net_device object is gone. Currently this flag is not
exposed to userspace.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The device is part of the hook configuration, so instead of a global
configuration per table, set it to each of the basechain that we create.
This patch reworks ebddf1a8d7 ("netfilter: nf_tables: allow to bind table to
net_device").
Note that this adds a dev_name field in the nft_base_chain structure which is
required the netdev notification subscription that follows up in a patch to
handle gone net_devices.
Suggested-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch adds the internal NFT_AF_NEEDS_DEV flag to indicate that you must
attach this table to a net_device.
This change is required by the follow up patch that introduces the new netdev
table.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add a flag to mark stateful expressions.
This is used for dynamic expression instanstiation to limit the usable
expressions. Strictly speaking only the dynset expression can not be
used in order to avoid recursion, but since dynamically instantiating
non-stateful expressions will simply create an identical copy, which
behaves no differently than the original, this limits to expressions
where it actually makes sense to dynamically instantiate them.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Preparation to attach expressions to set elements: add a set extension
type to hold an expression and dump the expression information with the
set element.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add helper functions for initializing, cloning, dumping and destroying
a single expression that is not part of a rule.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
This patch changes sets to support variable sized set element keys / data
up to 64 bytes each by using variable sized set extensions. This allows
to use concatenations with bigger data items suchs as IPv6 addresses.
As a side effect, small keys/data now don't require the full 16 bytes
of struct nft_data anymore but just the space they need.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add a size argument to nft_data_init() and pass in the available space.
This will be used by the following patches to support variable sized
set element data.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Switch the nf_tables registers from 128 bit addressing to 32 bit
addressing to support so called concatenations, where multiple values
can be concatenated over multiple registers for O(1) exact matches of
multiple dimensions using sets.
The old register values are mapped to areas of 128 bits for compatibility.
When dumping register numbers, values are expressed using the old values
if they refer to the beginning of a 128 bit area for compatibility.
To support concatenations, register loads of less than a full 32 bit
value need to be padded. This mainly affects the payload and exthdr
expressions, which both unconditionally zero the last word before
copying the data.
Userspace fully passes the testsuite using both old and new register
addressing.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add helper functions to parse and dump register values in netlink attributes.
These helpers will later be changed to take care of translation between the
old 128 bit and the new 32 bit register numbers.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Simple conversion to use u32 pointers to the beginning of the data
area to keep follow up patches smaller.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Only needlessly complicates things due to requiring specific argument
types. Use memcmp directly.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Replace the array of registers passed to expressions by a struct nft_regs,
containing the verdict as a seperate member, which aliases to the
NFT_REG_VERDICT register.
This is needed to seperate the verdict from the data registers completely,
so their size can be changed.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Change nft_validate_input_register() to not only validate the input
register number, but also the length of the load, and rename it to
nft_validate_register_load() to reflect that change.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
All users of nft_validate_register_store() first invoke
nft_validate_output_register(). There is in fact no use for using it
on its own, so simplify the code by folding the functionality into
nft_validate_register_store() and kill it.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The existing name is ambiguous, data is loaded as well when we read from
a register. Rename to nft_validate_register_store() for clarity and
consistency with the upcoming patch to introduce its counterpart,
nft_validate_register_load().
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
For values spanning multiple registers, we need to validate that enough
space is available from the destination register onwards. Add a len
argument to nft_validate_data_load() and consolidate the existing length
validations in preparation of that.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add an userdata set extension and allow the user to attach arbitrary
data to set elements. This is intended to hold TLV encoded data like
comments or DNS annotations that have no meaning to the kernel.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add a new "dynset" expression for dynamic set updates.
A new set op ->update() is added which, for non existant elements,
invokes an initialization callback and inserts the new element.
For both new or existing elements the extenstion pointer is returned
to the caller to optionally perform timer updates or other actions.
Element removal is not supported so far, however that seems to be a
rather exotic need and can be added later on.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Currently a set binding is assumed to be related to a lookup and, in
case of maps, a data load.
In order to use bindings for set updates, the loop detection checks
must be restricted to map operations only. Add a flags member to the
binding struct to hold the set "action" flags such as NFT_SET_MAP,
and perform loop detection based on these.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Use atomic operations for the element count to avoid races with async
updates.
To properly handle the transactional semantics during netlink updates,
deleted but not yet committed elements are accounted for seperately and
are treated as being already removed. This means for the duration of
a netlink transaction, the limit might be exceeded by the amount of
elements deleted. Set implementations must be prepared to handle this.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
That way we don't have to reinstantiate another nf_hook_state
on the stack of the nf_reinject() path.
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for element timeouts to nft_hash. The lookup and walking
functions are changed to ignore timed out elements, a periodic garbage
collection task cleans out expired entries.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
GC is expected to happen asynchrously to the netlink interface. In the
netlink path, both insertion and removal of elements consist of two
steps, insertion followed by activation or deactivation followed by
removal, during which the element must not be freed by GC.
The synchronization helpers use an unused bit in the genmask field to
atomically mark an element as "busy", meaning it is either currently
being handled through the netlink API or by GC.
Elements being processed by GC will never survive, netlink will simply
ignore them. Elements being currently processed through netlink will be
skipped by GC and reprocessed during the next run.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add helpers for GC batch destruction: since element destruction needs
a RCU grace period for all set implementations, add some helper functions
for asynchronous batch destruction. Elements are collected in a batch
structure, which is asynchronously released using RCU once its full.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add API support for set element timeouts. Elements can have a individual
timeout value specified, overriding the sets' default.
Two new extension types are used for timeouts - the timeout value and
the expiration time. The timeout value only exists if it differs from
the default value.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add set timeout support to the netlink API. Sets with timeout support
enabled can have a default timeout value and garbage collection interval
specified.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Set elements are the last object type not supporting transaction support.
Implement similar to the existing rule transactions:
The global transaction counter keeps track of two generations, current
and next. Each element contains a bitmask specifying in which generations
it is inactive.
New elements start out as inactive in the current generation and active
in the next. On commit, the previous next generation becomes the current
generation and the element becomes active. The bitmask is then cleared
to indicate that the element is active in all future generations. If the
transaction is aborted, the element is removed from the set before it
becomes active.
When removing an element, it gets marked as inactive in the next generation.
On commit the next generation becomes active and the therefor the element
inactive. It is then taken out of then set and released. On abort, the
element is marked as active for the next generation again.
Lookups ignore elements not active in the current generation.
The current set types (hash/rbtree) both use a field in the extension area
to store the generation mask. This (currently) does not require any
additional memory since we have some free space in there.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add some helper functions for building the genmask as preparation for
set transactions.
Also add a little documentation how this stuff actually works.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Return the extension area from the ->lookup() function to allow to
consolidate common actions.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
With the conversion to set extensions, it is now possible to consolidate
the different set element destruction functions.
The set implementations' ->remove() functions are changed to only take
the element out of their internal data structures. Elements will be freed
in a batched fashion after the global transaction's completion RCU grace
period.
This reduces the amount of grace periods required for nft_hash from N
to zero additional ones, additionally this guarantees that the set
elements' extensions of all implementations can be used under RCU
protection.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The set implementations' private struct will only contain the elements
needed to maintain the search structure, all other elements are moved
to the set extensions.
Element allocation and initialization is performed centrally by
nf_tables_api instead of by the different set implementations'
->insert() functions. A new "elemsize" member in the set ops specifies
the amount of memory to reserve for internal usage. Destruction
will also be moved out of the set implementations by a following patch.
Except for element allocation, the patch is a simple conversion to
using data from the extension area.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Add simple set extension infrastructure for maintaining variable sized
and optional per element data.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The network namespace is only needed for base chains to get at the
gencursor. Also convert to possible_net_t.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Conflicts:
net/netfilter/nf_tables_core.c
The nf_tables_core.c conflict was resolved using a conflict resolution
from Stephen Rothwell as a guide.
Signed-off-by: David S. Miller <davem@davemloft.net>
Since fab4085 ("netfilter: log: nf_log_packet() as real unified
interface"), the loginfo structure that is passed to nf_log_packet() is
used to explicitly indicate the logger type you want to use.
This is a problem for people tracing rules through nfnetlink_log since
packets are always routed to the NF_LOG_TYPE logger after the
aforementioned patch.
We can fix this by removing the trace loginfo structures, but that still
changes the log level from 4 to 5 for tracing messages and there may be
someone relying on this outthere. So let's just introduce a new
nf_log_trace() function that restores the former behaviour.
Reported-by: Markus Kötter <koetter@rrzn.uni-hannover.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Having to say
> #ifdef CONFIG_NET_NS
> struct net *net;
> #endif
in structures is a little bit wordy and a little bit error prone.
Instead it is possible to say:
> typedef struct {
> #ifdef CONFIG_NET_NS
> struct net *net;
> #endif
> } possible_net_t;
And then in a header say:
> possible_net_t net;
Which is cleaner and easier to use and easier to test, as the
possible_net_t is always there no matter what the compile options.
Further this allows read_pnet and write_pnet to be functions in all
cases which is better at catching typos.
This change adds possible_net_t, updates the definitions of read_pnet
and write_pnet, updates optional struct net * variables that
write_pnet uses on to have the type possible_net_t, and finally fixes
up the b0rked users of read_pnet and write_pnet.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Daniel Borkmann