Enhance IPVS to load balance SCTP transport protocol packets. This is done
based on the SCTP rfc 4960. All possible control chunks have been taken
care. The state machine used in this code looks some what lengthy. I tried
to make the state machine easy to understand.
Signed-off-by: Venkata Mohan Reddy Koppula <mohanreddykv@gmail.com>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Commit 2eff25c18c
(netfilter: xt_hashlimit: fix race condition and simplify locking)
added a mutex deadlock :
htable_create() is called with hashlimit_mutex already locked
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
with 32 bit userland and 64 bit kernels, it is unlikely but possible
that insertion of new rules fails even tough there are only about 2000
iptables rules.
This happens because the compat delta is using a short int.
Easily reproducible via "iptables -m limit" ; after about 2050
rules inserting new ones fails with -ELOOP.
Note that compat_delta included 2 bytes of padding on x86_64, so
structure size remains the same.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Normally, each connection needs a unique identity. Conntrack zones allow
to specify a numerical zone using the CT target, connections in different
zones can use the same identity.
Example:
iptables -t raw -A PREROUTING -i veth0 -j CT --zone 1
iptables -t raw -A OUTPUT -o veth1 -j CT --zone 1
Signed-off-by: Patrick McHardy <kaber@trash.net>
The error handlers might need the template to get the conntrack zone
introduced in the next patches to perform a conntrack lookup.
Signed-off-by: Patrick McHardy <kaber@trash.net>
It is one of these things that iptables cannot catch and which can
cause "Invalid argument" to be printed. Without a hint in dmesg, it is
not going to be helpful.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
call_rcu() will unconditionally reinitialize RCU head anyway.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Add TCP support, which is mandated by RFC3261 for all SIP elements.
SIP over TCP is similar to UDP, except that messages are delimited
by Content-Length: headers and multiple messages may appear in one
packet.
Signed-off-by: Patrick McHardy <kaber@trash.net>
When using TCP multiple SIP messages might be present in a single packet.
A following patch will parse them by setting the dptr to the beginning of
each message. The NAT helper needs to reload the dptr value after mangling
the packet however, so it needs to know the offset of the message to the
beginning of the packet.
Signed-off-by: Patrick McHardy <kaber@trash.net>
When requests are parsed, the "sip:" part of the SIP URI should be skipped.
Usually this doesn't matter because address parsing skips forward until after
the username part, but in case REGISTER requests it doesn't contain a username
and the address can not be parsed.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Make the output a bit more informative by showing the helper an expectation
belongs to and the expectation class.
Signed-off-by: Patrick McHardy <kaber@trash.net>
The static initial tables are pretty large, and after the net
namespace has been instantiated, they just hang around for nothing.
This commit removes them and creates tables on-demand at runtime when
needed.
Size shrinks by 7735 bytes (x86_64).
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
The respective xt_table structures already have most of the metadata
needed for hook setup. Add a 'priority' field to struct xt_table so
that xt_hook_link() can be called with a reduced number of arguments.
So should we be having more tables in the future, it comes at no
static cost (only runtime, as before) - space saved:
6807373->6806555.
Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
Rewrite COMPAT_XT_ALIGN in terms of dummy structure hack.
Compat counters logically have nothing to do with it.
Use ALIGN() macro while I'm at it for same types.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
As noticed by Jon Masters <jonathan@jonmasters.org>, the conntrack hash
size is global and not per namespace, but modifiable at runtime through
/sys/module/nf_conntrack/hashsize. Changing the hash size will only
resize the hash in the current namespace however, so other namespaces
will use an invalid hash size. This can cause crashes when enlarging
the hashsize, or false negative lookups when shrinking it.
Move the hash size into the per-namespace data and only use the global
hash size to initialize the per-namespace value when instanciating a
new namespace. Additionally restrict hash resizing to init_net for
now as other namespaces are not handled currently.
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Expectation hashtable size was simply glued to a variable with no code
to rehash expectations, so it was a bug to allow writing to it.
Make "expect_hashsize" readonly.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
nf_conntrack_cachep is currently shared by all netns instances, but
because of SLAB_DESTROY_BY_RCU special semantics, this is wrong.
If we use a shared slab cache, one object can instantly flight between
one hash table (netns ONE) to another one (netns TWO), and concurrent
reader (doing a lookup in netns ONE, 'finding' an object of netns TWO)
can be fooled without notice, because no RCU grace period has to be
observed between object freeing and its reuse.
We dont have this problem with UDP/TCP slab caches because TCP/UDP
hashtables are global to the machine (and each object has a pointer to
its netns).
If we use per netns conntrack hash tables, we also *must* use per netns
conntrack slab caches, to guarantee an object can not escape from one
namespace to another one.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
[Patrick: added unique slab name allocation]
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
As discovered by Jon Masters <jonathan@jonmasters.org>, the "untracked"
conntrack, which is located in the data section, might be accidentally
freed when a new namespace is instantiated while the untracked conntrack
is attached to a skb because the reference count it re-initialized.
The best fix would be to use a seperate untracked conntrack per
namespace since it includes a namespace pointer. Unfortunately this is
not possible without larger changes since the namespace is not easily
available everywhere we need it. For now move the untracked conntrack
initialization to the init_net setup function to make sure the reference
count is not re-initialized and handle cleanup in the init_net cleanup
function to make sure namespaces can exit properly while the untracked
conntrack is in use in other namespaces.
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a new target for the raw table, which can be used to specify conntrack
parameters for specific connections, f.i. the conntrack helper.
The target attaches a "template" connection tracking entry to the skb, which
is used by the conntrack core when initializing a new conntrack.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Support initializing selected parameters of new conntrack entries from a
"conntrack template", which is a specially marked conntrack entry attached
to the skb.
Currently the helper and the event delivery masks can be initialized this
way.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Add two masks for conntrack end expectation events to struct nf_conntrack_ecache
and use them to filter events. Their default value is "all events" when the
event sysctl is on and "no events" when it is off. A following patch will add
specific initializations. Expectation events depend on the ecache struct of
their master conntrack.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Split up the IPCT_STATUS event into an IPCT_REPLY event, which is generated
when the IPS_SEEN_REPLY bit is set, and an IPCT_ASSURED event, which is
generated when the IPS_ASSURED bit is set.
In combination with a following patch to support selective event delivery,
this can be used for "sparse" conntrack replication: start replicating the
conntrack entry after it reached the ASSURED state and that way it's SYN-flood
resistant.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Make sure not to assign a helper for a different network or transport
layer protocol to a connection.
Additionally change expectation deletion by helper to compare the name
directly - there might be multiple helper registrations using the same
name, currently one of them is chosen in an unpredictable manner and
only those expectations are removed.
Signed-off-by: Patrick McHardy <kaber@trash.net>
As noticed by Shin Hong <hongshin@gmail.com>, there is a race between
htable_find_get() and htable_put():
htable_put(): htable_find_get():
spin_lock_bh(&hashlimit_lock);
<search entry>
atomic_dec_and_test(&hinfo->use)
atomic_inc(&hinfo->use)
spin_unlock_bh(&hashlimit_lock)
return hinfo;
spin_lock_bh(&hashlimit_lock);
hlist_del(&hinfo->node);
spin_unlock_bh(&hashlimit_lock);
htable_destroy(hinfo);
The entire locking concept is overly complicated, tables are only
created/referenced and released in process context, so a single
mutex works just fine. Remove the hashinfo_spinlock and atomic
reference count and use the mutex to protect table lookups/creation
and reference count changes.
Signed-off-by: Patrick McHardy <kaber@trash.net>
The TCPMSS target is dropping SYN packets where:
1) There is data, or
2) The data offset makes the TCP header larger than the packet.
Both of these result in an error level printk. This printk has been
removed.
This change avoids dropping SYN packets containing data. If there
is also no MSS option (as well as data), one will not be added
because of possible complications due to the increased packet size.
Signed-off-by: Simon Arlott <simon@fire.lp0.eu>
Signed-off-by: Patrick McHardy <kaber@trash.net>
In a string like "v:SIP/2.0..." it was checking for !isalpha('S') when it
meant to be inspecting the ':'.
Patch by Greg Alexander <greqcs@galexander.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
This patch remove variable part from a debug message to have
message concatenation from syslog.
Signed-off-by: Eric Leblond <eric@inl.fr>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Make hashtable per-netns.
Make proc files per-netns.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Make recent table list per-netns.
Make proc files per-netns.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Simply pass hashtable to seqfile iterators, proc entry itself is not needed.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (56 commits)
sky2: Fix oops in sky2_xmit_frame() after TX timeout
Documentation/3c509: document ethtool support
af_packet: Don't use skb after dev_queue_xmit()
vxge: use pci_dma_mapping_error to test return value
netfilter: ebtables: enforce CAP_NET_ADMIN
e1000e: fix and commonize code for setting the receive address registers
e1000e: e1000e_enable_tx_pkt_filtering() returns wrong value
e1000e: perform 10/100 adaptive IFS only on parts that support it
e1000e: don't accumulate PHY statistics on PHY read failure
e1000e: call pci_save_state() after pci_restore_state()
netxen: update version to 4.0.72
netxen: fix set mac addr
netxen: fix smatch warning
netxen: fix tx ring memory leak
tcp: update the netstamp_needed counter when cloning sockets
TI DaVinci EMAC: Handle emac module clock correctly.
dmfe/tulip: Let dmfe handle DM910x except for SPARC on-board chips
ixgbe: Fix compiler warning about variable being used uninitialized
netfilter: nf_ct_ftp: fix out of bounds read in update_nl_seq()
mv643xx_eth: don't include cache padding in rx desc buffer size
...
Fix trivial conflict in drivers/scsi/cxgb3i/cxgb3i_offload.c
commit 8a27f7c90f
changed the output style of %pi4 to use fixed
width leading zero IP addresses "001.002.003.004".
It's useful when printing multiple lines of
addresses, but was a change in output style for
some existing uses.
Using %pI4 restores the previous output style.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Use the same format string as net/ipv4/netfilter/nf_nat_ftp.c
to encode an ipv4 address and port.
Both uses should be a single common function.
Signed-off-by: Joe Perches <joe@perches.com>
Acked-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
As noticed by Dan Carpenter <error27@gmail.com>, update_nl_seq()
currently contains an out of bounds read of the seq_aft_nl array
when looking for the oldest sequence number position.
Fix it to only compare valid positions.
Cc: stable@kernel.org
Signed-off-by: Patrick McHardy <kaber@trash.net>
I was very frustrated about the fact that I have to recompile the kernel
to change the hash size. So, I created this patch.
If IPVS is built-in you can append ip_vs.conn_tab_bits=?? to kernel
command line, or, if you built IPVS as modules, you can add
options ip_vs conn_tab_bits=??.
To keep everything backward compatible, you still can select the size at
compile time, and that will be used as default.
It has been about a year since this patch was originally posted
and subsequently dropped on the basis of insufficient test data.
Mark Bergsma has provided the following test results which seem
to strongly support the need for larger hash table sizes:
We do however run into the same problem with the default setting (212 =
4096 entries), as most of our LVS balancers handle around a million
connections/SLAB entries at any point in time (around 100-150 kpps
load). With only 4096 hash table entries this implies that each entry
consists of a linked list of 256 connections *on average*.
To provide some statistics, I did an oprofile run on an 2.6.31 kernel,
with both the default 4096 table size, and the same kernel recompiled
with IP_VS_CONN_TAB_BITS set to 18 (218 = 262144 entries). I built a
quick test setup with a part of Wikimedia/Wikipedia's live traffic
mirrored by the switch to the test host.
With the default setting, at ~ 120 kpps packet load we saw a typical %si
CPU usage of around 30-35%, and oprofile reported a hot spot in
ip_vs_conn_in_get:
samples % image name app name
symbol name
1719761 42.3741 ip_vs.ko ip_vs.ko ip_vs_conn_in_get
302577 7.4554 bnx2 bnx2 /bnx2
181984 4.4840 vmlinux vmlinux __ticket_spin_lock
128636 3.1695 vmlinux vmlinux ip_route_input
74345 1.8318 ip_vs.ko ip_vs.ko ip_vs_conn_out_get
68482 1.6874 vmlinux vmlinux mwait_idle
After loading the recompiled kernel with 218 entries, %si CPU usage
dropped in half to around 12-18%, and oprofile looks much healthier,
with only 7% spent in ip_vs_conn_in_get:
samples % image name app name
symbol name
265641 14.4616 bnx2 bnx2 /bnx2
143251 7.7986 vmlinux vmlinux __ticket_spin_lock
140661 7.6576 ip_vs.ko ip_vs.ko ip_vs_conn_in_get
94364 5.1372 vmlinux vmlinux mwait_idle
86267 4.6964 vmlinux vmlinux ip_route_input
[ horms@verge.net.au: trivial up-port and minor style fixes ]
Signed-off-by: Catalin(ux) M. BOIE <catab@embedromix.ro>
Cc: Mark Bergsma <mark@wikimedia.org>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
The ipvs code has a nifty system for doing the size of ioctl command
copies; it defines an array with values into which it indexes the cmd
to find the right length.
Unfortunately, the ipvs code forgot to check if the cmd was in the
range that the array provides, allowing for an index outside of the
array, which then gives a "garbage" result into the length, which
then gets used for copying into a stack buffer.
Fix this by adding sanity checks on these as well as the copy size.
[ horms@verge.net.au: adjusted limit to IP_VS_SO_GET_MAX ]
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: Patrick McHardy <kaber@trash.net>
