A logic error in the IPv6 Routing header parsing caused accidental
updating of the original packet buffer. The calculated extension
header lenght was set to the length field of the routing header,
causing it to be wrong.
This has 2 consequences:
1. defrag failure. As the now modified payload was used in defrag,
the decoding of the reassembled packet now contained a broken length
field for the routing header. This would lead to decoding failure.
The potential here is evasion, although it would trigger:
[1:2200014:1] SURICATA IPv6 truncated extension header
2. in IPS mode, especially the AF_PACKET mode, the modified and now
broken packet would be transmitted on the wire. It's likely that
end hosts and/or routers would reject this packet.
NFQ based IPS mode would be less affected, as it 'verdicts' based on
the packet handle. In case of replacing the packet (replace keyword
or stream normalization) it could broadcast the bad packet.
Additionally, the RH Type 0 address parsing was also broken. It too
would modify the original packet. As the result of this code was not
used anywhere else in the engine, this code is now disabled.
Reported-By: Rafael Schaefer <rschaefer@ernw.de>
AF_PACKET is not setting the engine mode to IPS when some
interfaces are peered and use IPS mode. This is due to the
fact, it is possible to peer 2 interfaces and run an IPS on
them and have a third one that is running in normal IDS mode.
In fact this choice is the bad one as unwanted side effect is
that there is no drop log and that stream inline is not used.
To fix that, this patch puts suricata in IPS mode as soon as
there is two interfaces in IPS mode. And it displays a error
message to warn user that the accuracy of detection on IDS only
interfaces will be low.
The request line scripts were added to the AMATCH list. However, there
is not AppLayerMatch function defined for lua scripts. So scripts
would not run.
This patch adds the request line scripts to the normal 'MATCH' list.
Bug #1273.
I found three somewhat serious IPv6 address bugs within the Suricata 2.0.x source code. Two are in the source module "detect-engine-address.c", and the third is in "util-radix-tree.c".
The first bug occurs within the function DetectAddressParse2(). When parsing an address string and a negated block is encountered (such as when parsing !$HOME_NET, for example), any corresponding IPv6 addresses were not getting added to the Group Heads in the DetectAddressList. Only IPv4 addresses were being added.
I discovered another bug related to IPv6 address ranges in the Signature Match Address Array comparison code for IPv6 addresses. The function DetectAddressMatchIPv6() walks a signature's source or destination match address list comparing each to the current packet's corresponding address value. The match address list consists of value pairs representing a lower and upper IP address range. If the packet's address is within that range (including equal to either the lower or upper bound), then a signature match flag is returned.
The original test of each signature match address to the packet was performed using a set of four compounded AND comparisons looking at each of the four 32-bit blocks that comprise an IPv6 address. The problem with the old comparison is that if ANY of the four 32-bit blocks failed the test, then a "no-match" was returned. This is incorrect. If one or more of the more significant 32-bit blocks met the condition, then it is a match no matter if some of the less significant 32-bit blocks did not meet the condition. Consider this example where Packet represents the packet address being checked, and Target represents the upper bound of a match address pair. We are testing if Packet is less than Target.
Packet -- 2001:0470 : 1f07:00e2 : 1988:01f1 : d468:27ab
Target -- 2001:0470 : 1f07:00e2 : a48c:2e52 : d121:101e
In this example the Packet's address is less than the target and it should give a match. However, the old code would compare each 32-bit block (shown spaced out above for clarity) and logically AND the result with the next least significant block comparison. If any of the four blocks failed the comparison, that kicked out the whole address. The flaw is illustrated above. The first two blocks are 2001:0470 and 1f07:00e2 and yield TRUE; the next less significant block is 1988:01f1 and a48c:2e52, and also yields TRUE (that is, Packet is less than Target); but the last block compare is FALSE (d468:27ab is not less than d121:101e). That last block is the least significant block, though, so its FALSE determination should not invalidate a TRUE from any of the more significant blocks. However, in the previous code using the compound logical AND block, that last least significant block would invalidate the tests done with the more significant blocks.
The other bug I found for IPv6 occurs when trying to parse and insert an IPv6 address into a Radix Tree using the function SCRadixAddKeyIPV6String(). The test for min and max values for an IPv6 CIDR mask incorrectly tests the upper limit as 32 when it should be 128 for an IPv6 address. I think this perhaps is an old copy-paste error if the IPv6 version of this function was initially copied from the corresponding IPv4 version directly above it in the code. Without this patch, the function will return null when you attempt to add an IPv6 network whose CIDR mask is larger than 32 (for example, the popular /64 mask will cause the function to return the NULL error condition).
(amended by Victor Julien)
Reported in bug 1238 is an issue where stream reassembly can be
disrupted.
A packet that was in-window, but otherwise unexpected set the
window to a really low value, causing the next *expected* packet
to be considered out of window. This lead to missing data in the
stream reassembly.
The packet was unexpected in various ways:
- it would ack unseen traffic
- it's sequence number would not match the expected next_seq
- set a really low window, while not being a proper window update
Detection however, it greatly hampered by the fact that in case of
packet loss, quite similar packets come in. Alerting in this case
is unwanted. Ignoring/skipping packets in this case as well.
The logic used in this patch is as follows. If:
- the packet is not a window update AND
- packet seq > next_seq AND
- packet acq > next_seq (packet acks unseen data) AND
- packet shrinks window more than it's own data size
THEN set event and skip the packet in the stream engine.
So in case of a segment with no data, any window shrinking is rejected.
Bug #1238.
Until now the time out handling in defrag was done using a single
uint32_t that tracked seconds. This lead to corner cases, where
defrag trackers could be timed out a little too early.
If a next header / protocol is encountered that we can't handle (yet)
set an event. Disabled the rule by default.
decode-event:ipv6.unknown_next_header;
Fix or rather implement handling of unfragmentable exthdrs in ipv6.
The exthdr(s) appearing before the frag header were copied into the
reassembled packet correctly, however the stripping of the frag header
did not work correctly.
Example:
The common case is a frag header directly after the ipv6 header:
[ipv6 header]->[frag header]->[icmpv6 (part1)]
[ipv6 header]->[frag header]->[icmpv6 (part2)]
This would result in:
[ipv6 header]->[icmpv6]
The ipv6 headers 'next header' setting would be updated to point to
whatever the frag header was pointing to.
This would also happen when is this case:
[ipv6 header]->[hop header]->[frag header]->[icmpv6 (part1)]
[ipv6 header]->[hop header]->[frag header]->[icmpv6 (part2)]
The result would be:
[ipv6 header]->[hop header]->[icmpv6]
However, here too the ipv6 header would have been updated to point
to what the frag header pointed at. So it would consider the hop header
as if it was an ICMPv6 header, or whatever the frag header pointed at.
The result is that packets would not be correctly parsed, and thus this
issue can lead to evasion.
This patch implements handling of the unfragmentable part. In the first
segment that is stored in the list for reassembly, this patch detects
unfragmentable headers and updates it to have the last unfragmentable
header point to the layer after the frag header.
Also, the ipv6 headers 'next hdr' is only updated if no unfragmentable
headers are used. If they are used, the original value is correct.
Reported-By: Rafael Schaefer <rschaefer@ernw.de>
Bug #1244.
coccinelle 1.0rc21 has a problem with regular expression handling.
This result in a Fatal Error when test system detects an coding
error.
This patch fixes the problem by using a simple blob inside
semantic patch instead of using a regular expression to define
the function.
It also fixes add an optimization on matching suppressing a
useless <.. ..> construction.
Fixes have been suggested by Julia Lawall.
AC_TRY_COMPILE puts the code in a function already, and Clang didn't like
the function within the function declaration. This lead to test failure.
Clang now properly detects __thread support.
When the config is missing, DefragPolicyGetHostTimeout will default
to returning -1. This will effectively set no timeout at all, leading
to defrag trackers being freed too early.
This patch is fixing an issue in defragmentation code. The
insertion of a fragment in the list of fragments is done with
respect to the offset of the fragment. But the code was using
the original offset of the fragment and not the one of the
new reconstructed fragment (which can be different in the
case of overlapping segment where the left part is trimmed).
This case could lead to some evasion techniques by causing
Suricata to analyse a different payload.
This patch fixes the following issue reported by valgrind:
31 errors in context 1 of 1:
Conditional jump or move depends on uninitialised value(s)
at 0x8AB2F8: UnixSocketPcapFilesCheck (runmode-unix-socket.c:279)
by 0x97725D: UnixCommandBackgroundTasks (unix-manager.c:368)
by 0x97BC52: UnixManagerThread (unix-manager.c:884)
by 0x6155F6D: start_thread (pthread_create.c:311)
by 0x6E3A9CC: clone (clone.S:113)
The running field in PcapCommand was not initialized.
This patch fixes an issue in unix socket handling. It is possible
that a socket did disconnect when analysing a command and because
the data treatment is done in a loop on clients this was leading
to a update of the list of clients during the loop. So we need
in fact to use TAILQ_FOREACH_SAFE instead of TAILQ_FOREACH.
Reported-by: Luigi Sandon <luigi.sandon@gmail.com>
Fix-suggested-by: Luigi Sandon <luigi.sandon@gmail.com>
It is possible to have a non-contiguous CPU set, which was not being
handled correctly on the TILE architecture.
Added a "rank" field in the ThreadVar to store the worker's rank separately
from the cpu for this case.
When applying wildcard thresholds (with sid = 0 and/or gid = 0) it's wrong
to exit on the first signature already having an event filter. Indeed,
doing so results in the theshold not being applied to all subsequent
signatures. Change the code in order to skip signatures with event
filters instead of breaking out of the loop.
Victor Julien
Eric Leblond
bmeeks8
sxhlinux
Alexander Gozman
sxhlinux
Ken Steele
Mats Klepsland
Alessandro Guido
Giuseppe Longo