]> git.ipfire.org Git - thirdparty/suricata.git/log
thirdparty/suricata.git
8 years agomagic: fix broken tests after CentOS6 update 2242/head
Victor Julien [Mon, 19 Sep 2016 07:12:28 +0000 (09:12 +0200)] 
magic: fix broken tests after CentOS6 update

8 years agompls: add missing event type + rule
Victor Julien [Mon, 12 Sep 2016 16:10:51 +0000 (18:10 +0200)] 
mpls: add missing event type + rule

8 years agobpf: fix file parsing memory handling
Victor Julien [Thu, 8 Sep 2016 08:39:51 +0000 (10:39 +0200)] 
bpf: fix file parsing memory handling

Fix improper fread string handling. Improve error handling.

Skip trailing spaces for slightly more pretty printing.

Coverity CID 400763.

Thanks to Steve Grubb for helping address this issue.

8 years agodetect: don't print (null) in --list-keywords=all
Victor Julien [Thu, 1 Sep 2016 13:06:11 +0000 (15:06 +0200)] 
detect: don't print (null) in --list-keywords=all

8 years agodetect: fix setup for some keywords
Eric Leblond [Fri, 16 Sep 2016 11:34:09 +0000 (13:34 +0200)] 
detect: fix setup for some keywords

Fix problems found by siginit.cocci.

8 years agococcinelle: add siginit test
Eric Leblond [Fri, 16 Sep 2016 09:47:20 +0000 (11:47 +0200)] 
coccinelle: add siginit test

Add a test that check an inversion during keyword setup where
we add a sigmatch to a signature and then do error handling on it.
This was causing a double free of some elements and ultimately a
segfault.

Proposed-by: Victor Julien <victor@inliniac.net>
8 years agodetect-flowbits: more unittest macro usage
Jason Ish [Thu, 15 Sep 2016 17:19:12 +0000 (11:19 -0600)] 
detect-flowbits: more unittest macro usage

Also cleanup some tests by removing extra code after a test was
determined to fail.

8 years agohostbits: use new unittest macros
Jason Ish [Thu, 15 Sep 2016 16:40:22 +0000 (10:40 -0600)] 
hostbits: use new unittest macros

8 years agohostbits: fail parse on unexpected trailing data
Jason Ish [Tue, 13 Sep 2016 23:09:58 +0000 (17:09 -0600)] 
hostbits: fail parse on unexpected trailing data

Address issue https://redmine.openinfosecfoundation.org/issues/1889
for hostbits. This involves updating the regular expresssion
to capture any trailing data as the regex already keeps
spaces out of the name.

A unit test was converted to new macros to find out which
line it was failing at after updating regex.

8 years agoflowbits: validate that there are no spaces in the name
Jason Ish [Tue, 13 Sep 2016 16:57:09 +0000 (10:57 -0600)] 
flowbits: validate that there are no spaces in the name

Fixes issue: https://redmine.openinfosecfoundation.org/issues/1889

To catch the issue where the ';' is missing we have to expand the
regex to capture the whole name string, not just the leading
valid stuff. Then verify that there are no spaces in the name
(Snort has the same restriction) and fail if there is.

8 years agounix-manager: fix output of version command
Eric Leblond [Wed, 14 Sep 2016 13:09:47 +0000 (15:09 +0200)] 
unix-manager: fix output of version command

Make it consistent with the output of version command line flag.

8 years agorule-parsing: reject unescaped double quote within content section
Andreas Herz [Wed, 7 Sep 2016 20:17:01 +0000 (22:17 +0200)] 
rule-parsing: reject unescaped double quote within content section

8 years agoUpdate Changelog for 3.1.2 suricata-3.1.2
Victor Julien [Tue, 6 Sep 2016 15:52:45 +0000 (17:52 +0200)] 
Update Changelog for 3.1.2

8 years agoeve-drop: allow logging all drops 2225/head
Victor Julien [Tue, 6 Sep 2016 09:29:09 +0000 (11:29 +0200)] 
eve-drop: allow logging all drops

- drop:
    alerts: yes      # log alerts that caused drops
    flows: all       # start or all: 'start' logs only a single drop
                     # per flow direction. All logs each dropped pkt.

8 years agodns: use nonnull attr for log functions
Victor Julien [Tue, 6 Sep 2016 08:29:57 +0000 (10:29 +0200)] 
dns: use nonnull attr for log functions

8 years agodns: fix coverity warning
Victor Julien [Tue, 6 Sep 2016 08:20:38 +0000 (10:20 +0200)] 
dns: fix coverity warning

** CID 1372324:  Null pointer dereferences  (FORWARD_NULL)
/src/output-json-dns.c: 532 in OutputAnswer()

________________________________________________________________________________________________________
*** CID 1372324:  Null pointer dereferences  (FORWARD_NULL)
/src/output-json-dns.c: 532 in OutputAnswer()
526             }
527         }
528
529         /* reset */
530         MemBufferReset(aft->buffer);
531         json_object_set_new(djs, "dns", js);
>>>     CID 1372324:  Null pointer dereferences  (FORWARD_NULL)
>>>     Dereferencing null pointer "entry".
532         if (likely(DNSRRTypeEnabled(entry->type, aft->dnslog_ctx->flags))) {
533             OutputJSONBuffer(djs, aft->dnslog_ctx->file_ctx, &aft->buffer);
534         }
535         json_object_del(djs, "dns");
536
537         return;

Move checks to the top of the functions. Should be more efficient too.

8 years agodetect-template: modernize 2222/head
Victor Julien [Fri, 2 Sep 2016 10:47:11 +0000 (12:47 +0200)] 
detect-template: modernize

8 years agoapp-layer templates: cleanups
Jason Ish [Wed, 31 Aug 2016 14:24:08 +0000 (08:24 -0600)] 
app-layer templates: cleanups

- cleanup file headers
- add todo section
- convert unit tests to new macros
- add markers to remove disabled by default behaviour

8 years agodecode-icmpv6: use FAIL macros in tests
Jason Ish [Mon, 5 Sep 2016 14:15:46 +0000 (08:15 -0600)] 
decode-icmpv6: use FAIL macros in tests

8 years agounittest: FAIL macro to unconditionally fail a test
Jason Ish [Mon, 5 Sep 2016 14:15:26 +0000 (08:15 -0600)] 
unittest: FAIL macro to unconditionally fail a test

8 years agoicmpv6: fix checksum verification if fcs present
Jason Ish [Mon, 5 Sep 2016 13:41:33 +0000 (07:41 -0600)] 
icmpv6: fix checksum verification if fcs present

Calculate the length of the ICMPv6 packet from decoded information
instead of off the wire length. This will provide the correct
length if trailing data like an FCS is present.

Fixes issue:
https://redmine.openinfosecfoundation.org/issues/1849

8 years agoaffinity: fix compilation on SunOS 2219/head
Victor Julien [Mon, 5 Sep 2016 08:17:40 +0000 (10:17 +0200)] 
affinity: fix compilation on SunOS

8 years agobyteswap: fix compilation on SunOS
Victor Julien [Mon, 5 Sep 2016 08:17:16 +0000 (10:17 +0200)] 
byteswap: fix compilation on SunOS

8 years agoconfigure: detect SunOS and link against required libs
Victor Julien [Mon, 5 Sep 2016 08:16:30 +0000 (10:16 +0200)] 
configure: detect SunOS and link against required libs

8 years agothreads: provide SCGetThreadIdLong for SunOS
Victor Julien [Mon, 5 Sep 2016 08:15:52 +0000 (10:15 +0200)] 
threads: provide SCGetThreadIdLong for SunOS

8 years agodecode: declare IPPROTO_IPIP if OS doesn't have it
Victor Julien [Mon, 5 Sep 2016 08:15:23 +0000 (10:15 +0200)] 
decode: declare IPPROTO_IPIP if OS doesn't have it

8 years agodecode: fix int types
Victor Julien [Mon, 5 Sep 2016 08:14:52 +0000 (10:14 +0200)] 
decode: fix int types

8 years agoconfigure: check for strings.h: used by SunOS
Victor Julien [Mon, 5 Sep 2016 08:14:04 +0000 (10:14 +0200)] 
configure: check for strings.h: used by SunOS

8 years agologfile: resolve name clash on SunOS
Victor Julien [Mon, 5 Sep 2016 08:12:54 +0000 (10:12 +0200)] 
logfile: resolve name clash on SunOS

8 years agoeve: reduce flow_id to 51 bits 2214/head
Victor Julien [Thu, 1 Sep 2016 18:34:26 +0000 (20:34 +0200)] 
eve: reduce flow_id to 51 bits

Evebox & ELK couldn't handle the large integers. It looks like (partly)
a javascript limitation that doesn't treat 64bit ints as real ints.

8 years agoconfigure: fix Ubuntu lua pkg suggestion
Victor Julien [Thu, 1 Sep 2016 09:13:48 +0000 (11:13 +0200)] 
configure: fix Ubuntu lua pkg suggestion

8 years agoeve: output more unique flow_id 2211/head
Victor Julien [Wed, 31 Aug 2016 13:08:58 +0000 (15:08 +0200)] 
eve: output more unique flow_id

8 years agoflow: introduce function to generate flow id
Victor Julien [Wed, 31 Aug 2016 13:04:58 +0000 (15:04 +0200)] 
flow: introduce function to generate flow id

The flow id itself is not stored in the flow, but generated based on
properties that do not change during the lifetime of the flow.

As it's meant for use with the json output, it is limited to a signed
64 bit integer (int64_t) because that is the time json_integer_t uses.

8 years agodetect: implement continue detect for dcepayload 2210/head
Victor Julien [Tue, 30 Aug 2016 19:44:44 +0000 (21:44 +0200)] 
detect: implement continue detect for dcepayload

Also fix a corner case in start detection.

Bug 1853.

8 years agodetect: cleanup
Victor Julien [Tue, 30 Aug 2016 18:54:35 +0000 (20:54 +0200)] 
detect: cleanup

8 years agodetect: fix ICMP error handling issue
Victor Julien [Tue, 30 Aug 2016 17:35:18 +0000 (19:35 +0200)] 
detect: fix ICMP error handling issue

The first packet in both directions of a flow looks up the rule group
(sgh) and stores it in the flow. This makes sure the lookup doesn't
have to be performed for each packet.

ICMPv4 error messages are connected to the TCP or UDP flow they apply
to. In the case of such an ICMP error being the first packet in a
flow's direction, this would lead to an issue.

The packet would look up the rule group based on the ICMP protocol,
not based on the embedded TCP/UDP. This makes sense, as the ICMP
packet is inspected as ICMP packet. The consequence however, was that
this rule group pointer (sgh) would be stored in the flow. This is
wrong, as TCP/UDP packets that follow the ICMP packet would have no sgh
or the wrong sgh.

In normal traffic this shouldn't normally happen, but it could be
used to evade Suricata's inspection.

8 years agooutput dns: fix bit declarations 2205/head
Victor Julien [Mon, 29 Aug 2016 11:17:31 +0000 (13:17 +0200)] 
output dns: fix bit declarations

8 years agocommon: introduce macro for bit declarations
Victor Julien [Mon, 29 Aug 2016 11:14:51 +0000 (13:14 +0200)] 
common: introduce macro for bit declarations

8 years agooutput-json-dns: dns output filtering.
Tom DeCanio [Thu, 4 Feb 2016 20:19:14 +0000 (12:19 -0800)] 
output-json-dns: dns output filtering.

8 years agoeve: make logging of tagged packets optional
Jason Ish [Tue, 23 Aug 2016 17:37:42 +0000 (11:37 -0600)] 
eve: make logging of tagged packets optional

But it is enabled in the default configuration.

8 years agoeve: log tag packets as packet events
Jason Ish [Mon, 1 Aug 2016 20:01:49 +0000 (14:01 -0600)] 
eve: log tag packets as packet events

Create a new eve event type, "packet" for logging packets that
are tagged as part of an event. The packet is still at the top
level to keep it consistent with alert event types.

In addition to the packet being logged, a packet_info object
is created to hold the linktype and any future meta data
we may want to add about the packet.

8 years agodetect: minor cleanup
Victor Julien [Thu, 14 Jul 2016 10:11:02 +0000 (12:11 +0200)] 
detect: minor cleanup

8 years agodetect: minor style fixes
Victor Julien [Fri, 19 Aug 2016 11:58:07 +0000 (13:58 +0200)] 
detect: minor style fixes

8 years agodetect: minor debug output cleanup
Victor Julien [Fri, 19 Aug 2016 10:58:33 +0000 (12:58 +0200)] 
detect: minor debug output cleanup

8 years agodetect: remove unused debug code
Victor Julien [Thu, 18 Aug 2016 07:39:48 +0000 (09:39 +0200)] 
detect: remove unused debug code

8 years agodetect: minor cleanups
Victor Julien [Mon, 18 Jul 2016 08:50:29 +0000 (10:50 +0200)] 
detect: minor cleanups

8 years agostream-tcp: fix ssn returning to wrong thread pool
Victor Julien [Fri, 6 May 2016 17:44:54 +0000 (19:44 +0200)] 
stream-tcp: fix ssn returning to wrong thread pool

8 years agoapp-layer-dcerpc-udp: style cleanups
Jason Ish [Wed, 24 Aug 2016 16:14:11 +0000 (10:14 -0600)] 
app-layer-dcerpc-udp: style cleanups

- consistent 4 space indent
- cleanup file header

8 years agodetect-flowbits: fix misleading indentation
Jason Ish [Wed, 24 Aug 2016 16:15:55 +0000 (10:15 -0600)] 
detect-flowbits: fix misleading indentation

detect-flowbits.c: In function â€˜FlowBitsTestSig02’:
detect-flowbits.c:475:4: warning: this â€˜if’ clause does not guard... [-Wmisleading-indentation]
    if(error_count == 5)
    ^~
detect-flowbits.c:478:5: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the â€˜if’
     SigGroupBuild(de_ctx);
     ^~~~~~~~~~~~~

8 years agoapp-layer-dcerpc-udp: fix missleading indentation
Jason Ish [Wed, 24 Aug 2016 16:06:26 +0000 (10:06 -0600)] 
app-layer-dcerpc-udp: fix missleading indentation

app-layer-dcerpc-udp.c: In function â€˜DCERPCUDPParserTest01’:
app-layer-dcerpc-udp.c:1105:5: warning: this â€˜if’ clause does not guard... [-Wmisleading-indentation]
     if (alp_tctx != NULL)
     ^~
app-layer-dcerpc-udp.c:1107:2: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the â€˜if’
  StreamTcpFreeConfig(TRUE);
  ^~~~~~~~~~~~~~~~~~~

8 years agodecode: support Cisco Fabric Path / DCE 2203/head
Jason Ish [Sun, 14 Aug 2016 18:44:51 +0000 (12:44 -0600)] 
decode: support Cisco Fabric Path / DCE

Cisco Fabric Path is ethernet wrapped in an ethernet like header
with 2 extra bytes.  The ethernet type is in the same location
so the ethernet decoder can be used with some validation
for the extra length.

8 years agooutput: use safer logic for fingerprint printing 2200/head
Victor Julien [Wed, 24 Aug 2016 07:50:06 +0000 (09:50 +0200)] 
output: use safer logic for fingerprint printing

8 years agoapp-layer-tls: accomodate trailing \0 in hash output 2198/head
Jason Ish [Tue, 23 Aug 2016 16:53:49 +0000 (10:53 -0600)] 
app-layer-tls: accomodate trailing \0 in hash output

8 years agooutput-json-dns: allocate correct size hexstring buffer
Jason Ish [Tue, 23 Aug 2016 16:00:54 +0000 (10:00 -0600)] 
output-json-dns: allocate correct size hexstring buffer

The buffer allocated for the hexstring was not large enough
for a ':' separated hex string.

8 years agodetect: log earlier that rule reload is happening 2191/head
Victor Julien [Wed, 17 Aug 2016 14:31:05 +0000 (16:31 +0200)] 
detect: log earlier that rule reload is happening

8 years agodns: fix OOB read on malformed TXT record
Victor Julien [Wed, 17 Aug 2016 11:12:40 +0000 (13:12 +0200)] 
dns: fix OOB read on malformed TXT record

8 years agounified2: fix logging of tagged packets
Jason Ish [Tue, 16 Aug 2016 19:39:48 +0000 (13:39 -0600)] 
unified2: fix logging of tagged packets

The structure for create the alert preceding each tagged packet
was not being initialized, preventing tagged packets from being
logged.

Note: Snort unified2 does not precede tagged packets with an
alert like is done here, so this just fixes what the code
intended to do, it does not make it Snort unified2
compatible.

Address issue:
https://redmine.openinfosecfoundation.org/issues/1854

8 years agodns: fix name parsing issue leading to events 2189/head
Victor Julien [Tue, 16 Aug 2016 13:54:43 +0000 (15:54 +0200)] 
dns: fix name parsing issue leading to events

8 years agopacket: remove empty and unused UDPVars struct
Victor Julien [Tue, 16 Aug 2016 11:54:30 +0000 (13:54 +0200)] 
packet: remove empty and unused UDPVars struct

8 years agopacket: make tcp/udp/icmp vars union non-anonymous
Victor Julien [Tue, 16 Aug 2016 11:34:38 +0000 (13:34 +0200)] 
packet: make tcp/udp/icmp vars union non-anonymous

Clean the whole thing after use.

8 years agocommandline: fix strlcpy usage
Victor Julien [Mon, 15 Aug 2016 09:06:27 +0000 (11:06 +0200)] 
commandline: fix strlcpy usage

9 years agopcap: don't fail with --pcap with no device present 2183/head
Jason Ish [Fri, 29 Jul 2016 16:47:23 +0000 (10:47 -0600)] 
pcap: don't fail with --pcap with no device present

Issue: 1856.

A device with the name of "" (empty string) was being added
with LiveRegisterDevice which failed to initialize causing
Suricata to fail.

9 years agoUpdate Changelog for 3.1.1 suricata-3.1.1
Victor Julien [Wed, 13 Jul 2016 07:48:54 +0000 (09:48 +0200)] 
Update Changelog for 3.1.1

9 years agooutput lua: improve debugging output 2174/head
Victor Julien [Tue, 12 Jul 2016 15:07:25 +0000 (17:07 +0200)] 
output lua: improve debugging output

9 years agooutput lua: set proper logging progress values
Victor Julien [Tue, 12 Jul 2016 15:07:10 +0000 (17:07 +0200)] 
output lua: set proper logging progress values

9 years agooutputs: small code cleanup
Victor Julien [Tue, 12 Jul 2016 06:49:48 +0000 (08:49 +0200)] 
outputs: small code cleanup

9 years agodns: directional logging 2173/head
Jason Ish [Mon, 11 Jul 2016 17:35:51 +0000 (11:35 -0600)] 
dns: directional logging

Register loggers for to server and to client so requests
and responses can be logged independently of each other.

This results in the request log having the actual timestamp of
the request instead of the reply.

9 years agologging: setup all registered loggers for a name
Jason Ish [Mon, 11 Jul 2016 17:34:45 +0000 (11:34 -0600)] 
logging: setup all registered loggers for a name

When setting up a configured logger, do so for all registered
loggers of that name instead of just the first registered one.

This allows a logger to register itself more than once, which
can allow for independent logging of requests and replies without
touching the core transaction handling logic.

We do this so just having "dns" in the eve-log can configured
multiple "dns" loggers instead of having something like "dns-tc"
and "dns-ts" in the configuration file.

9 years agolua smtp: fix SMTPGetMimeField arg checking 2171/head
Victor Julien [Mon, 11 Jul 2016 07:30:26 +0000 (09:30 +0200)] 
lua smtp: fix SMTPGetMimeField arg checking

Properly check argument before passing it on: CID 1363385: (NULL_RETURNS)

9 years agoconfigure: require libhtp 0.5.20 2169/head
Victor Julien [Fri, 8 Jul 2016 09:26:24 +0000 (11:26 +0200)] 
configure: require libhtp 0.5.20

Ticket #1839

9 years agocmdline: fix --list-keywords and --list-app-layer-protos
Victor Julien [Fri, 8 Jul 2016 09:24:27 +0000 (11:24 +0200)] 
cmdline: fix --list-keywords and --list-app-layer-protos

Ticket #1840

9 years agooutput: don't register loggers for disabled protocols 2168/head
Victor Julien [Wed, 6 Jul 2016 14:33:20 +0000 (16:33 +0200)] 
output: don't register loggers for disabled protocols

9 years agoapp-layer: add AppLayerParserIsTxAware
Victor Julien [Wed, 6 Jul 2016 14:21:33 +0000 (16:21 +0200)] 
app-layer: add AppLayerParserIsTxAware

This function globally checks if the protocol is registered and
enabled by testing for the per alproto callback:
StateGetProgressCompletionStatus

This check is to be used before enabling Tx-aware code, like loggers.

9 years agooutput: fix debug messages
Victor Julien [Wed, 6 Jul 2016 14:08:55 +0000 (16:08 +0200)] 
output: fix debug messages

9 years agosmb: style fix in log message
Victor Julien [Wed, 6 Jul 2016 11:48:14 +0000 (13:48 +0200)] 
smb: style fix in log message

9 years agoaf-packet: improve threads selection logic
Victor Julien [Fri, 1 Jul 2016 14:03:53 +0000 (16:03 +0200)] 
af-packet: improve threads selection logic

Only use RSS queue count when cluster_qm is used. Only use core count
when cluster_flow is used.

Use a local variable to simplify the check so that we don't have to deal
with the extra flags.

9 years agolua: add smtp for detection
Victor Julien [Wed, 6 Jul 2016 11:08:02 +0000 (13:08 +0200)] 
lua: add smtp for detection

9 years agolua output: expose smtp functions to output scripts
Victor Julien [Wed, 6 Jul 2016 10:58:37 +0000 (12:58 +0200)] 
lua output: expose smtp functions to output scripts

9 years agolua: SMTPGetRcptList use position as key, not value
Victor Julien [Wed, 6 Jul 2016 10:57:22 +0000 (12:57 +0200)] 
lua: SMTPGetRcptList use position as key, not value

9 years agoLua: SMTP support; Addresses feature ticket #1775; v5
tobiass1 [Wed, 15 Jun 2016 02:28:50 +0000 (22:28 -0400)] 
Lua: SMTP support; Addresses feature ticket #1775; v5

9 years agolua: support smtp tx logging
Victor Julien [Tue, 5 Jul 2016 14:09:32 +0000 (16:09 +0200)] 
lua: support smtp tx logging

9 years agolua-output: don't crash on script setup error
Victor Julien [Tue, 5 Jul 2016 14:00:31 +0000 (16:00 +0200)] 
lua-output: don't crash on script setup error

9 years agocuda: make sure we don't use cuda in proto detect
Victor Julien [Tue, 14 Jun 2016 12:08:34 +0000 (14:08 +0200)] 
cuda: make sure we don't use cuda in proto detect

9 years agoaf-packet: minor cleanups 2164/head
Victor Julien [Tue, 21 Jun 2016 15:07:15 +0000 (17:07 +0200)] 
af-packet: minor cleanups

9 years agoaf-packet: test if fanout is supported before use
Victor Julien [Tue, 21 Jun 2016 14:43:53 +0000 (16:43 +0200)] 
af-packet: test if fanout is supported before use

Older system may pretend they can support FANOUT but then fail to
work at runtime. CentOS6 is an example of this. It would fail to
start up with the default configuration with errors like:

[15770] 21/6/2016 -- 16:00:13 - (tm-threads.c:2168) <Notice> (TmThreadWaitOnThreadInit) -- all 4 packet processing threads, 4 management threads initialized, engine started.
[15785] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1907) <Error> (AFPCreateSocket) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Coudn't set fanout mode, error Protocol not available
[15785] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1337) <Error> (ReceiveAFPLoop) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Couldn't init AF_PACKET socket, fatal error
[15770] 21/6/2016 -- 16:00:13 - (suricata.c:2664) <Notice> (main) -- Signal Received.  Stopping engine.
[15787] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1907) <Error> (AFPCreateSocket) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Coudn't set fanout mode, error Protocol not available
[15788] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1907) <Error> (AFPCreateSocket) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Coudn't set fanout mode, error Protocol not available
[15786] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1907) <Error> (AFPCreateSocket) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Coudn't set fanout mode, error Protocol not available
[15789] 21/6/2016 -- 16:00:13 - (flow-manager.c:693) <Perf> (FlowManager) -- 0 new flows, 0 established flows were timed out, 0 flows in closed state
[15787] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1337) <Error> (ReceiveAFPLoop) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Couldn't init AF_PACKET socket, fatal error
[15788] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1337) <Error> (ReceiveAFPLoop) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Couldn't init AF_PACKET socket, fatal error
[15786] 21/6/2016 -- 16:00:13 - (source-af-packet.c:1337) <Error> (ReceiveAFPLoop) -- [ERRCODE: SC_ERR_AFP_CREATE(190)] - Couldn't init AF_PACKET socket, fatal error

This patch adds a test that if run before the number of threads
is determined. If the test fails, only 1 thread is created.

9 years agonetmap: fix coverity warning 1362789
Victor Julien [Tue, 21 Jun 2016 11:21:54 +0000 (13:21 +0200)] 
netmap: fix coverity warning 1362789

** CID 1362789:  Null pointer dereferences  (FORWARD_NULL)
/src/runmode-netmap.c: 247 in ParseNetmapConfig()

________________________________________________________________________________________________________
*** CID 1362789:  Null pointer dereferences  (FORWARD_NULL)
/src/runmode-netmap.c: 247 in ParseNetmapConfig()
241         strlcpy(aconf->iface_name, iface_name, sizeof(aconf->iface_name));
242         SC_ATOMIC_INIT(aconf->ref);
243         (void) SC_ATOMIC_ADD(aconf->ref, 1);
244
245         /* Find initial node */
246         netmap_node = ConfGetNode("netmap");
>>>     CID 1362789:  Null pointer dereferences  (FORWARD_NULL)
>>>     Comparing "netmap_node" to null implies that "netmap_node" might be null.
247         if (netmap_node == NULL) {
248             SCLogInfo("Unable to find netmap config using default value");
249         } else {
250             if_root = ConfFindDeviceConfig(netmap_node, aconf->iface_name);
251             if_default = ConfFindDeviceConfig(netmap_node, "default");
252         }

9 years agoutil-threshold-config: parse suppress rules with spaces in ip list 2160/head
Andreas Herz [Tue, 14 Jun 2016 21:14:40 +0000 (23:14 +0200)] 
util-threshold-config: parse suppress rules with spaces in ip list

This modified regex allows spaces witihn the ip list for supress rules
like [10.0.0.1, 10.0.0.2]

9 years agoprivs: add capability CAP_SYS_NICE.
Jason Ish [Mon, 13 Jun 2016 18:44:28 +0000 (12:44 -0600)] 
privs: add capability CAP_SYS_NICE.

Allows the setting of thread priorities after dropping privileges.

9 years agodetect: reduce verbosity, don't warn on empty files
Victor Julien [Fri, 10 Jun 2016 07:56:23 +0000 (09:56 +0200)] 
detect: reduce verbosity, don't warn on empty files

9 years agopfring: move output to 'Perf' level
Victor Julien [Fri, 10 Jun 2016 07:31:30 +0000 (09:31 +0200)] 
pfring: move output to 'Perf' level

9 years agooffloading: work around missing TOE support
Victor Julien [Mon, 13 Jun 2016 11:32:28 +0000 (13:32 +0200)] 
offloading: work around missing TOE support

9 years agooffloading: distinguish between csum and the rest
Victor Julien [Mon, 13 Jun 2016 10:55:29 +0000 (12:55 +0200)] 
offloading: distinguish between csum and the rest

As AF_PACKET handles csum offloading don't check for this type of
offloading. Other methods like pcap and netmap do require it to be
turned off.

Improve disable command suggestion wording.

9 years agooffloading: reduce verbosity
Victor Julien [Mon, 13 Jun 2016 10:36:27 +0000 (12:36 +0200)] 
offloading: reduce verbosity

9 years agoafpacket: update offloading warning
Victor Julien [Mon, 13 Jun 2016 10:33:51 +0000 (12:33 +0200)] 
afpacket: update offloading warning

9 years agooffloading: improve checks on FreeBSD
Victor Julien [Mon, 13 Jun 2016 08:16:33 +0000 (10:16 +0200)] 
offloading: improve checks on FreeBSD

Move FreeBSD specific (but not netmap specific) checks from the netmap
code to the general ioctl wrapper code.

Warn from the check functions now, so callers no longer need to.

9 years agooffloading: move linux specific into their own func
Victor Julien [Mon, 13 Jun 2016 06:59:26 +0000 (08:59 +0200)] 
offloading: move linux specific into their own func

9 years agooffloading: check for more offloading on Linux
Victor Julien [Sun, 12 Jun 2016 08:37:28 +0000 (10:37 +0200)] 
offloading: check for more offloading on Linux

9 years agonetmap: fix enabling promisc mode on FreeBSD
Victor Julien [Sat, 11 Jun 2016 14:12:26 +0000 (16:12 +0200)] 
netmap: fix enabling promisc mode on FreeBSD

In FreeBSD setting the IFF_PROMISC flag has no effect. Instead we
need to set the IFF_PPROMISC flag.

9 years agonetmap: redo config parsing
Victor Julien [Sat, 11 Jun 2016 11:30:16 +0000 (11:30 +0000)] 
netmap: redo config parsing

Normally we parse the config per interface only. But to properly
setup the bridge, netmap also needs the config of it's peering
interface. Instead of using a complicated peering scheme like in
afpacket, simply parse the peers config too.

9 years agochangelog: update for 3.1 and add missing 3.1rc1 tickets suricata-3.1
Victor Julien [Fri, 17 Jun 2016 13:51:07 +0000 (15:51 +0200)] 
changelog: update for 3.1 and add missing 3.1rc1 tickets