]> git.ipfire.org Git - thirdparty/suricata.git/log
thirdparty/suricata.git
5 days agolib: opt-in signal handlers master 13568/head
Jason Ish [Mon, 30 Jun 2025 21:55:21 +0000 (15:55 -0600)] 
lib: opt-in signal handlers

Instead of enabling signal handlers by default, require the user of
the library to opt-in. This is done with the call to
SCEnableDefaultSignalHandlers, which sets a flag to add the default
signal handlers.

This seems like the least invasive way to do this at this time, but it
will require some re-thinking for 9.0, especially if migrate globals
to engine instances, signal handling will need to be re-thought.

Ticket: #6814

5 days agogithub-actions: bump github/codeql-action from 3.28.18 to 3.29.2
dependabot[bot] [Tue, 1 Jul 2025 19:07:53 +0000 (19:07 +0000)] 
github-actions: bump github/codeql-action from 3.28.18 to 3.29.2

Bumps [github/codeql-action](https://github.com/github/codeql-action) from 3.28.18 to 3.29.2.
- [Release notes](https://github.com/github/codeql-action/releases)
- [Commits](https://github.com/github/codeql-action/compare/v3.28.18...v3.29.2)

---
updated-dependencies:
- dependency-name: github/codeql-action
  dependency-version: 3.29.2
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
5 days agodpdk: auto threads assign one too many threads 13565/head
Lukas Sismis [Tue, 1 Jul 2025 19:43:19 +0000 (21:43 +0200)] 
dpdk: auto threads assign one too many threads

Configuration option `threads: auto` in DPDK's interface node
overassigns available threads to the interface.
Commit 4dfd44d3 changed the signedness of the remaining threads counter,
which caused surpass of the counter initialization.
The if-clause is switched to first initialize and then use the counter.

Ticket: 7798

5 days agorust/Makefile: sort filenames
Jason Ish [Mon, 30 Jun 2025 17:31:57 +0000 (11:31 -0600)] 
rust/Makefile: sort filenames

Misc cleanup as these lists are growing.

5 days agorust/htp: follow suricata versioning
Jason Ish [Mon, 30 Jun 2025 14:56:55 +0000 (08:56 -0600)] 
rust/htp: follow suricata versioning

Have htp follow Suricata versioning so we don't have to worry about
version updates as it changes.

For example, between 8.0.0-beta1 and 8.0.0-rc1 there were changes to
the htp, however the version stayed at 2.0.0 making it impossible to
publish these changes to crates.io.

5 days agodetect-bytetest: remove meaningless NULL check on data_offset
Boris Tonofa [Mon, 30 Jun 2025 13:39:47 +0000 (16:39 +0300)] 
detect-bytetest: remove meaningless NULL check on data_offset

The condition data_offset == NULL can never be true: data_offset has
already been validated as non-NULL a few lines earlier. The guard seems
to have been intended for the offset argument, yet throughout the
codebase offset is never passed as NULL. (In the unit tests, offset is
NULL, but those tests pass the value parameter as NULL, which causes the
function to return before offset is dereferenced.)

Remove the pointless check to simplify control flow and silence
static-analysis warnings.

No functional change.

Bug 7767

7 days agohttp: do not yield after failed connect if already pipelined 13557/head
Philippe Antoine [Thu, 26 Jun 2025 09:03:11 +0000 (11:03 +0200)] 
http: do not yield after failed connect if already pipelined

Ticket: 7791

7 days agodoc: add upgrade note about change in inspection
Shivani Bhardwaj [Fri, 20 Jun 2025 11:00:44 +0000 (16:30 +0530)] 
doc: add upgrade note about change in inspection

7 days agosmtp: trigger raw stream inspection
Shivani Bhardwaj [Mon, 30 Jun 2025 10:24:35 +0000 (15:54 +0530)] 
smtp: trigger raw stream inspection

Internals
---------
Suricata's stream engine returns data for inspection to the detection
engine from the stream when the chunk size is reached.

Bug
---
Inspection triggered only in the specified chunk sizes may be too late
when it comes to inspection of smaller protocol specific data which
could result in delayed inspection, incorrect data logged with a transaction
and logs misindicating the pkt that triggered an alert.

Fix
---
Fix this by making an explicit call from all respective applayer parsers to
trigger raw stream inspection which shall make the data available for inspection
in the following call of the stream engine. This needs to happen per direction
on the completion of an entity like a request or a response.

Important notes
---------------
1. The above mentioned behavior with and without this patch is
affected internally by the following conditions.
- inspection depth
- stream depth
In these special cases, the inspection window will be affected and
Suricata may not consider all the data that could be expected to be
inspected.
2. This only applies to applayer protocols running over TCP.
3. The inspection window is only considered up to the ACK'd data.
4. This entire issue is about IDS mode only.

SMTP parser can handle multiple command lines per direction. Appropriate calls
to trigger raw stream inspection have been added on succesful parsing of each
request line and response line.

For the requests, the call to trigger inspection has been added in the
beginning rather than the completion of transactions. This does not
affect the inspection as it is actually triggered in the following call.
This covers the case for anomaly as well. There are two benefits for
this:
- immediate inspection for anomalous data
- flushing of the anomalous data making next data's inspection cleaner

Bug 7783

7 days agosmtp: revert ab01a1b
Shivani Bhardwaj [Mon, 30 Jun 2025 08:58:56 +0000 (14:28 +0530)] 
smtp: revert ab01a1b

to build the correct behavior. As a part of ab01a1b, in order to match
the behavior in master, the calls for triggering raw stream inspection
were made when communication in one direction for a transaction was
completed. However, it was incorrect to do so. Reliable inspection
requires any request line/response line to be completed.

Bug 7783

7 days agosrc: remove BUG_ON in packet path
Philippe Antoine [Thu, 26 Jun 2025 09:36:36 +0000 (11:36 +0200)] 
src: remove BUG_ON in packet path

Transforms them into debug validations

Ticket: 1484

Packet path is here considered FlowWorker and all lines
covered by SV tests are here fixed

7 days agosrc: remove direct calls to abort
Philippe Antoine [Thu, 26 Jun 2025 09:16:36 +0000 (11:16 +0200)] 
src: remove direct calls to abort

use BUG_ON macros instead

Ticket: 1484

10 days agoci: test with latest rust stable version 1.88 13545/head
Philippe Antoine [Fri, 27 Jun 2025 07:21:50 +0000 (09:21 +0200)] 
ci: test with latest rust stable version 1.88

10 days agorust: allow collapsible_else_if for debug logs
Philippe Antoine [Fri, 27 Jun 2025 07:20:56 +0000 (09:20 +0200)] 
rust: allow collapsible_else_if for debug logs

see https://github.com/rust-lang/rust-clippy/issues/15158

10 days agorust/htp: fix clippy uninlined_format_args
Philippe Antoine [Fri, 27 Jun 2025 07:02:10 +0000 (09:02 +0200)] 
rust/htp: fix clippy uninlined_format_args

   --> htp/src/decompressors.rs:703:64
    |
703 |                 std::io::Error::new(std::io::ErrorKind::Other, format!("{}", e))
    |                                                                ^^^^^^^^^^^^^^^^
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#uninlined_format_args

10 days agoconfigure: remove unused strtoul check
Fupeng Zhao [Thu, 19 Jun 2025 06:17:59 +0000 (14:17 +0800)] 
configure: remove unused strtoul check

strtoul is no longer used in the codebase, so the check was removed from AC_CHECK_FUNCS.

10 days agodetect/byte: update comments to reflect current parsing logic
Fupeng Zhao [Thu, 19 Jun 2025 06:14:30 +0000 (14:14 +0800)] 
detect/byte: update comments to reflect current parsing logic

10 days agoutil/coredump: refactor parsing and respect zero core dump limit
Fupeng Zhao [Thu, 19 Jun 2025 06:03:50 +0000 (14:03 +0800)] 
util/coredump: refactor parsing and respect zero core dump limit

- Replaced strtoul/strtoull with ByteExtractString* for safer and more consistent parsing.
- Allowed max-dump to be set to 0, and correctly apply a core dump limit of 0, maintaining behavior consistent with the commented default in suricata.yaml.in.
- Added and registered unit tests to validate the updated logic.

Ticket: #7212

10 days agodetect: replace strtoul with ByteExtractStringUint32
Fupeng Zhao [Thu, 19 Jun 2025 05:48:07 +0000 (13:48 +0800)] 
detect: replace strtoul with ByteExtractStringUint32

Also added and updated unit tests to ensure correctness.

Ticket: #7212

10 days agomisc/debug: Use SCConfDump 13541/head
Jeff Lucovsky [Thu, 26 Jun 2025 14:17:45 +0000 (10:17 -0400)] 
misc/debug: Use SCConfDump

The API entrypoint for ConfDump has changed so update the #ifdef'd
code to use it.

10 days agodpdk: complete function prototype definition
Lukas Sismis [Wed, 25 Jun 2025 09:40:20 +0000 (11:40 +0200)] 
dpdk: complete function prototype definition

Ticket: 7789

10 days agoaffinity: initialize CPU sets with online CPUs only
Lukas Sismis [Wed, 25 Jun 2025 09:18:56 +0000 (11:18 +0200)] 
affinity: initialize CPU sets with online CPUs only

When no CPU set is explicitly defined, switch from
UtilCpuGetNumProcessorsConfigured() (which counts all existing CPU
cores, even offline ones) to UtilCpuGetNumProcessorsOnline() (only
the available cores).
If Suricata initializes more threads than online CPUs it oversubscribes
the system. As Suricata does not support any runtime live reconfiguration
Suricata initializes only as many cores as online CPU cores.

Ticket: 7790

10 days agodetect/analyzer: Include ICMP icode information
Jeff Lucovsky [Sun, 15 Jun 2025 14:12:46 +0000 (10:12 -0400)] 
detect/analyzer: Include ICMP icode information

Issue: 6359

Extend the rules analysis to include ICMP icode information.

10 days agodetect/analyzer: Support u8 types
Jeff Lucovsky [Sun, 15 Jun 2025 14:12:08 +0000 (10:12 -0400)] 
detect/analyzer: Support u8 types

Issue: 6359

Support JSON output of u8 types

10 days agodoc/devguide: document app-layer protocol detection
Philippe Antoine [Tue, 17 Jun 2025 12:53:31 +0000 (14:53 +0200)] 
doc/devguide: document app-layer protocol detection

Ticket: 6022

10 days agodoc: document krb5 event type
Philippe Antoine [Tue, 17 Jun 2025 12:01:42 +0000 (14:01 +0200)] 
doc: document krb5 event type

Ticket: 6566

10 days agodoc: do not have bittorrent in the middle of SMB events
Philippe Antoine [Tue, 17 Jun 2025 11:42:39 +0000 (13:42 +0200)] 
doc: do not have bittorrent in the middle of SMB events

10 days agoschema: document kerberos fields
Philippe Antoine [Tue, 17 Jun 2025 11:42:22 +0000 (13:42 +0200)] 
schema: document kerberos fields

Ticket: 6566

10 days agoci: do not run workflows for etc/schema.json
Philippe Antoine [Tue, 17 Jun 2025 19:05:19 +0000 (21:05 +0200)] 
ci: do not run workflows for etc/schema.json

As it is rather a documentation file than code

2 weeks agonfq: suppress coverity thread warning 13519/head
Victor Julien [Sat, 21 Jun 2025 19:13:35 +0000 (21:13 +0200)] 
nfq: suppress coverity thread warning

CID 1593187: (#1 of 1): Data race condition (MISSING_LOCK)
2. missing_lock: Accessing (*p).nfq_v.mark without holding lock Packet_.persistent.tunnel_lock. Elsewhere, NFQPacketVars_.mark is written to with Packet_.persistent.tunnel_lock held 2 out of 5 times (2 of these accesses strongly imply that it is necessary).

No concurrency happening on non-tunnel packet, so no locking needed.

2 weeks agoutil/var-store: suppress coverity warnings
Victor Julien [Sat, 21 Jun 2025 18:59:01 +0000 (20:59 +0200)] 
util/var-store: suppress coverity warnings

2 weeks agodetect/multi-tenant: address various thread safety warnings
Victor Julien [Sat, 21 Jun 2025 18:54:16 +0000 (20:54 +0200)] 
detect/multi-tenant: address various thread safety warnings

2 weeks agopacketpool: improve thread safety 13518/head
Victor Julien [Sat, 21 Jun 2025 14:10:47 +0000 (16:10 +0200)] 
packetpool: improve thread safety

 lock_acquire: Calling pthread_mutex_lock acquires lock PktPoolLockedStack_.mutex.
 87        SCMutexLock(&my_pool->return_stack.mutex);

CID 1554228: (#1 of 1): Indefinite wait (BAD_CHECK_OF_WAIT_COND)
dead_wait: A wait is performed without ensuring that the condition is not already satisfied while holding lock PktPoolLockedStack_.mutex. This can cause a deadlock if the notification happens before the lock is acquired.
      Acquire the lock, then check the wait condition in a loop, without releasing with the lock before the wait. This will prevent deadlocks and failed conditions from spurious wakeups.

2 weeks agodetect/loader: add threading coverity warning
Victor Julien [Sat, 21 Jun 2025 10:42:17 +0000 (12:42 +0200)] 
detect/loader: add threading coverity warning

 lock_acquire: Calling pthread_mutex_lock acquires lock ThreadVars_.ctrl_mutex.
725        SCCtrlMutexLock(th_v->ctrl_mutex);

CID 1554214: (#1 of 1): Indefinite wait (BAD_CHECK_OF_WAIT_COND)
dead_wait: A wait is performed without ensuring that the condition is not already satisfied while holding lock ThreadVars_.ctrl_mutex. This can cause a deadlock if the notification happens before the lock is acquired.
      Acquire the lock, then check the wait condition in a loop, without releasing with the lock before the wait. This will prevent deadlocks and failed conditions from spurious wakeups.

2 weeks agoflow/manager: fix threading/locking coverity warnings
Victor Julien [Thu, 19 Jun 2025 12:24:18 +0000 (14:24 +0200)] 
flow/manager: fix threading/locking coverity warnings

In flow manager and recycler timed condition wait loops.

First check loop break conditions before entiring the timed wait.

CID 1638284: (#1 of 1): Indefinite wait (BAD_CHECK_OF_WAIT_COND)
dead_wait: A wait is performed without ensuring that the condition is not already satisfied while holding lock flow_manager_ctrl_mutex. This can cause a deadlock if the notification happens before the lock is acquired.

CID 1638293: (#1 of 1): Indefinite wait (BAD_CHECK_OF_WAIT_COND)
dead_wait: A wait is performed without ensuring that the condition is not already satisfied while holding lock flow_recycler_ctrl_mutex. This can cause a deadlock if the notification happens before the lock is acquired.

2 weeks agodefrag: improve thread safety in config logging
Victor Julien [Thu, 19 Jun 2025 10:52:32 +0000 (12:52 +0200)] 
defrag: improve thread safety in config logging

CID 1554235: (#1 of 1): Data race condition (MISSING_LOCK)
missing_lock: Accessing defragtracker_spare_q.len without holding lock DefragTrackerStack_.m. Elsewhere, DefragTrackerStack_.len is written to with DefragTrackerStack_.m held 2 out of 2 times.

2 weeks agodatasets: use locking wrappers everywhere
Victor Julien [Thu, 19 Jun 2025 10:33:16 +0000 (12:33 +0200)] 
datasets: use locking wrappers everywhere

To assist coverity, which got confused:

CID 1649393: (#1 of 1): Data race condition (MISSING_LOCK)
missing_lock: Accessing sets without holding lock sets_lock. Elsewhere, sets is written to with sets_lock held 2 out of 3 times.

2 weeks agoeve/schema: map mdns properties that have keywords
Jason Ish [Fri, 20 Jun 2025 18:07:11 +0000 (12:07 -0600)] 
eve/schema: map mdns properties that have keywords

Also add descriptions for the EVE index.

2 weeks agorules: add mdns rules
Jason Ish [Fri, 20 Jun 2025 18:17:12 +0000 (12:17 -0600)] 
rules: add mdns rules

2 weeks ago.gitignore: don't ignore rule files in rules/
Jason Ish [Fri, 20 Jun 2025 18:16:36 +0000 (12:16 -0600)] 
.gitignore: don't ignore rule files in rules/

These are rules we want to track edits to, as well as new and removed
files.

2 weeks agodoc/install: use our recommended header order
Jason Ish [Fri, 20 Jun 2025 15:12:17 +0000 (09:12 -0600)] 
doc/install: use our recommended header order

2 weeks agodoc/userguide: remove example with CentOS 7
Jason Ish [Wed, 18 Jun 2025 17:19:06 +0000 (11:19 -0600)] 
doc/userguide: remove example with CentOS 7

CentOS 7 is EOL.

Ticket: #7749

2 weeks agodoc/userguide: break out package installation
Jason Ish [Wed, 18 Jun 2025 17:16:02 +0000 (11:16 -0600)] 
doc/userguide: break out package installation

Break out RPM, Debian, and Ubuntu package installation into their own
pages.

Also break out other distributions like "Arch" into an "Other" section
with a note about how those packages are not supported by the OISF.

Ticket: #6252
Ticket: #6069

2 weeks agodoc/userguide: add appendix item on eve schema
Jason Ish [Thu, 19 Jun 2025 22:00:40 +0000 (16:00 -0600)] 
doc/userguide: add appendix item on eve schema

Add some basic documentation on our EVE schema, mainly to show users
that it exists.

2 weeks agolua: don't accept a table as a return value from match
Jason Ish [Thu, 19 Jun 2025 17:57:41 +0000 (11:57 -0600)] 
lua: don't accept a table as a return value from match

Remove the half finished support for accepting a table returned from a
Lua rule's match function. This is not documented, not tested, and not
really implemented.

Also, use lua_tointeger to get the return value from the match function
as an integer instead of a float.

Ticket: #6941

2 weeks agodetect/entropy: Clarify when entropy is logged
Jeff Lucovsky [Wed, 18 Jun 2025 13:14:16 +0000 (09:14 -0400)] 
detect/entropy: Clarify when entropy is logged

Clarify when entropy values are logged and associated with non-alert log
records.

2 weeks agodecode: add stats counters for ipv4/ipv6 over ipv4
Juliana Fajardini [Wed, 11 Jun 2025 21:29:58 +0000 (18:29 -0300)] 
decode: add stats counters for ipv4/ipv6 over ipv4

These existed for ipv6 over ipv6, and ipv4 over ipv6, but not for the
ipv4 counterpart.

Task #7758

2 weeks agodecode/ipv6: set packet flow in ip-in-ip
Juliana Fajardini [Wed, 11 Jun 2025 20:54:22 +0000 (17:54 -0300)] 
decode/ipv6: set packet flow in ip-in-ip

Bug #7752

2 weeks agodetect/analyzer: Extend analyzer output with dsize value info
Jeff Lucovsky [Sun, 15 Jun 2025 15:06:58 +0000 (11:06 -0400)] 
detect/analyzer: Extend analyzer output with dsize value info

Issue: 6357

2 weeks agodatasets: remove useless NULL check 13483/head
Eric Leblond [Mon, 16 Jun 2025 09:23:43 +0000 (11:23 +0200)] 
datasets: remove useless NULL check

Coverity did detect that the cleaning code is only reached with
Dataset *set being initialized so the check is useless.

** CID 1649392:       Null pointer dereferences  (REVERSE_INULL)
/src/datasets-context-json.c: 719           in DatajsonGet()
>>>     Null-checking "set" suggests that it may be null, but it has
already been dereferenced on all paths leading to the check.

** CID 1649391:       Null pointer dereferences  (REVERSE_INULL)
/src/datasets.c: 526           in DatasetGet()
>>>     Null-checking "set" suggests that it may be null, but it has
already been dereferenced on all paths leading to the check.

2 weeks agodatasets: remove unreachable code
Eric Leblond [Thu, 12 Jun 2025 09:16:16 +0000 (11:16 +0200)] 
datasets: remove unreachable code

As the DatasetGetOrCreate function is called before the creation
of the hash, it will never be available so we can remove the
cleaning.

2 weeks agoutil/mpm: grow state queue on demand
Philippe Antoine [Tue, 17 Jun 2025 13:06:27 +0000 (15:06 +0200)] 
util/mpm: grow state queue on demand

Ticket: 7678

2 weeks agoutil/mpm: factorize code
Philippe Antoine [Tue, 17 Jun 2025 13:06:08 +0000 (15:06 +0200)] 
util/mpm: factorize code

2 weeks agompm: allocate StateQueue on the heap for ks
Philippe Antoine [Tue, 17 Jun 2025 08:35:55 +0000 (10:35 +0200)] 
mpm: allocate StateQueue on the heap for ks

Completes commit 92fce2fdc069324e6060ce046ca80da6a351573c

Ticket: 6264

2 weeks agodoc/install: windows build documentation
Jason Ish [Mon, 16 Jun 2025 18:26:39 +0000 (12:26 -0600)] 
doc/install: windows build documentation

Ticket: #5911

2 weeks agoconfigure: display if we have npcap support
Jason Ish [Mon, 16 Jun 2025 18:28:10 +0000 (12:28 -0600)] 
configure: display if we have npcap support

2 weeks agorust: fix prototype for AppLayerParserTriggerRawStreamInspection
Philippe Antoine [Tue, 17 Jun 2025 09:37:15 +0000 (11:37 +0200)] 
rust: fix prototype for AppLayerParserTriggerRawStreamInspection

Ticket: 7762

As it modifies the flow indeed

2 weeks agorust: remove unnecessary cbindgen exclusions
Philippe Antoine [Tue, 17 Jun 2025 09:26:04 +0000 (11:26 +0200)] 
rust: remove unnecessary cbindgen exclusions

2 weeks agoutil/mpm: constify arg
Philippe Antoine [Tue, 17 Jun 2025 09:18:47 +0000 (11:18 +0200)] 
util/mpm: constify arg

and make rust version use the correct integer type

Ticket: 7762

2 weeks agoutil/mtu: Avoid excessive ioctls for MTU
Jeff Lucovsky [Mon, 16 Jun 2025 14:04:48 +0000 (10:04 -0400)] 
util/mtu: Avoid excessive ioctls for MTU

Issue: 7643

Use the cached livedev MTU value, when available.

2 weeks agopcap-log: fix memory leak on error paths after SCStrdup(prefix)
Boris Tonofa [Sat, 14 Jun 2025 08:44:57 +0000 (11:44 +0300)] 
pcap-log: fix memory leak on error paths after SCStrdup(prefix)

When PcapLogDataCopy() fails after duplicating pl->prefix,
the allocated 'prefix' string was not freed, leading to a leak.

Ticket: 7759

2 weeks agowindows: use _tzname instead of tzname 13467/head
Jason Ish [Mon, 16 Jun 2025 22:34:36 +0000 (16:34 -0600)] 
windows: use _tzname instead of tzname

tzname is a POSIX variable, WIN32 has prefixed many POSIX variables
with "_". While Mingw64 supports both, UCRT64 emits a compiler warning
on the usage of "tzname".

This triggered a rather large clang-format update.

2 weeks agogithub-ci: add windows ucrt build
Jason Ish [Mon, 16 Jun 2025 21:18:19 +0000 (15:18 -0600)] 
github-ci: add windows ucrt build

MSYS2 now recommends the UCRT environment as the default, so we should
probably add a CI job for it.

2 weeks agodoc/code-style: add rust; minor cleanups
Jason Ish [Mon, 16 Jun 2025 21:09:37 +0000 (15:09 -0600)] 
doc/code-style: add rust; minor cleanups

- Add small section on Rust code, and Rust code exposed to FFI.
- Other minor cleanups.

Ticket: #7078
Ticket: #6955

2 weeks agodoc/userguide/code-style: update header ordering
Jason Ish [Mon, 16 Jun 2025 20:26:16 +0000 (14:26 -0600)] 
doc/userguide/code-style: update header ordering

To match our recommended header ordering.

3 weeks agoversion: start development towards 8.0.0 13454/head
Shivani Bhardwaj [Mon, 16 Jun 2025 05:21:50 +0000 (10:51 +0530)] 
version: start development towards 8.0.0

3 weeks agorelease: 8.0.0-rc1; update changelog suricata-8.0.0-rc1
Shivani Bhardwaj [Thu, 12 Jun 2025 14:47:08 +0000 (20:17 +0530)] 
release: 8.0.0-rc1; update changelog

3 weeks agodetect/dataset: skip adding localstatedir if fullpath is provided 13437/head
Andreas Herz [Wed, 11 Jun 2025 08:47:45 +0000 (10:47 +0200)] 
detect/dataset: skip adding localstatedir if fullpath is provided

When the option to set a full path is enabled and a full path is
provided, skip adding the prefix (based on localstatedir) to the
directory since it would be unexpected and unwanted by a user.

Ticket: 7083

3 weeks agodatajson: limit impact of feature for non user
Eric Leblond [Wed, 11 Jun 2025 12:02:19 +0000 (14:02 +0200)] 
datajson: limit impact of feature for non user

The det_ctx structure was inflated by the additoin of the array to
handle JSON context. This commit updates the code to use a growing
buffer and limit the impact.

3 weeks agodetect/alert: optimize context cleanup
Victor Julien [Wed, 11 Jun 2025 20:15:24 +0000 (22:15 +0200)] 
detect/alert: optimize context cleanup

Don't always loop over each alert, but only do so if context was used.

3 weeks agolua: update to lua 5.4.8002
Jason Ish [Wed, 11 Jun 2025 20:38:33 +0000 (14:38 -0600)] 
lua: update to lua 5.4.8002

This updated crates remove tmpnam from being linked in, removing the
warning.

3 weeks agopacket: micro optimization in packet recycling 13432/head
Victor Julien [Wed, 11 Jun 2025 17:10:41 +0000 (19:10 +0200)] 
packet: micro optimization in packet recycling

3 weeks agodetect: move large array to bottom of detect thread ctx
Victor Julien [Wed, 11 Jun 2025 17:09:53 +0000 (19:09 +0200)] 
detect: move large array to bottom of detect thread ctx

3 weeks agodetect/pcre: add needed casts
Eric Leblond [Wed, 11 Jun 2025 13:34:06 +0000 (15:34 +0200)] 
detect/pcre: add needed casts

3 weeks agodatajson: add needed cast
Eric Leblond [Wed, 11 Jun 2025 12:52:22 +0000 (14:52 +0200)] 
datajson: add needed cast

3 weeks agodatajson: fix include in header
Eric Leblond [Wed, 11 Jun 2025 07:34:03 +0000 (09:34 +0200)] 
datajson: fix include in header

3 weeks agodetect: remove some useless tests
Eric Leblond [Tue, 10 Jun 2025 18:34:38 +0000 (20:34 +0200)] 
detect: remove some useless tests

3 weeks agodoc/userguide: fix some typos
Eric Leblond [Tue, 10 Jun 2025 18:07:45 +0000 (20:07 +0200)] 
doc/userguide: fix some typos

3 weeks agopacket: optimize json context cleaning
Eric Leblond [Tue, 10 Jun 2025 09:43:12 +0000 (11:43 +0200)] 
packet: optimize json context cleaning

We don't need to recycle the full alert array. This is going to
optimize packet recycle time.

3 weeks agodecode: light refactor on packet cleaning
Eric Leblond [Tue, 10 Jun 2025 09:42:12 +0000 (11:42 +0200)] 
decode: light refactor on packet cleaning

3 weeks agodatasets: improve lock handling
Eric Leblond [Mon, 9 Jun 2025 09:45:03 +0000 (11:45 +0200)] 
datasets: improve lock handling

DatasetGetOrCreate should now be called with the dataset lock
hold so the lock management is easier to read.

3 weeks agodetect/dataset: fix locking on datajson
Eric Leblond [Mon, 9 Jun 2025 09:38:22 +0000 (11:38 +0200)] 
detect/dataset: fix locking on datajson

3 weeks agodetect/dataset: check context_key validity
Eric Leblond [Mon, 9 Jun 2025 09:30:26 +0000 (11:30 +0200)] 
detect/dataset: check context_key validity

As context_key is an user entry and as it is used to build the JSON
string of alert, we could end up with an invalid event if the string
contains improper characters.

3 weeks agoeve/alert: add comment about string format
Eric Leblond [Mon, 9 Jun 2025 09:17:25 +0000 (11:17 +0200)] 
eve/alert: add comment about string format

3 weeks agodetect/pcre: fix error handling
Eric Leblond [Mon, 9 Jun 2025 09:14:37 +0000 (11:14 +0200)] 
detect/pcre: fix error handling

3 weeks agodetect/engine: put datajson related code in a func
Eric Leblond [Mon, 9 Jun 2025 09:10:20 +0000 (11:10 +0200)] 
detect/engine: put datajson related code in a func

3 weeks agodatasets: use SCLogError for errors
Eric Leblond [Mon, 9 Jun 2025 08:50:19 +0000 (10:50 +0200)] 
datasets: use SCLogError for errors

3 weeks agodetect/engine: remove unused keyword
Eric Leblond [Mon, 9 Jun 2025 08:35:02 +0000 (10:35 +0200)] 
detect/engine: remove unused keyword

3 weeks agodatasets: comment to specify usage of field
Eric Leblond [Mon, 9 Jun 2025 08:34:31 +0000 (10:34 +0200)] 
datasets: comment to specify usage of field

3 weeks agodecode: use pointer for PacketContextData
Eric Leblond [Mon, 9 Jun 2025 08:27:16 +0000 (10:27 +0200)] 
decode: use pointer for PacketContextData

By adding a pointer instead of the structure in the alert info
we spare some memory.

3 weeks agodecode: rename ExtraDataJsonList
Eric Leblond [Mon, 9 Jun 2025 08:14:27 +0000 (10:14 +0200)] 
decode: rename ExtraDataJsonList

3 weeks agodatasets: comments about format
Eric Leblond [Mon, 9 Jun 2025 08:09:34 +0000 (10:09 +0200)] 
datasets: comments about format

3 weeks agodatasets: rename function
Eric Leblond [Mon, 9 Jun 2025 08:07:04 +0000 (10:07 +0200)] 
datasets: rename function

DatasetCreateOrGet was in fact doing get or create.

3 weeks agodatasets: fix error handling in StringSet
Eric Leblond [Mon, 9 Jun 2025 08:04:35 +0000 (10:04 +0200)] 
datasets: fix error handling in StringSet

3 weeks agodatasets: factorize set operation
Eric Leblond [Mon, 9 Jun 2025 07:59:17 +0000 (09:59 +0200)] 
datasets: factorize set operation

3 weeks agodatasjson: fix include barrier name
Eric Leblond [Mon, 9 Jun 2025 07:52:13 +0000 (09:52 +0200)] 
datasjson: fix include barrier name

3 weeks agodetect/pcre: fix copyright date
Eric Leblond [Mon, 9 Jun 2025 07:50:18 +0000 (09:50 +0200)] 
detect/pcre: fix copyright date

3 weeks agodatajson: key length fix
Eric Leblond [Sun, 8 Jun 2025 08:17:50 +0000 (10:17 +0200)] 
datajson: key length fix

When using type string, we should not limit the size of the string.

This patch also fixes some implicit cast issues.

3 weeks agodatajson: value_key is necessary in json format
Eric Leblond [Sun, 8 Jun 2025 07:28:58 +0000 (09:28 +0200)] 
datajson: value_key is necessary in json format

3 weeks agodoc/userguide: enrichment_key is now context_key
Eric Leblond [Sun, 8 Jun 2025 07:05:13 +0000 (09:05 +0200)] 
doc/userguide: enrichment_key is now context_key