Daniel Golle [Fri, 4 Jul 2025 15:25:39 +0000 (16:25 +0100)]
lvm2: update to 2.03.33 and libdm to 1.02.207
LVM changes:
Version 2.03.33 - 27th June 2025
================================
Various spelling, grammar, formatting, test, and build script improvements.
Override LC_NUMERIC locale if unsuitable for json_std report format.
Repair raid arrays with transiently lost devices.
Version 2.03.32 - 05th May 2025
===============================
Lvconvert vdopool conversion properly validates acceptable LVs.
Accept thin pool data LV as cacheable LV.
Allow using zram block devices (likely for testing).
Fix lvresize when resizing COW snapshots already covering origin.
Fix lvmdbusd read of executed lvm commands output.
Fix construction of DM UUID for cachevol _cdata and _cmeta devices.
Ignore PV claims from old metadata when then PV belongs to a new VG.
Fix integrity metadata rounding.
Accept --autobackup option in pvresize.
libdm changes:
Version 1.02.207 - 27th June 2025
=================================
Escape the escape character itself on JSON report format output.
Fail dm_report_group_create if radix char from locale unsuitable for JSON_STD.
Version 1.02.206 - 05th May 2025
================================
Add support for using regex in selection criteria for string lists.
Fix string list selection when using [<item> || <item> ...].
Daniel Golle [Fri, 4 Jul 2025 15:24:35 +0000 (16:24 +0100)]
libwacom: update to 2.16.1
See release notes for changes
https://github.com/linuxwacom/libwacom/releases/tag/libwacom-2.16.0
https://github.com/linuxwacom/libwacom/releases/tag/libwacom-2.16.1
Daniel Golle [Fri, 4 Jul 2025 15:23:45 +0000 (16:23 +0100)]
libdeflate: update to 1.24
Version 1.24
------------
* The CMake-based build system now supports Apple Framework builds.
* libdeflate now builds for Windows ARM64EC.
* Made another small optimization to the x86 and ARM CRC32 code.
* Fixed a compiler warning on certain platforms (issue #416).
Version 1.23
------------
* Fixed bug introduced in 1.20 where incorrect checksums could be calculated if
libdeflate was compiled with clang at -O0 and run on a CPU supporting AVX512.
* Fixed bug introduced in 1.20 where incorrect checksums could be calculated in
rare cases on macOS computers that support AVX512 and are running an older
version of macOS that contains a bug that corrupts AVX512 registers. This
could occur only if code outside libdeflate enabled AVX512 in the thread.
* Fixed build error when using -mno-evex512 with clang 18+ or gcc 14+.
knot-resolver will attempt to use jemalloc if it's available.
Jemalloc was added to openwrt packages in PR: Add jemalloc and integrate into bind #26721 [1].
At this time, we would rather continue linking with the libc malloc,
so this PR disables jemalloc support explicitly in the knot-resolver build configuration.
,,_ -*> Snort++ <*-
o" )~ Version 3.9.1.0
'''' By Martin Roesch & The Snort Team
http://snort.org/contact#team
Copyright (C) 2014-2025 Cisco and/or its affiliates. All rights reserved.
Copyright (C) 1998-2013 Sourcefire, Inc., et al.
Using DAQ version 3.0.20
Using Hyperscan version 5.4.2 2025-06-30
Using libpcap version 1.10.5 (with TPACKET_V3)
Using LuaJIT version 2.1.0-beta3
Using LZMA version 5.6.2
Using OpenSSL 3.5.0 8 Apr 2025
Using PCRE2 version 10.42 2022-12-11
Using ZLIB version 1.3.1
Tianling Shen [Sun, 29 Jun 2025 08:48:11 +0000 (16:48 +0800)]
openlist: move from alist and update to 4.0.5
The alist was sold to a Chinese commercial company by the project owner
without any clarification. Though the company bought it nearly a half
year ago, they still cannot show a clear roadmap and privacy policy now.
They also tried to collect runtime information from the user, but the
code was not merged into the main tree "luckily".
Anyway, various behaviors from the company indicate they are untrustable,
so the other developers from the alist project have forked and made a new
project named OpenList, which will continue to be maintained by the community.
===== security note =====
The alist code up to 3.45.0 has been reviewed "clean", however the main
problem is from the author's private API that used to connect to cloud drives.
For security reasons, please reset your password and/or token if possible.
Dirk Brenken [Sat, 28 Jun 2025 20:10:34 +0000 (22:10 +0200)]
banIP: update 1.5.6-6
* limit nft logging to a rate 10/second to prevent possible log-flooding
* skip external feed processing if "allowlist-only" mode is fully enabled (in in- and outbound)
* remove needless default icmpv6 rule in wan-input
* refine the housekeeping script (uci-defaults)
* readme update
Paul Donald [Mon, 14 Apr 2025 15:12:36 +0000 (17:12 +0200)]
ddns-scripts: refactor get_current_ip()
Remove the awk based ifconfig output handling violence (omg) and instead
use the ip utility and its JSON output: this uses the available
system utility jsonfilter. Much cleaner and less brittle.
This fixes alias interface handling
Fixes issue #24922
Paul Donald [Fri, 11 Apr 2025 13:32:47 +0000 (15:32 +0200)]
ddns-scripts: refactor verify_host_port()
Leverage the resolveip utility - it does the same job that several
different resolvers do for a fraction of the resource usage. This...
to verify a host's connectivity.
resolveip dependency net delta: +2-3Kbytes
Also uses the ddns-scripts built-in 'timeout' function.
Request only 'interfaces' status, as we don't use 'connected' and
'policies' status. On my router with 5 wans / 2 tracking IPs per wan,
scrape time goes from 1.90s to 1.30s (still pretty slow but better)
hostapd exposes the `hostapd-auth` ubus object. In both
hostapd_ubus_stations.lua and hostapd_stations.lua this object is
incorrectly matched as if it were a wifi phy such as `hostapd.phy0-ap0`.
Using the literal period to find interfaces (^hostapd%.) instead of the
pattern-matching period (hostapd.) fixes this.
fix formatting consistency
Signed-off-by: Michael Payne <redacted>
[fix version conflict] Signed-off-by: Etienne Champetier <redacted>
Georgi Valkov [Fri, 13 Jun 2025 07:35:28 +0000 (10:35 +0300)]
fuse3: mount_util.c: restore symlink check
Fixes the following error when mounting iPhone using ifuse:
ifuse /mnt --container com.httpstorm.httpstorm
mount: mounting ifuse on /mnt failed: Invalid argument
Georgi Valkov [Thu, 12 Jun 2025 05:28:04 +0000 (08:28 +0300)]
fuse3: fix build warning in fuse_signals.c
BT_STACK_SZ and backtrace_buffer are not used when
HAVE_BACKTRACE is undefined. Wrap them in #ifdef
to avoid a build warning:
../lib/fuse_signals.c:31:14: warning: 'backtrace_buffer' defined but not used [-Wunused-variable]
31 | static void *backtrace_buffer[BT_STACK_SZ];
| ^~~~~~~~~~~~~~~~
Georgi Valkov [Thu, 12 Jun 2025 04:12:50 +0000 (07:12 +0300)]
fuse3: update to version 3.17.2
- Fixed uninitized bufsize value
(compilation warning and real issue when HAVE_SPLICE was not defined)
- Fixed initialization races related to buffer realocation when
large buf sizes are used
- Fixed a bug that may result in incorrect behaviour:
../util/fusermount.c:1069:17: warning: unsigned conversion from 'long long int' to 'long unsigned int' changes value from '8315462406243767374' to '1397118030' [-Woverflow]
1069 | 0x736675005346544e /* UFSD */,
| ^~~~~~~~~~~~~~~~~~
Til Kaiser [Fri, 20 Jun 2025 11:02:51 +0000 (13:02 +0200)]
rrdtool1: fix compilation with GCC 15
Currently, rrdtool1 produces the following build error
for various C files while building with GCC 15:
In file included from rrd_tool.h:93,
from rrd_cgi.c:11:
getopt.h:108:12: error: conflicting types for 'getopt'; have 'int(void)'
108 | extern int getopt ();
| ^~~~~~
In file included from include/fortify/unistd.h:22,
from rrd_tool.h:56:
include/unistd.h:127:5: note: previous declaration of 'getopt' with type 'int(int, char * const*, const char *)'
127 | int getopt(int, char * const [], const char *);
| ^~~~~~
This adds -std=gnu17 to TARGET_CFLAGS to fix the compilation error.