Antonio Pastor [Mon, 15 Sep 2025 22:38:09 +0000 (18:38 -0400)]
netatalk: update to 4.3.2
Netatalk 4.3.x adds the option to use sqlite as a CNID DB. This
is now a config option for the full package.
(mysql is also an option but this has not been included here yet).
As CNID DB backends are now managed by the netatalk meta-daemon
the init script has been updated to use it instead of starting
afpd & dbd manually.
1. Disable experimental feature: stackdump
due to issues, which were reported to upstream
2. Disabled example modules to avoid adding libstdc++.so.6 dependency
Fixes:
Package syslog-ng is missing dependencies for the following libraries:
libstdc++.so.6
Other changes
-------------
In syslog-ng 4.8.0, there was added possibility to use value "current"
as version in the config file, so use it, which confirm to use
the latest version instead of bumping the version in the file
manually.
Then in PKG_LICENSE I've included licenses which are used for headers,
source, or build system. I've also corrected PKG_LICENSE_FILES: license
texts are in the LICENSES directory, lgpl-relicensing ends with .md, and
I've added the overarching LICENSE.md.
Josef Schlehofer [Wed, 24 Sep 2025 10:27:48 +0000 (12:27 +0200)]
nft-qos: drop it as it does not work
There are several issues opened in our repository:
https://github.com/openwrt/packages/issues/16007
https://github.com/openwrt/packages/issues/19833
https://github.com/openwrt/packages/issues/20498
https://github.com/openwrt/packages/issues/20899
https://github.com/openwrt/packages/issues/24027
https://github.com/openwrt/packages/issues/24147
https://github.com/openwrt/packages/issues/24149
Unfortunately, maintainer @rosysong appears to be gone
and his domain is not working anymore.
For such reason as it is not maintained since 2021, drop it.
Josef Schlehofer [Wed, 24 Sep 2025 10:20:55 +0000 (12:20 +0200)]
dysk: drop unmaintained package here
This package was introduced in https://github.com/openwrt/packages/pull/22592
and it has not received any update in this repository despite
the upstream releases new versions.
Because, we dont have enough man power to keep it updated,
lets drop this.
sendmail: fix build on hosts with Berkley DB installed
Buildbot caught an error:
"Berkeley DB file locking needs flock() for version 5.x (and greater?)"
It is caused by leakage of host-installed Berkley DB into the build.
Since libmilter is not using the DB and because of convoluted build
process of sendmail, we do the workaround - define a macro which
prevents the error without affecting libmilter binary.
George Sapkin [Sun, 27 Apr 2025 13:28:59 +0000 (16:28 +0300)]
temperusb: drop package
- unmaintained upstream
- drop confirmed by maintainer
- last local commit:
commit: 4ca726ae02d92e0ab061c4d2b26d3b31f84b71d9
Author: Samuel Progin <redacted>
Date: Thu, 2 May 2019 21:46:30 +0200
George Sapkin [Sun, 27 Apr 2025 13:28:28 +0000 (16:28 +0300)]
dhcp-forwarder: drop package
- unmaintained upstream
- drop confirmed by maintainer
- last local commit:
commit: 4006865ae81b20b1793ae2a07db20235fefd2c71
Author: Etienne Champetier <redacted>
Date: Tue, 29 Aug 2017 21:41:14 -0700
George Sapkin [Sun, 27 Apr 2025 13:27:53 +0000 (16:27 +0300)]
intltool: drop package
- unmaintained upstream
- drop confirmed
- last local commit:
commit: cea49c620dbd53e79d206d0eba7f0dbce92e2d2f
Author: Ansuel Smith <redacted>
Date: Wed, 7 Aug 2019 13:42:26 +0200
George Sapkin [Sun, 27 Apr 2025 13:27:05 +0000 (16:27 +0300)]
pkg-config: drop package
- unmaintained upstream
- drop confirmed by maintainer
- last local commit:
commit: 243a1a13241dffc3d8da2830d825cbc535c1e33d
Author: Rosen Penev <redacted>
Date: Sat, 2 Nov 2019 11:14:50 -0700
It has been determined that the URL currently in use points to v1. The
previously used URL remains valid and is correct. If someone requires the
v1 URL, a new provider must be created.
Paul Donald [Thu, 18 Sep 2025 23:21:05 +0000 (01:21 +0200)]
ddns-scripts: stash the next check time
Calculating the next check time based on the last update time is not
very accurate if the next check is a large multiple forwards from the
last update time because the cumulative sleeps and wake times are not
exact but best effort of the OS. Other factors including clock-drift
give rise to a larger time discrepancy the further the next update is in
the future.
Stash the next check time which should be quite accurate since it's
only one sleep instance away. This is also for use in the GUI.
tunneldigger: add broker_selection option to expose load balancing capabilities
Using the broker_selection param makes it possible to decide by use (default),
always use the first available broker to connect or select a random broker
cryptography_project:cryptography has been deprecated in favour of
cryptography.io:cryptography:
https://nvd.nist.gov/products/cpe/detail/2EBA50FC-F3F9-40D5-82BD-EFB67F761153
gnuplot_project:gnuplot has been deprecated in favour of
gnuplot:gnuplot:
https://nvd.nist.gov/products/cpe/detail/DB68C9F5-3330-4749-A6F5-61FF041037CC
rom_walton:boinc has been deprecated in favour of
universityofcalifornia:boinc_client:
https://nvd.nist.gov/products/cpe/detail/DAC161C5-2154-44BF-916A-EACB524E8B8F
- Fix GCC 15 build with 2 patches from Gentoo [1][2]
- Refresh existing patch
- Extend 010-enable-nonroot-install.patch to remove "-o U -g G" from
more install targets (fixes 'invalid user buildbot' in CI)
John Audia [Thu, 4 Sep 2025 20:38:22 +0000 (16:38 -0400)]
snort3: add patch to unambiguously show vectorscan
When snort is run with the --version option, it advertises components'
versions in the output. Add a patch to modify the output to clearly
show vectorscan is in use.
John Audia [Tue, 2 Sep 2025 10:21:37 +0000 (06:21 -0400)]
snort3: replace hyperscan with vectorscan in deps
* Replacement of hyperscan-runtime reference with vectorscan-runtime
* Added support for all aarch64 targets which I believe is exhaustive
For x86 and x86/64, I found that vectorscan is truly a drop-in
replacement for hyperscan as assessed by speedtests with snort3 running
on my Intel N150 PC. CPU load during the test with each condition was
nearly saturating on a single core for both cases on a symmetrical
Gbps line.
John Audia [Tue, 2 Sep 2025 18:02:58 +0000 (14:02 -0400)]
hyperscan: remove package
Remove hyperscan since Intel announced a proprietary/closed source
license beginning with hyperscan 5.5[1,2] and a general lock of support
for the 5.4.x branch which has not seen a commit since 19-Apr-2023[3].
John Audia [Tue, 9 Sep 2025 23:41:00 +0000 (19:41 -0400)]
snort3: remove hyperscan specific patch
Drop 100-remove-HAVE_HS_COMPILE_LIT-to-work-around-upstream-b.patch as
it was only needed to fix the build against hyperscan. Vectorscan
builds fine without it.
John Audia [Sun, 21 May 2023 12:04:37 +0000 (08:04 -0400)]
vectorscan: new package for speeding up regex ops
Vectorscan is fork of Hyperscan, a high-performance multiple regex
matching library. It follows the regular expression syntax of the
commonly-used libpcre library, but is a standalone library with
its own C API.
Currently ARM NEON/ASIMD and Power VSX are 100% functional. ARM
SVE2 support is in ongoing with access to hardware now. More
platforms will follow in the future.
The performance difference of snort3 compiled against this is
sizable for aarch64 confirmed on two different SoCs:
Test SoC #1 flogic/glinet_gl-mt6000
IDS mode:
Download speed wo/ vectorscan: 91.2 ±0.21 Mbit/s (n=3)
Download speed using vectorscan: 331.0 ±27.34 Mbit/s (n=3)
Gain of 3.6x
IPS mode:
Download speed wo/ vectorscan: 30.0 ±0.06 Mbit/s (n=3)
Download speed using vectorscan: 52.9 ±0.78 Mbit/s (n=3)
Gain of 1.8x
Notes:
* Data generated on snapshot build on 12-Apr-2024 using kernel
6.6.26, snort 3.1.84.0, vectorscan 5.4.11.
* Speedtest script hitting the same server.
* Snort rules file of was 37,917 lines/22 MB.
* In all cases, single core CPU saturation occurred which
speaks to the efficiency gains supplied by vectorscan.
Test Soc #2 bcm2712/RPi5B
IPS mode:
Download speed wo/ vectorscan: 164.3 ±0.64 Mbit/s (n=3)
Download speed using vectorscan: 232.8 ±0.26 Mbit/s (n=3)
Gain of 1.4x
Notes:
* Data generated on snapshot build on 13-Apr-2024 using kernel
6.1.86, snort 3.1.84.0, vectorscan 5.4.11.
* Google fiber speedtest (https://fiber.google.com/speedtest/)
hitting the same server.
* Snort rules contained 39,801 rules/22 MB.
* In all cases, single core CPU saturation occurred which
speaks to the efficiency gains supplied by vectorscan.
David Härdeman [Thu, 18 Sep 2025 07:55:29 +0000 (09:55 +0200)]
bind: don't break IPv6 support
What started in #20183 as a attempt to clean up noise in the logfiles,
turned out to be causing denial-of-service for dual-stack and especially
IPv6-only environments.
Breaking core network functionality cannot possibly be less important
than cosmetic issues, and those affected by log spam can avoid it via
other means (e.g. "query-source-v6 none;" in named.conf).
There's no reliable heuristic for determining whether there's IPv6
connectivity at the time bind is started which will catch any and all
corner cases, as discussed in #26327.
So, remove this logic for now. If a suitable heuristic can be devised,
it can always be added in a subsequent patch, but I have my doubts.
(Also, quote one variable to make shellcheck happy)
Closes: #26327 Closes: #20468 Signed-off-by: David Härdeman <redacted>
- Remove obsolete patch
- Add patch replace NPN with ALPN for client connections \
because the default OpenSSL library does not enable NPN.
- Add newly libstdcpp dependency
Ruby 3.4.0 is a major release that introduces several changes:
- Adds `it` block parameter reference
- Switches default parser to Prism
- Implements Happy Eyeballs Version 2 in the socket library
- Improves YJIT
- Adds Modular GC
- And more (see changelog for full details)
- The project was archived on Mar 22, 2024.
- The maintainer of the package and the upstream maintainer are the
same person, who has expressed their intention not to maintain the
package/project. See the quote[1] below:
> I haven't been maintaining this and I don't plan to spend any more
> time on it. Happy to hand it off, if someone is willing to take it on.
- The latest significant commit for the package (no treewide changes)
is commit 2c71d5bcd4e09b926d63d0fbba39534fd2c587f7 from Mar 29, 2020.
- The latest upstream commit[2] is from Mar 22, 2024, but it is a
documentation or cosmetic change. After this, the latest commits[3]
are from Apr 6, 2019.
John Audia [Sun, 14 Sep 2025 12:00:10 +0000 (08:00 -0400)]
nfs-kernel-server: update to v2.8.4
Update to v2.8.4
Shortlog:
Anna Schumaker (4):
rpcctl: Add support for `rpcctl switch add-xprt`
rpcctl: Display new rpc_clnt sysfs attributes
rpcctl: Add support for the xprtsec sysfs attribute
rpcctl: Rename {read,write}_addr_file()
Antonio Alvarez Feijoo (3):
nfsroot-generator: do not fail if nfsroot is not configured
systemd: Add a generator to mount /sysroot via NFSv4 in the initrd
systemd: Allow nfs-idmapd.service to be started without the server
Scott Mayhew (3):
rpc-statd.service: define dependency on both rpcbind.service and rpcbind.socket
nfsdctl: fix lockd config during autostart
nfsdctl: debug logging fixups
Steve Dickson (3):
Release: 2.8.4
configure.ac: AC_PROG_GCC_TRADITIONAL is obsolete.
nfsdctl: Warning Clean Up
zhangyaqi (2):
gssd:fix the possible buffer overflow in get_full_hostname
nfsdcld:Fix a memory leak
While testing my integration of Snapcast, I didn't test with CONFIG_ALL*
and CONFIG_BUILDBOT flags, but the buildbots did and failed with a lot
of undefined reference to PulseAudio.
Fix it with preventing PulseAudio detection.
While at it, prevent Avahi detection too to save more flash space.
These features will be available later, through flavours.
David Härdeman [Sun, 27 Jul 2025 20:33:38 +0000 (22:33 +0200)]
bind: fix build and bump PKG_RELEASE
bind9 builds for me on 24.10, but it doesn't build on master with or without my
patches.
The build already dies on the configure stage (without my patches applied),
because the autoconf magic manages to mix up the host gcc and the
cross-compiling gcc.
Removing PKG_FIXUP:=autoreconf from the Makefile fixes that, but compilation chokes later instead on libtool magic:
make[7]: Entering directory '/home/build/openwrt/build_dir/target-arm_cortex-a9+vfpv3-d16_musl_eabi/bind-9.20.11/bin/rndc'
...
/bin/bash ../../libtool --tag=CC --mode=link arm-openwrt-linux-muslgnueabi-gcc ...
libtool: link: arm-openwrt-linux-muslgnueabi-gcc ...
.../bin/ld.bfd: warning: libns-9.20.11.so, needed by ../../lib/isccfg/.libs/libisccfg.so, not found (try using -rpath or -rpath-link)
...
collect2: error: ld returned 1 exit status
Which I did a (compile-tested only) quick and dirty fix for.
Also, BUILD_CC isn't defined anywhere in the current bind sources, so I removed
that as well.
David Härdeman [Sun, 3 Aug 2025 20:10:53 +0000 (22:10 +0200)]
isc-dhcp: create zones using rndc
This changes isc-dhcp's init script to create bind zones using the tools
bind provides for that scenario instead of crafting separate zone
configuration by hand.
At the same time, remove the use of /tmp/bind/named.conf.local and add
permissions for dynamic zone creation to bind.
David Härdeman [Sun, 27 Jul 2025 01:50:50 +0000 (03:50 +0200)]
bind: update conffiles list
The previous patches removed a number of conffiles that weren't necessary,
meaning we can now assume that any changes or additional files in /etc/bind
are things that the user wants to keep.
Since /var/lib/bind is the standard location for longer-lived zone data
(i.e. not zones that secondary servers have obtained via XFER), we symlink
it to /etc/bind/zones so that it survives a sysupgrade.
Temporary files (such as XFER:ed zones for secondaries) stay in
/var/cache/bind.
David Härdeman [Sun, 27 Jul 2025 01:02:31 +0000 (03:02 +0200)]
bind: use default rndc control channel
The present logic recreates what is already the default in bind [1], and writes
the rndc key twice to two different files. In addition, the rndc key is
regenerated every time bind is restarted. Simplify this by relying on the
default behaviour instead.
David Härdeman [Sat, 26 Jul 2025 22:39:22 +0000 (00:39 +0200)]
bind: update named.conf
First, change the "directory" to point to /var/cache/bind, which ensures that
e.g. simple file "something" statements for secondary servers will create files
there (the directory is created by the init script, so we're sure that it
exists). This behaviour matches e.g. what Debian does.
Second, remove the "auth-nxdomain" stanza which is the default in bind since
9.0 was released.
Third, change "master" to "primary" (the wording used in the bind reference
docs)
git.99rst.org / openwrt-packages.git / log