The helper binary is being dropped and etherwake is enabled
for CAP_NET_RAW. This allows execution by unprivileged users
as needed by the web user interface (nobody).
Reported-by: Albert Schwarzkopf <ipfire@quitesimple.org>
Fixes: #12562
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This reverts commit f1d98a1c3f.
The new version of make seems to break building u-boot on armv5tel.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The SSH init script only kills the main daemon which leads to any child
processes (for remaining connections) being untouched.
killproc returns 4 (unknown error) when not all processes were killed
which is not intended here. Therefore we ignore the error and do not
pause the shut down process for a minute.
Fixes: #12544
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Newer kernels seem to return this in lowercase format which makes the
comparison to "EC2" fail.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
DEFAULT_GATEWAY is used only in RED_STATIC config so it
fits better to this menu and is only selectable if red
is set to static mode.
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
* Add patch to support listing of token providers.
* Add fix to proper encode python string.
Signed-off-by: Stefan Schantl <stefan.schantl@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- Update fuse from 2.9.7 to 3.10.1
- Update also required by sshfs update
- Changelog is available at https://github.com/libfuse/libfuse/releases
- Build had to be changed from autools to meson/ninja
- Rootfiles changed
- namespace conflict fix patch no longer required. Fix now built into kernel.h
Signed-off-by: Adolf Belka <ahb.ipfire@gmail.com>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
System capabilities are stored in extended file system attributes
which are by default not stored in tar balls.
This patch ensures that they are packaged and extracted.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- Update iperf3 from 3.7 to 3.9
- No changes to rootfiles
- Update patch file to remove pg flag
- Release notes from version 3.8 onwards:-
iperf 3.9 2020-08-17
--------------------
* Notable user-visible changes
* A --timestamps flag has been added, which prepends a timestamp to
each output line. An optional argument to this flag, which is a
format specification to strftime(3), allows for custom timestamp
formats (#909, #1028).
* A --server-bitrate-limit flag has been added as a server-side
command-line argument. It allows a server to enforce a maximum
throughput rate; client connections that specify a higher bitrate
or exceed this bitrate during a test will be terminated. The
bitrate is expressed in bits per second, with an optional trailing
slash and integer count that specifies an averaging interval over
which to enforce the limit (#999).
* A bug that caused increased CPU usage with the --bidir option has
been fixed (#1011).
* Notable developer-visible changes
* Fixed various minor memory leaks (#1023).
iperf 3.8.1 2020-06-10
----------------------
* Notable user-visible changes
* A regression with "make install", where the libiperf shared
library files were not getting installed, has been fixed (#1013 /
#1014).
iperf 3.8 2020-06-08
--------------------
* Notable user-visible changes
* Profiled libraries and binaries are no longer built by default
(#950).
* A minimal Dockerfile has been added (#824).
* A bug with burst mode and unlimited rate has been fixed (#898).
* Configuring with the --enable-static-bin flag will now cause
a statically-linked iperf3 binary to be built (#989).
* Configuring with the --without-sctp flag will now prevent SCTP
from being auto-detected (#1008). This flag allows building a
static binary (see above item) on a CentOS system with SCTP
installed, because no static SCTP libraries are available.
* Clock skew between the iperf3 client and server will no longer
skew the computation of jitter during UDP tests (#842 / #990).
* A possible buffer overflow in the authentication feature has been
fixed. This was only relevant when configuration authentication
using the libiperf3 API, and did not affect command-line usage.
Various other improvements and fixes in this area were also made
(#996).
* Notable developer-visible changes
* The embedded version of cJSON has been updated to 1.7.13 (#978).
* Some server authentication functions have been added to the API
(#911).
* API access has been added to the connection timeout parameter
(#1001).
* Tests for some authentication functions have been added.
* Various compiler errors and warnings have been fixed.
Signed-off-by: Adolf Belka <ahb.ipfire@gmail.com>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- update from 3.2.8 to 3.3.16
This is also an update from procps to procps-ng
The previous version was no longer being maintained.
- Added autogen.sh into lfs as ity is needed to create the config script.
- Added libdir=/lib line into configure command as default is /usr/lib
- Added mv commands for kill, ps & sysctl to place them into the same locations
as the previous version of procps
- Moved lfsmake2 procps line to after pkg-config in make.sh
The autogen line requires autoconf, libtool, gettext and pkg-config
to be available so procps moved to after them.
- procps-3.2.8-fix_unknown_HZ_value.patch no longer required with new
version so removed.
- rootfile updated.
- libprocps library being maintained by the same people now maitaining this
version of procps.
- information on the releases from 3.3.13 to 3.3.16 available on
https://gitlab.com/procps-ng/procps/-/releases
Signed-off-by: Adolf Belka <ahb.ipfire@gmail.com>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Bacula install used the bacula initscript for starting and stopping bacula.
This works fine but results in no pid or memory input in the addons table
under services.
Using the IPFire initscript also successfully starts and stops bacula with
no problems but also provides the pid and memory information in the services
addons table.
- rootfiles adjusted to remove the reference to bacula-ctl-fd
- lfs/bacula adjusted to remove the init.d/bacula link generation
remove the "rm -f /root/.rnd" command. This file is not present
and I have not seen this command in any other lfs file that I
have looked at.
- new bacula initscript created
Signed-off-by: Adolf Belka <ahb.ipfire@gmail.com>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This patch disables the output of 'iptables' in 'summary.dat' by
modifying '/usr/share/conf/logwatch.conf'.
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
When safe search is enabled, it is being enabled on YouTube, too.
This creates problems in some scenarios like schools where politics
is being tought as well as other subjects that might be censored by
YouTube (i.e. election TV spots).
Therefore it is now possible to exclude YouTube from Safe Search
but keep it enabled for the search engines.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Those fix some unintentional behaviour regarding autocompletion I
stumbled across the other day. While there seems nothing security
relevant in this, it irons out a few bugs.
The full and up-to-date list of all Bash 5.0 patches can be obtained
from https://ftp.gnu.org/gnu/bash/bash-5.0-patches/ .
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The SSH daemon was not terminated properly because killproc
tried to terminate all processes with that name. That caused
that the master daemon respawned some processed which were
therefore not killed because killproc determined a list of
PIDs only once before starting sending signals.
This patch only kills the master process which is being
determined by using sshd's pid file.
That results in all established connections not being
interrupted any more.
Furthermore, the loadproc function checks if any processes
with the given name are already running which could be true
if there are any connections still open.
That check is being disabled with the -f switch and sshd
will always be launched.
"/etc/init.d/sshd stop" might now print FAIL if only the
master process, but no connection processes were terminated.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
