Use the systemd class to correctly plug the package into the systemd
infrastructure.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
We already depend on systemd in DISTRO_FEATURES so adding it to
RDEPENDS is redundant. We also rdepend on two python packages, so
there's no need to explicitly depend on python3.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Use distro_features_check to check for systemd in DISTRO_FEATURES
instead of a hand-crafted python function.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
some plugins are installed into libdir/opensaf but we were making a copy
into libdir as well, this patch changes that so the packaged files are
appearing only once
create_empty_library should be using cross compiler with linker flags,
existig code in this area is not cross compile friendly
skip dev-so QA test since some .so are stubs which are packages in PN
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Do not try to compile ptests for snmp-bc if it is not in PACKGECONFIG.
Signed-off-by: Joe Slater <joe.slater@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
PKG_CONFIG is hardcoded to /usr/bin/pkg-config which is not cross
compile safe and ends up with build errors especially on hosts where
pkgconf is used it ends up with errors like
/usr/bin/pkg-config: line 11: exec: pkgconf: not found
/usr/bin/pkg-config: line 11: exec: pkgconf: not found
Override it to use own native pkg-config which can deal with sysroot
correctly
Signed-off-by: Khem Raj <raj.khem@gmail.com>
This gives users a proper error message when trying to build
a known non-building package.
netkit-rsh already had COMPATIBLE_HOST_libc-musl = 'null'.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
There are several packages which all provide /usr/lib/sendmail
when lsb is enabled. So use alternative to manage it.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
* it was upgraded to 0.2.1 in:
commit 50979151e6
Author: Oleksandr Kravchuk <open.source@oleksandr-kravchuk.com>
Date: Sat Mar 16 18:22:59 2019 +0100
igmpproxy: update to 0.2.1
* which includes _GNU_SOURCE:
https://github.com/pali/igmpproxy/commit/7d790aa2f16afad63418e573db13ea39d76ed141
* and includes sys/types.h in os-linux.h:
https://github.com/pali/igmpproxy/commit/6f92242b23178e979f4a806abe4ac9652b51c593
* and the patch doesn't apply cleanly:
ERROR: Fuzz detected:
checking file src/igmpproxy.h
Hunk #1 succeeded at 51 with fuzz 2 (offset 5 lines).
patching file src/igmpproxy.h
Hunk #1 succeeded at 51 with fuzz 2 (offset 5 lines).
The context lines in the patches can be updated with devtool:
devtool modify lib32-igmpproxy
devtool finish --force-patch-refresh lib32-igmpproxy <layer_path>
Don't forget to review changes done by devtool!
ERROR: QA Issue: Patch log indicates that patches do not apply cleanly. [patch-fuzz]
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The openhpi-fix-testfail-errors.patch tries to fix 2 saftest failures.
But the second fix is not correct. It should not return false when there
is no surrogate pairs since for the code points in UTF-16 Basic
Multilingual Plane (BMP), there is no need surrogate pairs. We should
update saftest case to fix this failure.
Also add more description for the first fix.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Install missing openhpi.conf to ptest directory to fix failure of ptest
case ohpi_035:
(process:3559): openhpid-CRITICAL **: 08:41:04.036:
../../openhpi-3.8.0/openhpid/conf.c:623: Configuration file
'../../../../openhpi-3.8.0/openhpid/t/ohpi/openhpi.conf' could not be
opened.
FAIL: ohpi_035
Also chage the configure files mode from 755 to 644.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
- refresh and remove obsolete patches
- add openssl and esi as package options
- add missing header for std::bind implementation
Signed-off-by: Andrej Valek <andrej.valek@siemens.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Backport a patch to fix 32bit build issues with
string length overflow as below:
| idbm.c:679:27: error: '%d' directive output may be truncated writing between 1 and 10 bytes into a region of size 8 [-Werror=format-truncation=]
| snprintf(passwd_len, 8, "%d", (int)strlen(value)); \
Signed-off-by: Mingli Yu <Mingli.Yu@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Copyright year has changed in COPYRIGHTS file, thus the hash change.
Signed-off-by: Pascal Bach <pascal.bach@siemens.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The ptpd2 daemon consumes 100% CPU (of a single core) after
some amount of stable runtime. This fix added minimum POSIX
timer interval to prevent from timers firing to quickly for
the process to handle, resulting in 100% CPU and endless signal queue.
Reference: https://github.com/ptpd/ptpd/blob/master/ChangeLog
Signed-off-by: Haiqing Bai <Haiqing.Bai@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
seccomp is activated by default in vsftpd and
this has caused compatibility issues with some
kernel versions. This was fixed as one can see
as https://bugzilla.redhat.com/show_bug.cgi?id=845980,
but can still cause issues with newer kernels with
kernel 4.18+.
And there is even a patch 0034-Turn-off-seccomp-sandbox-because-it-is-too-strict.patch
in fedora[https://dl.fedoraproject.org/pub/fedora/linux/releases/29/Everything/source/tree/Packages/v/vsftpd-3.0.3-28.fc29.src.rpm]
turning off seccomp sandbox for vsftpd by default
as below which means fedora doesn't limit the syscall
any more by default.
[snip]
- tunable_seccomp_sandbox = 1;
+ tunable_seccomp_sandbox = 0;
tunable_allow_writeable_chroot = 0;
tunable_accept_timeout = 60;
[snip]
Refresh 0001-vsftpd-allow-sysinfo-in-the-seccomp-sandbox.patch
to allow one more syscall getdents64 in the
seccomp sandbox apart from the previous one in
below commit:
fbffcf3f3 vsftpd: allow sysinfo() in the seccomp sandbox
before this patch:
root@qemux86-64:~# tnftp 127.0.0.1
Connected to 127.0.0.1.
220 (vsFTPd 3.0.3)
Name (127.0.0.1:root): anonymous
331 Please specify the password.
Password:
230 Login successful.
Remote system type is UNIX.
Using binary mode to transfer files.
ftp> dir
229 Entering Extended Passive Mode (|||8352|)
150 Here comes the directory listing.
500 OOPS: priv_sock_get_cmd
ftp>
after this patch:
root@qemux86-64:~# tnftp 127.0.0.1
Connected to 127.0.0.1.
220 (vsFTPd 3.0.3)
Name (127.0.0.1:root): anonymous
331 Please specify the password.
Password:
230 Login successful.
Remote system type is UNIX.
Using binary mode to transfer files.
ftp> dir
229 Entering Extended Passive Mode (|||22610|)
150 Here comes the directory listing.
226 Directory send OK.
ftp>
Reference: https://wiki.archlinux.org/index.php/Very_Secure_FTP_Daemon#vsftpd:_Error_500_with_kernel_4.18+
Signed-off-by: Mingli Yu <mingli.yu@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
We don't offer /etc/radvd.conf but only radvd.conf.example which would
cause a startup error:
Starting radvd:
* /etc/radvd.conf does not exist or is empty.
Remove update-rc.d settings to make it doesn't start by default.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The recipe wants to install a script under init.d but does not
want to it be started by default. It did so by inheriting update-rc.d
and setting INITSCRIPT_PARAMS to "remove". This is not correct.
We could just not inherit 'update-rc.d' to achieve such effect.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
When building native package, the do_compile function tries to check
libnsl.so and libresolv.so on host machine and add -lnsl and -lresolv to
SYSLIBS if they exist. But finally it will link the libnsl.so from
${STAGING_LIBDIR_NATIVE}. Actually there is no need to check them since
the libnsl2 is specified in DEPENDS and libresolv.so is from c libarary.
So add -lnsl and -lresolv to SYSLIBS directly.
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
The patch:
0001-Check-for-glibc-before-setting-CANT_USE_SEND_RECV_MS.patch
has ben integrated upstream.
update the upstream check regex
Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
F_SETLKW64 and F_SETLK64 are defined in
include/asm-generic/fcntl.h on musl target
but just including this header does not work
since both include/asm-generic/fcntl.h and include/fcntl.h
define same structures resulting in
conflicting declaration of structs.
Having local definitions of these values seems reasonable here.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
While building plugins, each <plugin>.c requires a <plugin>_init.c,
and the <plugin>_init.c is dynamically generated by makeinit.sh.
But the makeinit.sh generates all *_init.c (13 mechanism plugins,
3 auxprop plugins) at one time, if there are multiple plugins,
there will be multiple makeinit.sh invoking.
It caused a parallel issue, the *_init.c files will be generated
repeatedly.
It occasionally generate dapdb_init.c incorrectly
[snip plugins/ldapdb_init.c]
SASL_CANONUSER_PLUG_INIT( ldapdb )
SASL_CANONUSER_PLUG_INIT( ldapdb )
SASL_CANONUSER_PLUG_INIT( ldapdb )
[snip plugins/ldapdb_init.c]
Let makeinit.sh generate the expected <plugin>_init.c which
is exactly required by <plugin>.c.
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
[snip]
| powerpc-wrs-linux-gcc [snip] -I../common
|../../git/saslauthd/lak.c:58:10: fatal error: crypto-compat.h:
No such file or directory
[snip]
The crypto-compat.h locates in git/common/, it should be |
`-I../../git/common'
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
This adds a recipe for networkd-dispatcher.
It's a simple package containing a python script that needs to be
installed together with a systemd service and an example config
file. There's nothing to build.
Tested with current poky & meta-openembedded master branches.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
A previous commit set CLEANBROKEN to "1" to claim to fix the following
error.
Fixed rebuild error:
configure: error: changes in the environment can compromise the build
configure: error: run `make distclean' and/or `rm .././config.cache' and start over
configure: error: ./configure failed for saslauthd
However, I'm still seeing these errors!
The actual problem is about autotools.bbclass not cleaning things up. It just
uses 'make clean' while maybe 'make distclean' and 'rm -f ${B}/config.cache' also
needs to be there.
In fact, setting CLEANBROKEN will do no cleanup except removing some .la files.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Khem Raj
Bartosz Golaszewski
Slater, Joseph
Changqing Li
Adrian Bunk
Jan Kaisrlik
Zang Ruochen
Qi.Chen@windriver.com
Oleksandr Kravchuk
Martin Jansa
Yi Zhao
Andrej Valek
Mingli Yu
Pascal Bach
Hong Liu
Haiqing Bai
Randy MacLeod
Hongxu Jia