Fixed an HTTP request smuggling with certain error_page configurations
which could have allowed unauthorized web page reads.
This issue affects nginx prior to 1.17.7, so only the recipe for 1.16.1
needs the patch applied.
Fix is taken directly from
c1be55f972
Signed-off-by: Ralph Siemsen <ralph.siemsen@linaro.org>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
a) use option 7z to build the lib7z.so library
This is needed for android-tools for building fastboot
from android-tools
b) Packaged the lib7z.so and codec libraries as a part of this recipe
Fastboot RDepends on it lib7z.so
c) Fixed a C++17 forbidden error when lib7z.so is built
fixes the below error
| ../../../../CPP/7zip/Archive/Wim/WimHandler.cpp: In member function 'virtual LONG NArchive::NWim::CHandler::GetArchiveProperty(PROPID, PROPVARIANT*)':
| ../../../../CPP/7zip/Archive/Wim/WimHandler.cpp:308:11: error: use of an operand of type 'bool' in 'operator++' is forbidden in C++17
| 308 | numMethods++;
| | ^~~~~~~~~~
| ../../../../CPP/7zip/Archive/Wim/WimHandler.cpp:318:9: error: use of an operand of type 'bool' in 'operator++' is forbidden in C++17
| 318 | numMethods++;
Signed-off-by: Nisha Parrakat <Nisha.Parrakat@kpit.com>
Signed-off-by: Nisha Parrakat <nishaparrakat@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Akash Hadke <Akash.Hadke@kpit.com>
Signed-off-by: Akash Hadke <hadkeakash4@gmail.com>
(cherry picked from commit 3c36a8efe2)
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The master branch has been renamed to main in the github repo.
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Fix CVE-2021-22570.
Link: https://koji.fedoraproject.org/koji/buildinfo?buildID=1916865
Link: 394beeacb5/f/CVE-2021-22570.patch
Remove first and second hunk because the second argument in
InsertIfNotPresent() function is of type const char* const& but the
first and second hunk makes the type of second argument as const string
which is not compatible with the type of second argument in
InsertIfNotPresent().
Signed-off-by: Sana Kazi <Sana.Kazi@kpit.com>
Signed-off-by: Sana Kazi <sanakazisk19@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Without the udevrules cryptsetup luksOpen will be hanging with "Udev
cookie 0xd4de0f6 (semid 5) waiting for zero".
Signed-off-by: Khem Raj <raj.khem@gmail.com>
(cherry picked from commit 60b33e376b2331cd20950f0745336397790d2201)
Signed-off-by: Armin Kuster <akuster808@gmail.com>
(cherry picked from commit 32f1d758a1)
[Minor fixup for Dunfell]
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The `dot` tool requires to be run once after installation in order to
create its configuration file.
The do_prepare_recipe_sysroot task uses do_populate_sysroot in order to
prepare the recipe-sysroot-native. Package postinstall scripts are not
executed for -native packages, but files under ${BINDIR}/postinst-* are.
This is quite the same as graphviz-setup.sh does for nativesdk. The
general idea has been taken from
OECORE/meta/classes/pixbufcache.bbclass.
Signed-off-by: Christian Eggers <ceggers@arri.de>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
This way yocto cve-check can find open CVE's. See also:
http://lists.openembedded.org/pipermail/openembedded-core/2017-July/139897.html
"Results from cve-check are not very good at the moment.
One of the reasons for this is that component names used in CVE
database differ from yocto recipe names. This series fixes several
of those name mapping problems by setting the CVE_PRODUCT correctly
in the recipes. To check this mapping with after a build, I'm exporting
LICENSE and CVE_PRODUCT variables to buildhistory for recipes and
packages."
Value added is based on:
https://nvd.nist.gov/products/cpe/search/results?keyword=netcat&status=FINAL&orderBy=CPEURI&namingFormat=2.3
Signed-off-by: Andre Carvalho <andrestc@fb.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Sana Kazi <sanakazisk19@gmail.com>
Signed-off-by: Sana Kazi <Sana.Kazi@kpit.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Changelog:
==========
*) SECURITY: CVE-2021-44790: Possible buffer overflow when parsing
multipart content in mod_lua of Apache HTTP Server 2.4.51 and
earlier (cve.mitre.org)
A carefully crafted request body can cause a buffer overflow in
the mod_lua multipart parser (r:parsebody() called from Lua
scripts).
The Apache httpd team is not aware of an exploit for the
vulnerabilty though it might be possible to craft one.
This issue affects Apache HTTP Server 2.4.51 and earlier.
*) SECURITY: CVE-2021-44224: Possible NULL dereference or SSRF in
forward proxy configurations in Apache HTTP Server 2.4.51 and
earlier (cve.mitre.org)
A crafted URI sent to httpd configured as a forward proxy
(ProxyRequests on) can cause a crash (NULL pointer dereference)
or, for configurations mixing forward and reverse proxy
declarations, can allow for requests to be directed to a
declared Unix Domain Socket endpoint (Server Side Request
Forgery).
This issue affects Apache HTTP Server 2.4.7 up to 2.4.51
(included).
*) http: Enforce that fully qualified uri-paths not to be forward-proxied
have an http(s) scheme, and that the ones to be forward proxied have a
hostname, per HTTP specifications.
*) OpenSSL autoconf detection improvement: pick up openssl.pc in the
specified openssl path.
*) mod_proxy_connect, mod_proxy: Do not change the status code after we
already sent it to the client.
*) mod_http: Correctly sent a 100 Continue status code when sending an interim
response as result of an Expect: 100-Continue in the request and not the
current status code of the request. PR 65725
*) mod_dav: Some DAV extensions, like CalDAV, specify both document
elements and property elements that need to be taken into account
when generating a property. The document element and property element
are made available in the dav_liveprop_elem structure by calling
dav_get_liveprop_element().
*) mod_dav: Add utility functions dav_validate_root_ns(),
dav_find_child_ns(), dav_find_next_ns(), dav_find_attr_ns() and
dav_find_attr() so that other modules get to play too.
*) mpm_event: Restart stopping of idle children after a load peak. PR 65626.
*) mod_http2: fixes 2 regressions in server limit handling.
1. When reaching server limits, such as MaxRequestsPerChild, the
HTTP/2 connection send a GOAWAY frame much too early on new
connections, leading to invalid protocol state and a client
failing the request. See PR65731.
The module now initializes the HTTP/2 protocol correctly and
allows the client to submit one request before the shutdown
via a GOAWAY frame is being announced.
2. A regression in v1.15.24 was fixed that could lead to httpd
child processes not being terminated on a graceful reload or
when reaching MaxConnectionsPerChild. When unprocessed h2
requests were queued at the time, these could stall.
See <https://github.com/icing/mod_h2/issues/212>.
*) mod_ssl: Add build support for OpenSSL v3.
*) mod_proxy_connect: Honor the smallest of the backend or client timeout
while tunneling.
*) mod_proxy: SetEnv proxy-nohalfclose (or alike) allows to disable TCP
half-close forwarding when tunneling protocols.
*) core: Be safe with ap_lingering_close() called with a socket NULL-ed by
a third-party module. PR 65627.
*) mod_md: Fix memory leak in case of failures to load the private key.
PR 65620
*) mod_md: adding v2.4.8 with the following changes
- Added support for ACME External Account Binding (EAB).
Use the new directive `MDExternalAccountBinding` to provide the
server with the value for key identifier and hmac as provided by
your CA.
While working on some servers, EAB handling is not uniform
across CAs. First tests with a Sectigo Certificate Manager in
demo mode are successful. But ZeroSSL, for example, seems to
regard EAB values as a one-time-use-only thing, which makes them
fail if you create a seconde account or retry the creation of the
first account with the same EAB.
- The directive 'MDCertificateAuthority' now checks if its parameter
is a http/https url or one of a set of known names. Those are
'LetsEncrypt', 'LetsEncrypt-Test', 'Buypass' and 'Buypass-Test'
for now and they are not case-sensitive.
The default of LetsEncrypt is unchanged.
- `MDContactEmail` can now be specified inside a `<MDomain dnsname>`
section.
- Treating 401 HTTP status codes for orders like 403, since some ACME
servers seem to prefer that for accessing oders from other accounts.
- When retrieving certificate chains, try to read the repsonse even
if the HTTP Content-Type is unrecognized.
- Fixed a bug that reset the error counter of a certificate renewal
and prevented the increasing delays in further attempts.
- Fixed the renewal process giving up every time on an already existing
order with some invalid domains. Now, if such are seen in a previous
order, a new order is created for a clean start over again.
See <https://github.com/icing/mod_md/issues/268>
- Fixed a mixup in md-status handler when static certificate files
and renewal was configured at the same time.
*) mod_md: values for External Account Binding (EAB) can
now also be configured to be read from a separate JSON
file. This allows to keep server configuration permissions
world readable without exposing secrets.
*) mod_proxy_uwsgi: Remove duplicate slashes at the beginning of PATH_INFO.
PR 65616.
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
(cherry picked from commit ea76fc6437)
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Backport a patch to fix build against glibc 2.34 (e.g. on Fedora 35)
Fixes:
| In file included from attr_clnt.c:88:
| /usr/include/unistd.h:363:13: error: conflicting types for
‘closefrom’; have ‘void(int)’
| 363 | extern void closefrom (int __lowfd) __THROW;
| | ^~~~~~~~~
| In file included from attr_clnt.c:87:
| ./sys_defs.h:1506:12: note: previous declaration of ‘closefrom’ with
type ‘int(int)’
| 1506 | extern int closefrom(int);
| | ^~~~~~~~~
Signed-off-by: Yi Zhao <yi.zhao@windriver.com>
Add patch to fix CVE-2021-43527 which causes heap overflow in nss.
Signed-off-by: Sana Kazi <Sana.Kazi@kpit.com>
Signed-off-by: Sana Kazi <sanakazisk19@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Port patch to fix CVE-2020-8927 for brotli from Debian Buster
CVE: CVE-2020-8927
Signed-off-by: Jan Kraemer <jan@spectrejan.de>
[Fixup to apply with URL changes]
Signed-off-by: Armin Kuster <akuster808@gmail.com>
According to the upstream [1], the bug happens only if the programmer
does not follow the API definition.
[1] https://github.com/akheron/jansson/issues/548
Signed-off-by: Marta Rybczynska <marta.rybczynska@huawei.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
* with PTEST_ENABLED it enables with-tests PACKAGECONFIG which
instead of using system googletest gmock, tries to fetch googletest
from github and fails because branch was recently renamed from master to main
| -- Found PkgConfig: /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/recipe-sysroot-native/usr/bin/pkg-config (found version "0.29.2")
| -- Checking for module 'libsystemd>=236'
| -- Found libsystemd, version 249
| -- Building with tests
| Fetching googletest...
| [1/9] Creating directories for 'googletest-populate'
| [1/9] Performing download step (git clone) for 'googletest-populate'
| Cloning into 'googletest-src'...
| fatal: invalid reference: master
| CMake Error at googletest-subbuild/googletest-populate-prefix/tmp/googletest-populate-gitclone.cmake:40 (message):
| Failed to checkout tag: 'master'
|
|
| FAILED: googletest-populate-prefix/src/googletest-populate-stamp/googletest-populate-download
| cd /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/build/_deps && /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/recipe-sysroot-native/usr/bin/cmake -P /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/build/_deps/googletest-subbuild/googletest-populate-prefix/tmp/googletest-populate-gitclone.cmake && /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/recipe-sysroot-native/usr/bin/cmake -E touch /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/build/_deps/googletest-subbuild/googletest-populate-prefix/src/googletest-populate-stamp/googletest-populate-download
| ninja: build stopped: subcommand failed.
|
| CMake Error at /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/recipe-sysroot-native/usr/share/cmake-3.19/Modules/FetchContent.cmake:989 (message):
| Build step for googletest failed: 1
| Call Stack (most recent call first):
| /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/recipe-sysroot-native/usr/share/cmake-3.19/Modules/FetchContent.cmake:1118:EVAL:2 (__FetchContent_directPopulate)
| /OE/tmp-glibc/work/qemux86-oe-linux/sdbus-c++/0.8.3-r0/recipe-sysroot-native/usr/share/cmake-3.19/Modules/FetchContent.cmake:1118 (cmake_language)
| tests/CMakeLists.txt:17 (FetchContent_Populate)
|
|
| -- Configuring incomplete, errors occurred!
* unfortunately this backported patch fixes the fetching failure, because
it uses release-${GOOGLETEST_VERSION} tag instead of now non-existent
master branch, but is not enough to prevent fetching from github during
do_configure:
-- Building with tests
-- Could NOT find GTest (missing: GTest_DIR)
-- Checking for module 'gmock>=1.10.0'
-- No package 'gmock' found
Fetching googletest...
we also need to add googletest dependency to with-tests PACKAGECONFIG was fixed in meta-oe/master with the upgrade to 1.0.0:
b26b66e5da (diff-48a847e7323703994fd2ce0fcb731ff860fa955a77cdfe39d71a9cc84a042c06L15)
then it's ok and not fetching:
-- Building with tests
-- Looking for pthread.h
-- Looking for pthread.h - found
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
lmsensors will build isadump and isaset only on x86 architecture.
Depending on this package breaks lmsensors on all non-x86 machines. Fix
this by enabling ${PN}-isatools dependency only on x86.
Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
This patch updates SRC_URIs using git to include branch=master if no branch is set
and also to use protocol=https for github urls as generated by the conversion script
in OE-Core.
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Updates the python3-fasteners package considering that the 0.15 version
doesn't support Python 3.8, only Python 3.5 or less. The 0.16.3 version
support Python 3.8.
The non-support of Python 3.8 results in the malfunction of many layers
in the dunfell branch.
Signed-off-by: Maíra Canal <maira.canal@usp.br>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Since it uses multiple fetch URIs make it explicit to define SRCREV_FORMAT
Signed-off-by: Andreas Weger <weger@hs-mittweida.de>
Change-Id: Id1d0a1062d09f690123b2a1c06137ae5c04d7b20
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Since it uses multiple fetch URIs make it explicit to define SRCREV_FORMAT
Signed-off-by: Andreas Weger <weger@hs-mittweida.de>
Change-Id: Ib24fce16b3986a465f1c5854166b8f28446b5186
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Since it uses multiple fetch URIs make it explicit to define SRCREV_FORMAT
Signed-off-by: Andreas Weger <weger@hs-mittweida.de>
Change-Id: I062eb971a83594315cc674ccb6eba67a14d5656f
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Since it uses multiple fetch URIs make it explicit to define SRCREV_FORMAT
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Since it uses multiple fetch URIs make it explicit to define SRCREV_FORMAT
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Cherry-pick the following patch from upstream/master branch,
as this fixes the following error(s) seen while running the
'pm-qa' scripts on the targets:
cpufreq_01.sh: line 28: ../include/functions.sh: No such file or directory
-----------------------------------------------
A commit in the repo of pm-qa:
"adf9df9 Fix path to library files and change shebang line"
Changed the text that sed was using to replace relative to
absolute paths.
As a result sed was not effectively finding the text
"source ../include" to replace it, as the sed should be now
searching for ". ../include".
Similarly for "../Switches"
Signed-off-by: Anastasios Kavoukis <anastasios.kavoukis@arm.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
(cherry picked from commit 06a93a04ef)
Signed-off-by: Bhupesh Sharma <bhupesh.sharma@linaro.org>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Bug fix only updates. see: https://raw.githubusercontent.com/antirez/redis/5.0/00-RELEASENOTES
Including these cves:
5.0.14
Security Fixes:
* (CVE-2021-41099) Integer to heap buffer overflow handling certain string
commands and network payloads, when proto-max-bulk-len is manually configured
to a non-default, very large value [reported by yiyuaner].
* (CVE-2021-32762) Integer to heap buffer overflow issue in redis-cli and
redis-sentinel parsing large multi-bulk replies on some older and less common
platforms [reported by Microsoft Vulnerability Research].
* (CVE-2021-32687) Integer to heap buffer overflow with intsets, when
set-max-intset-entries is manually configured to a non-default, very large
value [reported by Pawel Wieczorkiewicz, AWS].
* (CVE-2021-32675) Denial Of Service when processing RESP request payloads with
a large number of elements on many connections.
* (CVE-2021-32672) Random heap reading issue with Lua Debugger [reported by
Meir Shpilraien].
* (CVE-2021-32628) Integer to heap buffer overflow handling ziplist-encoded
data types, when configuring a large, non-default value for
hash-max-ziplist-entries, hash-max-ziplist-value, zset-max-ziplist-entries
or zset-max-ziplist-value [reported by sundb].
* (CVE-2021-32627) Integer to heap buffer overflow issue with streams, when
configuring a non-default, large value for proto-max-bulk-len and
client-query-buffer-limit [reported by sundb].
* (CVE-2021-32626) Specially crafted Lua scripts may result with Heap buffer
overflow [reported by Meir Shpilraien].
5.0.11
Integer overflow on 32-bit systems (CVE-2021-21309):
Redis 4.0 or newer uses a configurable limit for the maximum supported bulk
input size. By default, it is 512MB which is a safe value for all platforms.
If the limit is significantly increased, receiving a large request from a client
may trigger several integer overflow scenarios, which would result with buffer
overflow and heap corruption.
5.0.10
This release fixes a potential heap overflow when using a heap allocator other
than jemalloc or glibc's malloc. See:
https://github.com/redis/redis/pull/7963
Signed-off-by: Armin Kuster <akuster808@gmail.com>
