libevent
libevent copied to clipboard
monotonic_prc_fallback failure
After a brief look, I would say that these failure because util/monotonic_prc_fallback
depends on time, but we should do something with this eventually to avoid false-positive failures/reports.
BTW in which env did you test? (OS and other related details)
>uname -a
Darwin mac 17.7.0 Darwin Kernel Version 17.7.0: Thu Dec 20 21:47:19 PST 2018; root:xnu-4570.71.22~1/RELEASE_X86_64 x86_64
>autoconf --version | head -1
autoconf (GNU Autoconf) 2.69
>automake --version | head -1
automake (GNU automake) 1.16.1
>/opt/llvm/bin/clang --version
clang version 7.0.0 (tags/RELEASE_700/final)
Target: x86_64-apple-darwin17.7.0
Thread model: posix
InstalledDir: /opt/llvm/bin
From the shell:
CC="/opt/llvm/bin/clang"; # gcc
CMAKE_C_COMPILER="/opt/llvm/bin/clang";
CPP="/opt/llvm/bin/clang -E"; # cpp
CXX="/opt/llvm/bin/clang++"; # g++
CMAKE_CXX_COMPILER="/opt/llvm/bin/clang++"; # g++
AR="/opt/llvm/bin/llvm-ar"; # ar
CMAKE_AR="/opt/llvm/bin/llvm-ar"; # ar
CMAKE_LINKER="/usr/bin/ld -v"; # ld
CLANG_DEFAULT_LINKER="$CMAKE_LINKER";
NM="/opt/llvm/bin/llvm-nm"; # nm
CMAKE_NM="/opt/llvm/bin/llvm-nm"; # nm
CMAKE_OBJDUMP="/opt/llvm/bin/llvm-objdump";
RANLIB="/opt/llvm/bin/llvm-ranlib";
CMAKE_RANLIB="/opt/llvm/bin/llvm-ranlib";
CFLAGS="$CFLAGS -arch x86_64";
LDFLAGS="$LDFLAGS -Wl,-z,relro,-z,now";
CFLAGS="$CFLAGS -fstack-protector-all";
CFLAGS="$CFLAGS -fno-omit-frame-pointer";
CFLAGS="$CFLAGS -fPIC -fPIE";
LDFLAGS="$LDFLAGS -pie";
CFLAGS="$CFLAGS -O2";
CPPFLAGS="$CPPFLAGS -D_FORTIFY_SOURCE=2";
CFLAGS="$CFLAGS -Wformat -Wformat-security -Werror=format-security";
libevent:
config_options="--prefix=$prefix \
--mandir=$man \
--docdir=$doc \
--enable-function-sections \
--disable-openssl \
--enable-gcc-hardening \
--disable-largefile \
--disable-dependency-tracking \
--disable-samples \
--disable-debug-mode";
I have name servers and PF running. I closed them all, and run the tests again, just in case. The test results did not change.
I think you want a debug trace. Tell me what you need.
I still see timeouts:
[msg] Nameserver 127.0.0.1:64954 has failed: request timed out. [msg] All nameservers have failed
This is not an error per se (this is just tests that uses timeouts)
It is the monotonic_prc_fallback from test/regress_util.c that is failing.
[...]
regress:
FAIL test/regress_util.c:1374: assert(diff.tv_sec == 0): 1 vs 0util/monotonic_prc_fallback:
[monotonic_prc_fallback FAILED]
[warn] Trying to disable lock functions after they have been set up will probaby not work.
[warn] Trying to disable lock functions after they have been set up will probaby not work.
[msg] Nameserver 127.0.0.1:51316 has failed: request timed out.
[msg] All nameservers have failed
[msg] Nameserver 127.0.0.1:61798 has failed: request timed out.
[msg] All nameservers have failed
[msg] Nameserver 127.0.0.1:56829 has failed: request timed out.
[msg] All nameservers have failed
[msg] Nameserver 127.0.0.1:65024 has failed: request timed out.
[msg] All nameservers have failed
[msg] Nameserver 127.0.0.1:59416 has failed: request timed out.
[msg] All nameservers have failed
[warn] Error from accept() call: Invalid argument
1/287 TESTS FAILED. (33 skipped)
FAILED
regress_debug:
FAIL test/regress_util.c:1374: assert(diff.tv_sec == 0): 1 vs 0util/monotonic_prc_fallback:
[monotonic_prc_fallback FAILED]
[...]
[msg] All nameservers have failed
[warn] Error from accept() call: Invalid argument
1/287 TESTS FAILED. (33 skipped)
FAILED
make[3]: *** [test_runner_select] Error 1
make[2]: *** [check-TESTS] Error 2
make[1]: *** [check-am] Error 2
make: *** [check] Error 2
It is the monotonic_prc_fallback from test/regress_util.c that is failing.
Yep, but it does not do any dns requests, I will take a look on weekends