You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A friend who has attempted to build ntf-core by following the provided instructions passed this along. While he managed to get past the error, this might be an opportunity to fix the issue at source. Also, he says a couple of tests failed. His platform/toolchain details: gcc 9.4 on Ubuntu 20.04.
Here is a private email he sent me:
===========================================================================
I was unable to build ntf-core with bde installed in a non-standard location. I installed bde into /third-party/bde, which was fine. But then trying to configure ntf-core, it kept bombing out looking for bde in /opt/bb, despite having set —prefix. Eventually I gave up and installed bde into /opt/bb. That got me a bit further, but then the build failed with an error about _FORTIFY_SOURCE being redefined. I ended up having to delete the definition of _FORTIFY_SOURCE in repository.cmake. That allowed my to build.
Running the tests, I get two failures:
The following tests FAILED:
46 - ntsu_socketutil.t (Subprocess aborted)
141 - ntcf_system.t (Subprocess aborted)
I ran the tests manually, and got this:
./ntsu_socketutil.t
…
Running test case 13
Running test case 14
Running test case 15
Running test case 16
Running test case 17
[ F ][ 24MAR2023_03:37:58.106 ][ main ][ ntsu_socketutil.t.cpp:5613 ]: Assertion true: context.softwareTimestamp().isNull()
Aborted (core dumped)
./ntcf_system.t
...
[ W ][ 24MAR2023_03:39:04.415 ][ main ][ ntcf_system.t.cpp:7160 ]: ECONNREFUSED/ETIMEDOUT (x5)
Running test case 15
[ W ][ 24MAR2023_03:39:05.117 ][ main ][ ntcf_system.t.cpp:5519 ]: Testing driver EPOLL (static)
[ W ][ 24MAR2023_03:39:05.117 ][ main ][ ntcf_system.t.cpp:7280 ]: ETIMEDOUT (x5)
[ F ][ 24MAR2023_03:39:05.134 ][ main ][ ntcf_system.t.cpp:7333 ]: Assertion failed: connectResult.event().context().error() == ntsa::Error(ntsa::Error::e_CONNECTION_TIMEOUT)
Found: Connection refused (ECONNREFUSED) (connectResult.event().context().error())
Expected: Connection timed out (ETIMEDOUT) (ntsa::Error(ntsa::Error::e_CONNECTION_TIMEOUT))
Aborted (core dumped)
The text was updated successfully, but these errors were encountered:
For the issue installing BDE into /thirdparty/bde, but the build system thinking that BDE is actually installed into /opt/bb, see bloomberg/bde#290. This appears to be a problem with the build system used by https://github.com/bloomberg/bde: it installed pkgconfig meta-data stating that the "prefix=/opt/bb" regardless of the prefix specified when configuring the build of BDE. I doubled-checked this by building and installing BDE into /thirdparty/BDE, manually fixing-up the pkgconfig meta-data in /thirdparty/bde, then configuring ntf-core with --prefix /thirdparty/bde and I was able to build ntf-core successfully. We'll have to wait for the maintainers of BDE to fix bloomberg/bde#290.
For the remaining issues, I've created three new issues to address them individually:
A friend who has attempted to build
ntf-core
by following the provided instructions passed this along. While he managed to get past the error, this might be an opportunity to fix the issue at source. Also, he says a couple of tests failed. His platform/toolchain details:gcc 9.4 on Ubuntu 20.04.
Here is a private email he sent me:
===========================================================================
I was unable to build ntf-core with bde installed in a non-standard location. I installed bde into /third-party/bde, which was fine. But then trying to configure ntf-core, it kept bombing out looking for bde in /opt/bb, despite having set —prefix. Eventually I gave up and installed bde into /opt/bb. That got me a bit further, but then the build failed with an error about _FORTIFY_SOURCE being redefined. I ended up having to delete the definition of _FORTIFY_SOURCE in repository.cmake. That allowed my to build.
Running the tests, I get two failures:
The following tests FAILED:
46 - ntsu_socketutil.t (Subprocess aborted)
141 - ntcf_system.t (Subprocess aborted)
I ran the tests manually, and got this:
./ntsu_socketutil.t
…
Running test case 13
Running test case 14
Running test case 15
Running test case 16
Running test case 17
[ F ][ 24MAR2023_03:37:58.106 ][ main ][ ntsu_socketutil.t.cpp:5613 ]: Assertion true: context.softwareTimestamp().isNull()
Aborted (core dumped)
./ntcf_system.t
...
[ W ][ 24MAR2023_03:39:04.415 ][ main ][ ntcf_system.t.cpp:7160 ]: ECONNREFUSED/ETIMEDOUT (x5)
Running test case 15
[ W ][ 24MAR2023_03:39:05.117 ][ main ][ ntcf_system.t.cpp:5519 ]: Testing driver EPOLL (static)
[ W ][ 24MAR2023_03:39:05.117 ][ main ][ ntcf_system.t.cpp:7280 ]: ETIMEDOUT (x5)
[ F ][ 24MAR2023_03:39:05.134 ][ main ][ ntcf_system.t.cpp:7333 ]: Assertion failed: connectResult.event().context().error() == ntsa::Error(ntsa::Error::e_CONNECTION_TIMEOUT)
Found: Connection refused (ECONNREFUSED) (connectResult.event().context().error())
Expected: Connection timed out (ETIMEDOUT) (ntsa::Error(ntsa::Error::e_CONNECTION_TIMEOUT))
Aborted (core dumped)
The text was updated successfully, but these errors were encountered: