If a test returns 0 but writes an XML logfile that contains FAIL or a
corrupted XML file, then qt-testrunner considers it a CRASH and exits
with 3.
Previously any test execution returning 0 (success) was considered
a PASS. Changing this behavior with this patch finds a lot of test
crashes on Android (QTBUG-100470), because the tests are run indirectly
on the emulator and the test wrapper process could not detect the crash,
thus returning 0 to qt-testrunner. But the corrupt XML file is caught
now.
Likewise, if a test returns != 0 but the XML logfile contains only PASS,
qt-testrunner considers it a FAIL. This used to be the case but now
tests are added.
Finally changed logging for such cases from INFO to WARNING.
Task-number: QTBUG-100470
Change-Id: I404c9d2211c7de027bf776d1914519d37f513ca1
Reviewed-by: Assam Boudjelthia <assam.boudjelthia@qt.io>
QNX requires its own TESTRUNNER script in order to SSH to the QEMU guest
and run the tests. Simplify it a lot and fix the way it passes arguments
so that they retain spaces and quotation.
Also wrap it with the generic TESTRUNNER script for CI, which is
qt-testrunner.py, so that tests can re-run in case of flakiness.
Delete prefix.sh as the environment variables are now properly set in
the Coin platform config files.
Avoid executing extra chmod commands to make scripts executable,
but use the right Coin instruction for that. For reference, we
use 493 for file mode which equals to 755 in octal, as Coin expects
the file mode in decimal.
Change-Id: Ife4d1caef606f48b92ba1da1cfb14ec0dea11ef2
Reviewed-by: Pasi Petäjäjärvi <pasi.petajajarvi@qt.io>
Previously the qt-testrunner script was avoiding to write XML log files
for the flaky test re-runs, since parsing them always provides limited
information about what went wrong (for example unparseable files) and
might miss issues completely (for example a crashed test might not even
write a log file), or even mislead if multiple different data points
about a single testcase end up into the database..
But currently Coin is the system that uploads test information to the
testresults database and expects to find such log files in an attempt to
send as much information as possible regarding failures.
Re-enabling these log files will allow Coin to deduce as much
information as possible about the test runs.
This is a temporary step at least, until the test-uploading logic is
separated from Coin. Then it will be easier to find the full information
about what happened with flaky or crashed tests by special logs from
qt-testrunner.py, which is the test re-running orchestrator and
therefore has more knowledge about what went wrong.
Fixes: QTQAINFRA-4754
Change-Id: I69e4469cbe6c0a6ca5b2473eaf8d034632564342
Reviewed-by: Fabian Kosmale <fabian.kosmale@qt.io>
Reviewed-by: Toni Saario <toni.saario@qt.io>
Reviewed-by: Simo Fält <simo.falt@qt.io>
Some function names are special for qtestlib, in the sense that they can
not be specified as a command line argument to run individually.
In such cases qt-testrunner treats the failure specially and tries once
to re-run the full test executable.
Fixes: QTBUG-89011
Change-Id: I0cc25f91c57374e5ac65ade10e2e223fe969f211
Reviewed-by: Volker Hilsheimer <volker.hilsheimer@qt.io>
Reviewed-by: Daniel Smith <Daniel.Smith@qt.io>
It causes crashes because of interleaved execution of Quick tests.
Task-number: QTBUG-98350
Change-Id: I61e26f5da5617e06810db8eb7991f17eecdd3acb
Reviewed-by: Tor Arne Vestbø <tor.arne.vestbo@qt.io>
Script that wraps Qt test execution in order to iron-out flakiness.
Task-number: QTBUG-96353
Change-Id: Ie8ee780e3f4f8d086c080a7784f9f68fd1867be2
Reviewed-by: Daniel Smith <Daniel.Smith@qt.io>
