-
Notifications
You must be signed in to change notification settings - Fork 16.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Numerous tests failing autotest for many many builds #10360
Comments
there is something wrong on autotest server as travis is able to pass the test ... |
travis and autotest are not running the same tests. |
yes they are ... we only have on set of autotest |
@khancyr FYI if I'm not mistaken: autotest runs build_autotest.sh which takes several hours to complete. Ci runs build_ci.sh. There is a small overlap of tests with autotest, but it is different. The problem is not with the autotest server. The problem is our program fails the tests we wrote. |
|
Autotest.sh is calling autotest.py like build_ci.sh the test are the same. But I am pretty sure that the problem come from our build server... Something with cache management, like usual... |
I've had a look at the full log from the most recent autotest. At the very end of the log we can see dive.ArduSub fails from some missing file Some other problems with missing dependencies are visible at the very end of the file also. Is there any information available about how autotest is set up and deployed? $ curl http://autotest.ardupilot.org/autotest-output.txt | tail -n50
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 16.8M 100 16.8M 0 0 12.2M 0 0:00:01 0:00:01 --:--:-- 12.2M
MANUAL> log download latest /storage/autotest/APM/buildlogs/APMrover2-log.bin
MANUAL> Downloading log 2 as /storage/autotest/APM/buildlogs/APMrover2-log.bin
Flight battery warning
Finished downloading /storage/autotest/APM/buildlogs/APMrover2-log.bin (9207808 bytes 35 seconds, 260.1 kbyte/sec 0 retries)
AUTOTEST: Drained 2051 messages from mav (110.192381/s)
AUTOTEST: Drained 0 messages from mav (0.000000/s)
Uploading 115 logs to http://firmware.ardupilot.org/CI-Logs/2019-02-04-12-42
@ERROR: Unknown module 'CI-Logs'
rsync error: error starting client-server protocol (code 5) at main.c(1653) [sender=3.1.1]
AUTOTEST: PASSED: "DownLoadLogs (Download logs)"
AUTOTEST: Failing tests:
DiveMission (Dive mission) ([Errno 2] No such file or directory: '/storage/autotest/APM/APM/test.ArduSub/logs/2019-02-04/flight15/way.txt') (see /storage/autotest/APM/buildlogs/ArduSub-DiveMission.txt)
>>>> FAILED STEP: dive.ArduSub at Mon Feb 4 12:42:43 2019
>>>> RUNNING STEP: convertgpx at Mon Feb 4 12:42:43 2019
Running: ("/bin/rm -f logs/*.BIN logs/LASTLOG.TXT") in (.)
step=convertgpx
Running: ("/storage/autotest/APM/APM/modules/mavlink/pymavlink/tools/mavtogpx.py --nofixcheck /storage/autotest/APM/buildlogs/ArduSub-test.tlog") in (.)
Created /storage/autotest/APM/buildlogs/ArduSub-test.tlog.gpx with 2277 points
Running: ("gpsbabel -i gpx -f /storage/autotest/APM/buildlogs/ArduSub-test.tlog.gpx -o kml,units=m,floating=1,extrude=1 -F /storage/autotest/APM/buildlogs/ArduSub-test.tlog.kml") in (.)
Running: ("zip /storage/autotest/APM/buildlogs/ArduSub-test.tlog.kmz /storage/autotest/APM/buildlogs/ArduSub-test.tlog.kml") in (.)
updating: storage/autotest/APM/buildlogs/ArduSub-test.tlog.kml (deflated 98%)
Running: ("mavflightview.py --imagefile=/storage/autotest/APM/buildlogs/ArduSub-test.tlog.png /storage/autotest/APM/buildlogs/ArduSub-test.tlog") in (.)
Traceback (most recent call last):
File "/home/autotest/.local/bin/mavflightview.py", line 4, in <module>
__import__('pkg_resources').run_script('MAVProxy==1.8.5', 'mavflightview.py')
File "/usr/local/lib/python2.7/dist-packages/pkg_resources/__init__.py", line 742, in run_script
self.require(requires)[0].run_script(script_name, ns)
File "/usr/local/lib/python2.7/dist-packages/pkg_resources/__init__.py", line 1510, in run_script
exec(script_code, namespace, namespace)
File "/home/autotest/.local/lib/python2.7/site-packages/MAVProxy-1.8.5-py2.7.egg/EGG-INFO/scripts/mavflightview.py", line 11, in <module>
File "build/bdist.linux-x86_64/egg/MAVProxy/modules/mavproxy_map/mp_slipmap.py", line 19, in <module>
File "build/bdist.linux-x86_64/egg/MAVProxy/modules/lib/win_layout.py", line 4, in <module>
ImportError: No module named wx
>>>> FAILED STEP: convertgpx at Mon Feb 4 12:42:45 2019 (Command '['mavflightview.py --imagefile=/storage/autotest/APM/buildlogs/ArduSub-test.tlog.png /storage/autotest/APM/buildlogs/ArduSub-test.tlog']' returned non-zero exit status 1)
Traceback (most recent call last):
File "APM/Tools/autotest/autotest.py", line 589, in run_tests
if run_step(step):
File "APM/Tools/autotest/autotest.py", line 434, in run_step
return convert_gpx()
File "APM/Tools/autotest/autotest.py", line 210, in convert_gpx
util.run_cmd("mavflightview.py --imagefile=%s.png %s" % (m, m))
File "/storage/autotest/APM/APM/Tools/autotest/pysim/util.py", line 73, in run_cmd
return check_call(cmd, shell=shell, cwd=directory)
File "/usr/lib/python2.7/subprocess.py", line 541, in check_call
raise CalledProcessError(retcode, cmd)
CalledProcessError: Command '['mavflightview.py --imagefile=/storage/autotest/APM/buildlogs/ArduSub-test.tlog.png /storage/autotest/APM/buildlogs/ArduSub-test.tlog']' returned non-zero exit status 1
check step: convertgpx
FAILED 9 tests: ['run.examples', 'fly.ArduPlane', 'fly.QuadPlane', 'drive.APMrover2', 'drive.balancebot', 'fly.ArduCopter', 'fly.CopterAVC', 'dive.ArduSub', 'convertgpx'] |
I've installed |
I've merged a fix for not understanding where MAVProxy's files are being put into master. |
eyyy! @peterbarker it worked! http://autotest.ardupilot.org/ There remain only two failing tests now. |
On Tue, 5 Feb 2019, Jacob Walser wrote:
eyyy! @peterbarker it worked! http://autotest.ardupilot.org/
🎉
There remain only two failing tests now.
I've created another PR to fix some remaining issues:
https://github.com/ArduPilot/ardupilot/pull/10422/commits
|
We have currently 10 failing autotests, and tests have been failing on every build for several months.
What is the point of doing these tests if we do not make use of the results? There are obviously problems here that are not being resolved.
I think we need to strive to police this better, increase the visibility of these results, and make some sort of rule about fixing failing tests within some amount of time. Preventing merge of code that fails tests (by adding autotest to ci) would be even better.
Related #9589 and #9592
The text was updated successfully, but these errors were encountered: