zuul/zuul-jobs
Code Issues Proposed changes

Remove OOM check from tox role

Browse Source 
This is something we did in openstack-infra but not sure it make sense
for default tox jobs.  It is possible we want to make this an ansible
role for playbooks to opt into post-run. But with properly logging on
a job, it should be straightforward to determine why a tox test
failed.

Change-Id: I746fae586a3df40a07ceca9c0bebe579f9f74f34
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
This commit is contained in:
Paul Belanger 2017-07-19 18:36:30 -04:00
parent c7037bb2b4
commit cd6902638f
No known key found for this signature in database
GPG Key ID: 611A80832067AF38
1 changed files with 0 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
roles/tox/tasks/main.yaml
View File

@ -38,20 +38,6 @@
fi
}
function check_oom {
$script_path/jenkins-oom-grep.sh post
oomresult=$?
if [ $oomresult -ne "0" ]; then
echo
echo "This test has failed because it attempted to exceed configured"
echo "memory limits and was killed prior to completion. See above"
echo "for related kernel messages."
echo
exit 1
fi
}
function check_nose_html {
htmlreport=$(find . -name $NOSE_HTML_OUT_FILE)
if [ -f "$htmlreport" ]; then
@ -81,7 +67,6 @@
result=$?
process_testr_artifacts
check_oom
check_nose_html
# Rename tox .log files to .log.txt so that Apache serves them as