openstack
/
puppet-openstack-integration
Code Issues Proposed changes

Skip searching selinux alerts 

In integration job, there is a task to search selinux alerts but any
alerts don't effectively make the whole job fail, and we often observe
timeout caused slow sealert command with many denial records.

This change disables the task until we fix all of the current denials.
We capture audit.log in CI so we can review that raw file to find out
any denials.

Change-Id: Ic850acfb2b576224c194ceb361e690c5fbf89838
This commit is contained in:
Takashi Kajinami 2022-02-22 15:56:21 +09:00
parent d65f7b5f45
commit 682e8a0dda
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
run_tests.sh
View File

@ -229,8 +229,8 @@ set -e
if [ $RESULT -ne 0 ] && [ $RESULT -ne 2 ]; then
print_header 'First Puppet run contains errors in catalog.'
catch_puppet_failures
print_header 'SELinux Alerts (1st time)'
catch_selinux_alerts
#print_header 'SELinux Alerts (1st time)'
#catch_selinux_alerts
exit 1
fi
timestamp_puppet_log
@ -244,8 +244,8 @@ set -e
if [ $RESULT -ne 0 ]; then
print_header 'Second Puppet run is not idempotent.'
catch_puppet_failures
print_header 'SELinux Alerts (2nd time)'
catch_selinux_alerts
#print_header 'SELinux Alerts (2nd time)'
#catch_selinux_alerts
exit 1
fi
timestamp_puppet_log
@ -402,7 +402,7 @@ elif [ -d .stestr ]; then
fi
subunit2html /tmp/openstack/tempest/testrepository.subunit /tmp/openstack/tempest/testr_results.html
print_header 'SELinux Alerts (Tempest)'
catch_selinux_alerts
#catch_selinux_alerts
print_header 'Done (run_tests.sh)'
exit $RESULT