Stress Test - cleanUp() removing all test resources as an admin
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
tempest |
Fix Released
|
Undecided
|
Julien Leloup |
Bug Description
After a successfull run of a stress test using "run_stress.py", a cleanUp() method is called, even after the call of tearDown() implemented in the stress test.
This cleanUp() method is using an admin manager in order to delete all resources available on the OpenStack it is running : all servers (using "all_tenants" paremeter), snapshots, volumes, key pairs, ...
Even it this behaviour make sense on a private OpenStack installation dedicated to testing (like a local devstack), this behaviour is potentially dangerous when running a stress test on a shared OpenStack used by multiple teams, even isolated on different tenants.
I propose to have a new parameter in tempest.conf called "full_clean_stack" in the [stress test] section, with a default value set to False :
- True : cleanUp() is called at the end of the stress test run in order to have the actual behaviour (get a pristine environment)
- False : cleanUp() is not called and only the tearDown() implemented in the stress test will be used to remove onlt test resources, leaving other resources untouched.
Note : this defect is slightly linked to this one "https:/
Changed in tempest: | |
status: | New → Confirmed |
Changed in tempest: | |
assignee: | nobody → Julien Leloup (julien-llp) |
Fix proposed to branch: master /review. openstack. org/73274
Review: https:/