Break up os::cleanup::all #14982
Break up os::cleanup::all #14982
Conversation
The previous attempt to DRY out cleanup code went too far. Generating jUnit, cleaning up after a cluster and describing the outcome of a script are separate steps and it makes sense to choose to run them independently. This allows scripts like the unit tests to not spend time tearing down clusters, etc. Signed-off-by: Steve Kuznetsov <skuznets@redhat.com>
|
@stevekuznetsov were you going to add support for a SKIP_CLEANUP var? |
|
Sure, is easy |
Signed-off-by: Steve Kuznetsov <skuznets@redhat.com>
|
@csrwng added |
|
LGTM |
|
Evaluated for origin test up to e416e24 |
|
@dobbymoodge why did it take 2h for rosie to pick this up? |
|
thanks! |
|
@stevekuznetsov It didn't? Trigger comment was 11:55 AM, p_p_r_o picked it up at 12:07 PM (https://ci.openshift.redhat.com/jenkins/job/process_pull_requests_origin/64723/), and it kicked off the test job at 12:08 PM: https://ci.openshift.redhat.com/jenkins/job/test_pull_request_origin/2881/ Now that's not the job it's running now. I'll figure out why it retriggered the test. |
|
@stevekuznetsov Looks like it retriggered at 1:55 PM because the base commit ID changed: https://ci.openshift.redhat.com/jenkins/job/process_pull_requests_origin/64736/console |
|
continuous-integration/openshift-jenkins/test SUCCESS (https://ci.openshift.redhat.com/jenkins/job/test_pull_request_origin/2887/) (Base Commit: b75e645) (PR Branch Commit: e416e24) |
|
Ah I see. [merge] |
|
@openshift-bot, the last build failed from the following flakes:
re[merge] |
|
@openshift-bot, the last build failed from the following flakes:
re[merge] |
|
#14575 [merge] |
|
Evaluated for origin merge up to e416e24 |
|
continuous-integration/openshift-jenkins/merge SUCCESS (https://ci.openshift.redhat.com/jenkins/job/merge_pull_request_origin/1232/) (Base Commit: c9f829c) (PR Branch Commit: e416e24) |
The previous attempt to DRY out cleanup code went too far. Generating
jUnit, cleaning up after a cluster and describing the outcome of a
script are separate steps and it makes sense to choose to run them
independently. This allows scripts like the unit tests to not spend time
tearing down clusters, etc.
Signed-off-by: Steve Kuznetsov skuznets@redhat.com
[test]
/cc @csrwng @DirectXMan12