Fix defer-cleanup flag functionality for system testrunner

[renekalff]

Currently the system testrunner is calling TearDown itself bypassing the defer-cleanup functionality. This MR resolves this by leaving calling TearDown to the testrunner main function.

[elasticmachine]

💚 Build Succeeded

the below badges are clickable and redirect to their specific view in the CI or DOCS

Expand to view the summary

Build stats

• Build Cause: mtojek commented: /test

• Start Time: 2021-06-30T12:53:14.476+0000

• Duration: 23 min 5 sec

• Commit: 68e3390

Test stats 🧪

Test	Results
Failed	0
Passed	315
Skipped	2
Total	317

Trends 🧪

[mtojek]

Hi @renekalff , thank you for the PR. Could you please describe manual testing steps to verify the fix?

[renekalff]

Hi @renekalff , thank you for the PR. Could you please describe manual testing steps to verify the fix?

After applying this fix run elastic-package test system —defer-cleanup 1m -v in a package with system testing defined like the nginx package.

After loading the resources a debug message will be displayed that cleanup is deferred for one minute. Now manual inspection is possible, for example using kibana.

After one minute cleanup is performed and the documents are removed from the datastream indexes.

Setting a longer delay will allow you to develop the kibana dashboards based on the test events. Changes made in the dashboards or visualizations won’t be lost in the cleanup process nor will they be overwritten on subsequent system test executions.

[mtojek]

/test

[mtojek]

/test

[mtojek]

LGTM. Thanks for the fix!