[KULRICE-14239] Performance improvements of AFT tests Created: 17/Apr/15  Updated: 30/Mar/16

Status: Open
Project: Kuali Rice Development
Component/s: Development
Affects Version/s: None
Fix Version/s: 2.6
Security Level: Public (Public: Anyone can view)

Type: Task Priority: Major
Reporter: Claus Niesen Assignee: Unassigned
Resolution: Unresolved Votes: 0
Labels: LongTerm
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Sprint: Rice Sprint 2015-04-15
KAI Review Status: Not Required
KTI Review Status: Not Required
Code Review Status: Not Required
Include in Release Notes?:
Story Points: 3


Currently our rice-sampleapp and krad-sampleapp AFTs each take about 12 hours to complete. Can we shorten that time?

Comment by Claus Niesen [ 17/Apr/15 ]

WebDriverUtils.waitFor method still uses the implicitWait in addition to the Thread.sleep causing a full 2 seconds delay between polling. Two seconds seems to be an excessive delay between polling.

FluentWait might provide a nicer solution to checking the existence of web elements.

Comment by Claus Niesen [ 17/Apr/15 ]

Side note: When refactoring the code it wouldn't hurt to remove the "wait" and "jiraAware" from the general AFT methods. When a test needs to click on an item, the test doesn't care about the implementation specifics (i.e. if we wait when the element isn't there right away or that our tests are jira aware). Naming should be plain so implementation specific things can be changed as desired.

Comment by Claus Niesen [ 12/May/15 ]

The following change caused the run time to decrease by about 2h but caused 12 failures.
Investigation on whether the failure could be fixed with case specific waits or other AFT fixes.

Generated at Tue Oct 27 06:57:46 CDT 2020 using JIRA 7.0.11#70121-sha1:19d24976997c1d95f06f3e327e087be0b71f28d4.