RFR: 8283222: improve diagnosability of runtime/8176717/TestInheritFD.java timeouts

Daniel D.Daugherty dcubed at openjdk.java.net
Tue Mar 15 21:32:57 UTC 2022


On Tue, 15 Mar 2022 21:13:32 GMT, Daniel D. Daugherty <dcubed at openjdk.org> wrote:

> While working on the following bug:
> 
>     [JDK-8282321](https://bugs.openjdk.java.net/browse/JDK-8282321) runtime/8176717/TestInheritFD.java times out during stress testing
> 
> I've made changes to the test to improve the diagnostics when test failures occur.

The improvements to the test:
- the three VMs now identify their names and their PIDs
- the third VM now waits for the second VM on all platforms and not just windows
- encapsulated all of the first VM's waiting logic into the waitForSubPids() method and the first VM now waits of all of the subpids and not just for the last output line
- added timeout handling for the subpid waits, the wait time is 12.5% of default timeout (15 seconds) * the timeoutFactor specified by the execution environment
- added a test debug/simulation point for the second VM timing out
- added a test debug/simulation point for the third VM timing out
- corrected the test's header comments

The test has been exercised in Mach5 Tier[1345678] runs and
in stress kit runs on my Ubuntu 20.04 machine.

-------------

PR: https://git.openjdk.java.net/jdk/pull/7828


More information about the hotspot-runtime-dev mailing list