RFR(XS): JDK-8021774: [TESTBUG] Tests failed because of a warning in its output
christian.tornqvist at oracle.com
Mon Nov 25 10:43:20 PST 2013
This is a problem with our infrastructure, we shouldn't have stale pid files around and they should be cleaned up. However I don't see that this will happen anytime soon. This test will continue to generate noise and people will start to mentally ignore any failures seen, making the test useless.
While I agree that it's not an optimal solution, I think that modifying this test is an acceptable solution to this issue.
From: hotspot-runtime-dev-bounces at openjdk.java.net [mailto:hotspot-runtime-dev-bounces at openjdk.java.net] On Behalf Of David Holmes
Sent: Sunday, November 24, 2013 7:38 PM
To: Mikhailo Seledtsov
Cc: hotspot-runtime-dev; GEORGE_TRIANTAFILLOU
Subject: Re: RFR(XS): JDK-8021774: [TESTBUG] Tests failed because of a warning in its output
IIRC this test was checking to see that systems that did not support NMT did not generate a warning when NMT was explicitly disabled. So this fix now removes that check.
I'm not sure there is a bug here, the test execution gave out the warning because there is an infrastructure issue with running the test and removing the attach file. With this change such problems will be masked instead of being investigated. I'd rather see such "failures"
reported and then marked as invalid when investigated. We don't expect there to be warnings in the normal case so any warning indicates a problem that needs to be resolved.
On 23/11/2013 3:10 AM, Mikhailo Seledtsov wrote:
> Please review this small change to the JVM RT test.
> I am posting this web rev on behalf of our new JVM Runtime SQE team
> member, George Triantafillou.
> Please note, this change is part of the test stabilization effort,
> therefore it could still be checked-in to the JDK-8.
> JBS: https://bugs.openjdk.java.net/browse/JDK-8021774
More information about the hotspot-runtime-dev