RFR of JDK-8170704: java/rmi/activation/Activatable/* tests fails intermittently with "output improperly annotated"

Roger Riggs Roger.Riggs at Oracle.com
Tue Dec 6 15:27:17 UTC 2016


I'm rather partial to fractional timeoutFactors, especially those < 1.0, 
useful for debugging and not having to wait.
Can the conversion to long be deferred until it is used?

(The other TestLibrary has a Util function to scale the timeout, but not 
the RMI one...  arg technical debt!)


On 12/6/2016 5:19 AM, Daniel Fuchs wrote:
> Looks reasonable to me Hamlin!
> best regards,
> -- daniel
> On 06/12/16 07:18, Hamlin Li wrote:
>> Would you please review the below patch?
>> bug: https://bugs.openjdk.java.net/browse/JDK-8170704
>> webrev: http://cr.openjdk.java.net/~mli/8170704/webrev.00/
>> * Root Cause:
>> The issue can be reproduced by adjust sleeping time to smaller values,
>> so on busy machine it's possible that the sleeping time is not long
>> enough for rmi system to pass myRMI err/out to rmid err/out.
>> * Solution:
>> scale timeout by time factor.
>> Thank you
>> -Hamlin

More information about the core-libs-dev mailing list