HI Mark-
We have a boatload of tests that were originally contributed
by Sun that are now part of this project. I know Brian Murphy
did some work this year to get them buildable and runnable
(I'll dig up that commit and forward) -- so they should be there,
runnable, and help to validate the release.
Don't you agree?
-Jim
On Sep 11, 2008, at 4:03 AM, Mark Brouwer wrote:
Hi Dan,
Is there someplace I could read up on what we would need for a proper
test env? Is it along the lines of just running all the existing
tests
under various OS's? (more than what Hudson is doing now?)
I'm currently running automated builds for some other OSS projects on
Win2k, WinXP, Ubuntu 8.04 [all i386 for now], though might be able to
add amd64 if needed.
(Unfortunately, my OpenSolaris i386 toasted itself a few too many
times
when it locked and zfs went bye, bye after reboot - [note: this is
not
a
slam on zfs, the hardware involved was prehistoric] - I will try it
again after the next release. Of course if some sparc hardware fell
out
the sky, I'd certainly add it to my basement build farm).
Maybe I've found something I could finally help with? ;)
The problem is I'm not sure what a proper test environment really
would mean or looks like. All I know is this article by Nigel Daley
who also setup the current tests with Hudson.
http://weblogs.java.net/blog/nidaley/archive/2006/04/jini_300000_tes.html
Maybe we don't have to go that far, but as I understood the current
tests are only a subset to keep the test time to a reasonable minimum
and that for the AR1 release Sun did the additional testing. I think
this is really somebody more qualified than I am should answer.
Nevertheless it would be great if you could help a hand, the same
applies BTW to Dennis Reedy. Still have to answer his questions ...
Regards,
--
Mark Brouwer
