Please review this addition of diagnostic output to an intermittently
failing Runtime.exec test.
The Solaris implementation of ProcessBuilder.start uses posix_spawn and
a trampoline
technique that invokes a small spawnhelper executable which in turn
forks the new executable.
The test has been intermittently failing on Solaris and is not reproducible.
The diagnostic output is only triggered if the test is failing and is
conditional
since the programs are only available on Solaris.
Other suggestions for diagnostics or the solution are welcome.
[The test fix will be pushed using a different bug number since it does
not attempt to fix the issue.]
Webrev:
http://cr.openjdk.java.net/~rriggs/webrev-sleepycat-8044841/
Thanks, Roger
