Argh. Another JBlanket file I/O failure.
Cedric: I have a new idea. I am assuming that you checked and that the
file that Ant claims it can't find does indeed exist on disk. Correct?
If so, here's a possible workaround: don't invoke JBlanketReport during
this run of Ant. Instead, fire up a separate invocation of Ant after this
one completes whose sole purpose is to run JBlanketReport. If it's actually
the case that there's some file closure related problem, then separating
JBlanketReport into its own subsequent execution stream should solve the
problem (if, indeed that is the problem).
Other ideas:
* The new build system is going to be easier to fiddle around with, and I
am going to implement JBlanket-based execution a bit differently. Perhaps
that will help things.
* I wonder why we are having these problems consistently now. We ran
JBlanket for literally years without this issue coming up, and now it comes
up regularly.
* Another possibility is to switch to Emma or Clover if we can't get this
resolved, although I hate to give up on JBlanket.
Cheers,
Philip
--On Saturday, October 29, 2005 3:25 AM -1000 Qin Zhang <[EMAIL PROTECTED]>
wrote:
Hackystat build (configuration Hackystat-ALL) failed.
Build report is available at
http://xenia.ics.hawaii.edu/hackyDevSite/configurationBuildReport.do?year
=2005&month=10&day=29&configuration=Hackystat-ALL Build Time Stamp: Sat
Oct 29 03:25:33 HST 2005
