I'm doing some refactoring on the testsuite to remove the need for thirdparty,
so some of the recent failures are my fault. I'll try to fix those asap.
It would be helpful if Hudson was a little more reliable. It seems like about
half the Hudson jobs I kick off for the AS testsuite build end up failing
randomly (failed svn checkout, out out memory, etc). I don't know if there is a
way to improve this, but it would make testing a lot easier.
Andrew Lee Rubinger wrote:
The AS TestSuite in trunk has again gotten out of line. I've got
a
4100+ line patch coming in, and need to be able to spot potential
regressions reliably.
We need to stress the importance of checking up on your commits to
ensure they don't bring the house down. Would like to see:
1) More self-imposed responsibility, looking after the builds in Hudson
if you've made commits to trunk recently
2) More rollbacks / notices in case of regression. "Hey Jim you broke
it fix it. Love, Me".
3) Maybe an appointed "watchdog" each week to act as TestSuite
gatekeeper in case 1 and 2 /fail.
S,
ALR