= Overview = We've made some notable accomplishments since our last report, but it doesn't feel as much like progress. We had to do a fair amount of work just to keep the testing framework working, thanks to challenges with lxc, our zope.testing fork, and ec2 precise images, among others. We have fixed all of the test failures that we had identified by the time of the last meeting, but, thanks perhaps in part to the increased rigor of adding --shuffle to our tests to ferret out more isolation bugs, we have more to fix. We have made our buildbot process a bit better, by introducing better reporting, better testr support, and better handling of exceptional error cases. We are still waiting on IS for the testing machine, and will have much more to do with them before the project is done. We expect, therefore, to have time to work on additional, related projects, possibly including work such as lpsetup and a parallel version of "ec2 test". = Progress towards biweekly action items = * [yellow] Bug fixes for test failures discovered in parallel test runs. * We addressed all items identified in the last meeting and several more. We have new test failures to fix. * [yellow] [carry over from past four weeks] Deliver scaling assessment based on experimental results, using ec2 * Still blocked by IS * [yellow] Get data center box running tests, and have a single comparison run with ec2. * Still blocked by IS * /dev/random exhaustion solution approved and installed * IS requested that we first determine whether it is a problem on the data center machines, or if it is specific to the ec2 environment. If we determine that it is a problem, we will address then. = Other accomplishments = * [carry over from two weeks] blocked on Clint Byrum releasing python-shelltoolbox and Python charm helpers * some lpsetup progress = Progress on tracked items = * No progress on testr bugs * Bug:914166 * Bug:949950 * Bug:957145 * Bug:961103 * Bug:974622 [new] [high priority for us] * LXC * Bug:968371 [new] [resolved] * Bug:959352 [kernel] [no progress] = Goals for next meeting = * Bug fixes for test failures discovered in parallel test runs. * Bug:974584 (celery tests; reliable, at least on 3 runs on 8-core ec2) * Bug:974585 (another UncleanReactorError failure) * Bug:974586 (bug import tests; intermittent) * Bug:974608 (bug test; intermittent) * Bug:974617 (test_operational_view_integration; intermittent) * Bug:974632 (memcache disconnection; also lpbuildbot; intermittent) * Provide fixes for some of the testr bugs (esp. Bug:974622) * Deliver scaling assessment based on experimental results, using ec2 (carry over from previous four weeks) * (carry over from previous two weeks) Get data center box running tests, and have a single comparison run with ec2.