• Still stuck in tinderbox hell, tracking down and fixing problems on the tinderboxes that are unrelated to recorded scripts. I've verified that the trunk fails to run on the recorded_scripts tinderboxes. I get failures running my recorded_scripts tinderboxes that don't happen on my machine and I've occationally seen failures on my machine that I don't see on the tinderboxes. I suspect these probably are related to speed of the machines running the tinderboxes, configuration differences between machines, or some other difference unrelated to the Chandler code.
• Narrowed down bug #11202: Debug build crashes on first attempt to open repository created by end-user build for same version to a repository call that, incorrectly causes events to be processed during a save.
• Fixed a bug in the recorded scripts branch where TestI18N^?.py merges from the trunk were missing. Interestingly, Heikki saw this same error independently on the Trunk, so I suspect some SVN wierdness might be involved.
• Found a problem with a recorded script that fails on Sunday. Tracked down the problem to a time dependency of the script. Rerecorded the script to avoid the time depencency. The new script runs locally, but fails on the tinderboxes. I suspect that this might be a real problem related to the speed of the machines running the tinderboxes.
• Worked narrowing down a bug that always occurs on the branch tinderboxes, used to appear on the trunk tinderboxes, but currently doesn't appear on the trunk tinderboxes. This particular bug is probably a timing dependent bug in the old functional test framework. I'm pretty close to understanding the cause of this bug.
• Still waiting for Bear and Heikki to get the Linux tinderboxes running on the recorded scripts branch
• Installed Leopard. My disk was always getting full, so I installed a larger disk first. I now know why installing a disk in a MacBook^? Pro voids the warranty. I ran into a few snags, but it's looking like my world is almost normal.

-- JohnAnderson - 06 Nov 2007