On 8/12/2010 12:59 PM, Jim Newsham wrote:
Would you be willing to run the same test on the jdbm2 codebase?  It would interesting to know if the problem exists in both places (if it does, it would give at least a hint of where the problem is coming from).
  Sure, I'll give that a try.

Ok I'm running the same test for jdbm2 right now.  There were some very minor api changes, but basically the same code.  I'll report back on the results when I have them.  I immediately noticed that the check() method runs in faster -- in 10ms as opposed to 25ms for jdbm1.

By the way, when I google for ["page header magic for block" "not ok"], I get a small handful of people reporting problems for other projects which apparently use jdbm internally (livesite, novell identity manager, gemfire, opendedup...).  I just skimmed, but looks like they generally recommend to delete .db and .log files and restart.  So the problem is known to be happening in the wild.