I suspect there’s a bad leveldb corruption bug that’s been present for a while so I’m going to write a test to verify that it’s present.
… so I’m trying to write a test. My idea is to write 1000 messages or so and keep consuming/producing more and giving them unique counters. If we lose some of the messages between restarts I know that there’s a data loss bug. my plan is to use the failover protocol and then tune leveldb for a small log size. But my question is about buffers.. if a message goes into a log buffer, but it’s not written to a disk, and it’s restarted, I assume it vanishes? the variables are here: http://activemq.apache.org/leveldb-store.html … so I’m trying to figure out exactly how to configure leveldb to handle full persistence between restarts. -- Founder/CEO Spinn3r.com Location: *San Francisco, CA* blog: http://burtonator.wordpress.com … or check out my Google+ profile <https://plus.google.com/102718274791889610666/posts> <http://spinn3r.com>
