I just ran encountered a corrupted index problem after a deploy. I
blew away the index, restarted the drb server, and rebuilt the entire
index. Exceptions were still trickling up to my mongrels.
The problem was solved by restarting the mongrels. Does this make any
sense? I thought the mongrels/AAF just access the drb server as a
service, and were unaware of the state of the service behind the port?
Thanks for any insight,
John