El 09/06/2010, a las 04:31, Peter Schuller escribió: >> I've had a server crash > > As jbellis points out there may be hardware issues, but if, in > particular, the crash in question was a power outage a very common > problem is running on a system which does not honor write barriers. > Was it a power outage?
It was a server crash, this Cassandra instance is running (for internal testing purposes) on a VPS, and its host crashed completely as my service provider told me. I suppose that if I was running a 2-node cluster with a RF=2 this wouldn't be an issue of course, as soon as the 2 nodes were on different hosts :-) As an exercise I'll try Jonathan's advice. Thanks guys! -- Lucas Di Pentima - Santa Fe, Argentina Jabber: lu...@di-pentima.com.ar MSN: ldipent...@hotmail.com