Would it be reasonable to have (possibly configurable) caps on the maximum size of any internal Cassandra queues that are directly populated by client requests? I understand this might mean sometimes breaking the API contract for writers using CL.ZERO by blocking on those calls, but on the other hand the server crashing isn't part of the API contract either :)
On Jun 11, 2010, at 9:09 AM, Gary Dusbabek wrote: >> >> Does this seem like a Cassandra bug or is it well known that Cassandra always >> needs more than 1GB of heap space? > > Not a bug at all. You were putting a lot of memory pressure on > Cassandra by writing at CL.ZERO. > > Gary.
