I am running into exact same issue where >= queries on secondary indexes don't work reliably, even in single node environment (using 2.1.0-rc5) and quering from same machine where cassandra server is running.
If secondary indices can't give results, when using cassandra is one supposed to create schema like this that uses int_val as primary index? CREATE TABLE foo_int_table ( foo_name ascii, foo_shard bigint, int_val bigint, PRIMARY KEY ((int_val)) ) WITH read_repair_chance=0.1; -Subodh On Wed, Aug 13, 2014 at 11:17 AM, Robert Coli <rc...@eventbrite.com> wrote: > On Wed, Aug 13, 2014 at 7:33 AM, Ian Rose <ianr...@fullstory.com> wrote: >> >> I'm starting to wonder if range queries on secondary indexes aren't >> supported at all (although if that is the case, I would certainly prefer an >> error rather than a timeout!). I've been scouring the web trying to find a >> definitive answer on this but all I have come up with is this (old, >> non-authoritative) blog post which states "Cassandra’s native index is like >> a hashed index, which means you can only do equality query and not range >> query." > > > Somewhere in google I'm pretty sure you can find me on this list explaining > the basic case for why I do not recommend using Secondary Indexes. > > An endless trickle of edge cases such as this one where trivial operations > do not work is one of the reasons given. The "convenience" they provide can > really only be convenience if they actually, you know, work. > > =Rob
