cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Schuller <>
Subject Re: Cassandra node's replication factor two with random partition non Bootstrap node problem
Date Tue, 12 Apr 2011 20:43:03 GMT
> I have two cassandra node's,If Boot strapped  nodes goes down my service
> remains alive,But if my non Bootstrap (master) node goes down my live site
> goes down as well,I am using cassandra 0.6.3 can any elaborate on this
> problem.

Assuming your RF is 2 (not 1), and that you are reading at consistency
level ONE (not QUORUM, which would be 2 in the case of RF=2),
single-node failures should be tolerated.

In order for people to help you'd have to specify some more
informaton. For example, your "site goes down" - but what is the
actual error condition w.r.t. Cassandra? What is the error reported by
the Cassandra client (and which client is it)?

I'm not sure what you mean w.r.t. boostrap/master etc. All nodes
should be entirely equal, with the exception of nodes that are marked
as seed nodes. But seed nodes going down should not cause reads and
writes to fail.

/ Peter Schuller

View raw message