Hi Bhuvan, You’re correct that large unlogged batch on one partition isn’t an issue. The logging behaviour has been/is being changed - see this jira for a detailed discussion: https://issues.apache.org/jira/browse/CASSANDRA-10876
Cheers Ben On Mon, 9 May 2016 at 00:58 Bhuvan Rawal <bhu1ra...@gmail.com> wrote: > Hi, > > I was testing unlogged batches of size 15 and encountered bunch of these > messages which are filling up my disk: > 2016-05-08 20:22:58,338 [WARNING] cassandra.protocol: Server warning: > *Unlogged > batch covering 1 partition detected* against tables [xyz.abc, xyz.123]. > You should use a logged batch for atomicity, or asynchronous writes for > performance. > > If I have identified the partition and writing bunch of rows into that > using unlogged batch, isnt that an ideal case? > > Regards, > Bhuvan > -- ———————— Ben Slater Chief Product Officer, Instaclustr +61 437 929 798
