On Sun, Jan 3, 2016 at 5:54 PM, Shuo Chen wrote:
> There are client operation in these days. Besides most columnfamily in the
> cluster are supercolumnfamily created by cassandra-cli. Most rows have
> average 30 sub-rows and each sub-row has 20 columns.
>
Supercolumns, especially pre-CQL impleme
The following is the snippets of nodetool compactions history. It shows
many records relates to keyspace system.
Compaction History:
id keyspace_name
columnfamily_namecompacted_at bytes_in
bytes_out rows_merged
8e4f8830-b04f-11e5
If you are lucky that might mask the real issue, but I doubt it… that is an
insane number of compaction tasks and indicative of another problem. I would
check release notes of 2.0.6+, if I recall that was not a stable version and
may have had leaks.
Aside from that, just FYI, if you use native_
Simplest option is to use java 8 with G1 gc.
> On 31 Dec 2015, at 10:23 a.m., Shuo Chen wrote:
>
> I have a cassandra 2.0.6 cluster with four nodes as backup database. The only
> operation is posting data into db. Recently, the full gc of the nodes
> increases apparently and blocks cluster op
