Piotr Jastrzębski

I'm trying to understand your use case. Why are you using cloud load balancer instead of nodes in the cluster, @congguosn ? How do you intend to read system.local on...

Version is good. We just haven't released the right artefacts yet.

What is a front page readme.md? What's exactly wrong on http://docs.scylladb.com/index?

The interaction is as follows: 1. Pooling options set core connections per host and max connections per host 2. If shard awareness is enabled then each shard gets coreConnectionsSize /...

@slivne ping. Would you like me to do anything with this or can we close the issue?

There's a SIZE_ESTIMATES table that has some info about tables sizes. https://github.com/scylladb/scylla/blob/master/db/system_keyspace.cc#L455

Have you tried setting `data_file_directories` in conf/cassandra.yaml in tools to the data directory? It helped for me.

It would probably be good as the default option should be the one that performs better. I won't insist though :)

@penberg @elcallio I would wait with enabling this by default. I was testing sstable loader with data generated by cassandra stress. Default schema is: CREATE KEYSPACE keyspace1 WITH replication =...

@slivne @elcallio @avikivity @amoskong