Hi Community,
We currently have a 3-node Cassandra 3.11 production cluster that was
initially set up with the following configuration:
-
*Snitch:* endpoint_snitch: SimpleSnitch
-
*Keyspace replication:* {'class': 'SimpleStrategy',
'replication_factor': '3'}
Unfortunately, this setup was not ideal for a production environment, and
we're now looking to correct it. Specifically, we want to:
1.
*Switch to a more appropriate snitch* (e.g., GossipingPropertyFileSnitch)
2.
*Update keyspaces to use* NetworkTopologyStrategy
I would appreciate any advice on the best practices for making these
changes in a running production cluster. My main concerns are:
-
What is the safest order of operations?
-
Will switching the snitch or replication strategy cause any *data loss
or availability issues*?
-
Do we need to perform any additional steps (e.g., full repair, rolling
restart, etc.)?
-
Are there specific * known issues* when migrating from SimpleSnitch +
SimpleStrategy to a production-grade setup?
Thanks in advance for your guidance!
