You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The cluster:refresh() method is now safe to be called at runtime,
allowing cluster topology changes (added or removed nodes) to be taken into
consideration. #134
The cluster:refresh() method now accepts an optional timeout argument,
and returns the topology changes (if any). #134
Improve host still considered down error logs to include the error causing
each node to be considered as "DOWN", as well as the duration for which it
will still be considered "DOWN". #129
Environments with DNS load-balancing in effect for contact_points provided
as hostnames (e.g. Kubernetes with contact_points = { "cassandra" }) could
result in no host details for <peer IP> errors when using multiple
instances of the Cluster module. This is now fixed. #118
The new request-aware load-balancing policies can now be used in the
init_by_lua* context when using the lua-resty-core module (only applies
to OpenResty environments). #117
New cafile option for the Cluster module. This allows supporting SSL
connections to Cassandra clusters when lua-cassandra is used in contexts
that do not support cosockets, and fallback on LuaSocket. #95