Github user f1yegor commented on the issue:
https://github.com/apache/flink/pull/1771
Could somebody point to a documentation how we could manage state (pool of
connections) when Sink works inside single JVM.
---
If your project is set up for it, you can reply to this email and have
Github user zentol commented on the issue:
https://github.com/apache/flink/pull/1771
You are correct that every CassandraSink instance opens a separate
connection to the cluster. I don't see how we could avoid this without using
Singletons, which should be avoided.
---
If your proje
Github user f1yegor commented on the issue:
https://github.com/apache/flink/pull/1771
Looking at the code of `CassandraSinkBase` I assume that every new sink
will create own session to the cluster.
I'm creating arount 1000 streams and sinks for them, so cassandra cluster
would be
Github user rmetzger commented on the issue:
https://github.com/apache/flink/pull/1771
The merging will probably be a bit delayed because there are some memory
issues:
```
---
T E S T S
--
Github user rmetzger commented on the issue:
https://github.com/apache/flink/pull/1771
Thank you for testing the cassandra connector.
I'll merge the pull request now.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as wel
Github user theomega commented on the issue:
https://github.com/apache/flink/pull/1771
I tried out this branch and it works like it should in the scenario I set
up:
I wrote (so only using the Sink) to a quiet complex columnfamily in a 8
node cassandra cluster. I was using a comple
Github user rmetzger commented on the issue:
https://github.com/apache/flink/pull/1771
I've reviewed the connector again.
The issues I've seen previously (failure on restart) are resolved.
However, I found new issues:
- The Cassandra Sink doesn't fail (at least not within 15
