I have the following cluster setup:
- 2 nodes
And have encountered the following problem during the rolling release redeployment:
Cluster Re-preparing already prepared query is generally an anti-pattern and will likely affect performance. Consider preparing the statement only once. Query=...
Plenty of such messages were registered on the new nodes and the affected queries were unavailable through the driver client, though from the code these queries are prepared only once.
The release was as follows:
- 3rd node was introduced to the cluster
- 1nd node was asked to create a couple of new tables and did it successfully
- 3rd node tried to prepare queries for these new tables and entered in an infinite loop of above mentioned warnings
- 1st node was removed from the cluster
- 4th node was introduced to the cluster
- 4th node tried to prepare queries for these new tables and entered in infinite loop of above mentioned warnings
- 2nd node was removed from the cluster
It all continued up until next full redeploymen. And then queries were prepared successfully and everything worked well.
Anyone can give some clues as what could be misconfigured in my case?